From nobody Wed Mar 2 00:44:30 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 661B11A905A for ; Wed, 2 Mar 2016 00:44:29 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.507 X-Spam-Level: X-Spam-Status: No, score=-14.507 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.006, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id B1Y4zrEx6u1O for ; Wed, 2 Mar 2016 00:44:27 -0800 (PST) Received: from alln-iport-4.cisco.com (alln-iport-4.cisco.com [173.37.142.91]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8AC7E1A9058 for ; Wed, 2 Mar 2016 00:44:27 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=3535; q=dns/txt; s=iport; t=1456908267; x=1458117867; h=date:from:to:cc:subject:message-id:references: mime-version:in-reply-to; bh=zrAq6gE15M8drHv1GqfIZFZVrH3qWXjX+ZJfsbprKRw=; b=IbhLEutPUIlaPkYfvNPqJHrGutRMyWo6kYLpnV4lvfkbBzm7pyPKv21k jluBAIHsr/iXe4bB9KvtYBK5mousfh6hwiE9efHZDI92FJcDxTZlUlo1C 7LpYgtCuZUJv0/6K9ddsbhdL/3fG1MScyzXbL9wNZcrRq+l7rPK1GKdIj A=; X-IronPort-AV: E=Sophos;i="5.22,527,1449532800"; d="scan'208";a="243770241" Received: from alln-core-4.cisco.com ([173.36.13.137]) by alln-iport-4.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 02 Mar 2016 08:44:27 +0000 Received: from mcast-linux1.cisco.com (mcast-linux1.cisco.com [172.27.244.121]) by alln-core-4.cisco.com (8.14.5/8.14.5) with ESMTP id u228iQGe021764 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Wed, 2 Mar 2016 08:44:26 GMT Received: from mcast-linux1.cisco.com (localhost.cisco.com [127.0.0.1]) by mcast-linux1.cisco.com (8.13.8/8.13.8) with ESMTP id u228iQRE020667; Wed, 2 Mar 2016 00:44:26 -0800 Received: (from eckert@localhost) by mcast-linux1.cisco.com (8.13.8/8.13.8/Submit) id u228iQER020666; Wed, 2 Mar 2016 00:44:26 -0800 Date: Wed, 2 Mar 2016 00:44:26 -0800 From: Toerless Eckert To: Brian E Carpenter , Michael Behringer Message-ID: <20160302084425.GC18386@cisco.com> References: <56B29E9D.5030508@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <56B29E9D.5030508@gmail.com> User-Agent: Mutt/1.4.2.2i Archived-At: Cc: Anima WG Subject: Re: [Anima] GRASP Issue 28: Does ACP support secure link-local multicast? X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 02 Mar 2016 08:44:29 -0000 Brian, Michael: Once we agree on the answers, i think those answers need to be documented somewhere. ACP draft ? Michael ? 1. We may have one instance of GRASP run across L2 to find candidate ACP neighbors. This would be a minimal instance as its insecure. And we have to figure out if we want GRASP here or use some other potentially pre-eisting protocol like mDNS. Eg "Find ACP service on LAN". I am not a fan of using mDNS for this because i see a high likelyhood of creating deployment problems when so many hosts on LANs speak mDNS. Once neighbor is seen in this discovery, we would start secure negotiation with - still to pick IPsec/(DIKE)/(d)TLS. So, this initial instance would be insecure. 2. For GRAP run inside of ACP, there will of course be support for link-local multicast. There is one interesting level of design that would be good to work through and write down: All "simple" security associations negotiation are p2p. Eg: IKE, SSL. so i would like to standardize for ACP first only such p2p ACP neighborships. Doing p2mp ACP tunnels should be a later extension if demand justifies it. I am saying this because i know how complex GDOI etc. become. Every L2 interface (with ACP neighbors) would have one or more ACP "tunnels" - one per neighbor. This leads to two possible APIs (this is all stuff we've done in IP multicast since the 90th): a) L2-subnet API / "L2 pseudo broadcast": You send a multicast (link-local) to an L2-subnet. ACP does send a copy to every ACP p2p tunnel on that interface. b) per-neighbor API: You send multicast (link-local) to a specific ACP tunnel. The important thing is really just that any application that wants to use LL multicast understands how to use these options and uses them consistently. If you want to do flooding, option a) is usually easier: GARP on node AN1 on LANx sends LL-MC to LANx. ACP replicates to all ACP tunnels associated with LANx. AN2 on LANx receives the LL-MC, passes to GARP. GARP replicates/ modifies messages. Replicates it out to all other LANs, but must not send a copy back to LANx. Option b) is better when you have some multicast that only needs to go to some subset of ACP neighbors on a LAN. Then the application, like GARP would need to take care of replicating per neighbor. In Option b) the app like GARP could of course still know for every neighbor/p2p-interface which underlying L2 interface it's associated with and therefore replicate the behavior of a) by itself if desired. Cheers toerless On Thu, Feb 04, 2016 at 01:43:09PM +1300, Brian E Carpenter wrote: > o 28. Does ACP support secure link-local multicast? > > The issue here is that GRASP has no choice but to use multicast > for discovery or flood messages, and such messages are intrinsically > insecure. So this isn't really an open issue for the protocol, but it > is a question for the ACP designers. Can they provide an ACP virtual > interface that supports secure LL multicast? > > For the GRASP draft, this should probably be changed from being an open > issue into a specific paragraph in the Security Considerations. > > Brian > > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima -- --- Toerless Eckert, eckert@cisco.com From nobody Wed Mar 2 01:06:06 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DD25D1A9145 for ; Wed, 2 Mar 2016 01:06:04 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.65 X-Spam-Level: X-Spam-Status: No, score=-1.65 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FR_ALMOST_VIAG2=10.357, MANGLED_VIAGRA=2.5, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.006, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qNGIG2Frs7OY for ; Wed, 2 Mar 2016 01:05:59 -0800 (PST) Received: from rcdn-iport-5.cisco.com (rcdn-iport-5.cisco.com [173.37.86.76]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B1A1A1A90DD for ; Wed, 2 Mar 2016 01:05:59 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=5983; q=dns/txt; s=iport; t=1456909559; x=1458119159; h=date:from:to:cc:subject:message-id:references: mime-version:in-reply-to; bh=+S2vD41L8KoLJnqODEe7eUgRUMOLjE/DtAs0IHOGwkA=; b=DP1cE5KHZv/gz99QkwlXQS2uRFohH1QAY04b4UPVnhCdcyXC8wyL7L+9 ErYYwqSAAQrgHNVxXgkTjcRklEZVb8ADUCjCGDX2bNqWFePTEe5dVYSb+ 61SLzeFV8XtGOTgQEi0E5L2ykH6/WY4xFmfJMqhZfc6RVP9HP3CAeFpST c=; X-IronPort-AV: E=Sophos;i="5.22,527,1449532800"; d="scan'208";a="78756039" Received: from rcdn-core-1.cisco.com ([173.37.93.152]) by rcdn-iport-5.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 02 Mar 2016 09:05:59 +0000 Received: from mcast-linux1.cisco.com (mcast-linux1.cisco.com [172.27.244.121]) by rcdn-core-1.cisco.com (8.14.5/8.14.5) with ESMTP id u2295wiw024572 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Wed, 2 Mar 2016 09:05:58 GMT Received: from mcast-linux1.cisco.com (localhost.cisco.com [127.0.0.1]) by mcast-linux1.cisco.com (8.13.8/8.13.8) with ESMTP id u2295vOx021751; Wed, 2 Mar 2016 01:05:58 -0800 Received: (from eckert@localhost) by mcast-linux1.cisco.com (8.13.8/8.13.8/Submit) id u2295vnr021750; Wed, 2 Mar 2016 01:05:57 -0800 Date: Wed, 2 Mar 2016 01:05:57 -0800 From: Toerless Eckert To: Brian E Carpenter Message-ID: <20160302090557.GD18386@cisco.com> References: <56B29FDE.3000407@gmail.com> <38de8937546a4d67a82353cb99887804@XCH-RCD-006.cisco.com> <56B3A11B.2090304@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <56B3A11B.2090304@gmail.com> User-Agent: Mutt/1.4.2.2i Archived-At: Cc: Anima WG , "Michael Behringer \(mbehring\)" Subject: Re: [Anima] GRASP Issue 42: ACP bypass? X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 02 Mar 2016 09:06:05 -0000 I have several times suggested that the GRASP text needs the notion of an "instance". Ther MUST be a "secure/ACP" instance, meaning that any information that an agent passes into this "instance" MUST be guaranteed by GRAP to be only passed on across other ACP links. That IMHO is the core instance of GRASP we need in ANIMA. There are a lot of reasons for addition instances that hae other purposes. I menioned in my last mail eg: an instance per L2 interface to help finding neighbors for the ACP. Thats an example of insecure instances. What would be an example of sending a ot of traffic via GRASP ? Eg: WOuld i try to send a router-firmware-upgrade via GRASP ? ;-)) If i need to do that, then i could see how extensions to GRASP could transparently and securely use the data-plane: -> GRASP wants p2p communication with a remote neighbor. It also knows it needs throughput: -> It needs to use the secure/ACP instance of GRAP to learn of that peer a) data-plane address (IPv6/IPv4) b) ACP cert. -> It can then optimize communication by setting up an SSL data plane connection to that peer. If we have a clear use-case, we could investigate this further. Remember that two AN devices that can talk to each other via ACP may not be able to talkk to each other via data plane, eg: there could be NAT, VRFs etc. pp. be in the way. If the GARP p2p connection is TCP/SCTP, then we could also do the switchover not in the GARP layer, but inside of MP-TCP/SCTP. These protocols would aleviate the need to exchange the cert because they do exchange (if i recall correctly) some challenge token across the established connection (ACP), that they then use to setup another connection - eg: via data-plane. I do describe the MP-TCP/SCTP idea in the stable connectivity draft. The main complexity would be that we would want "auto-security" so the additional path across data-plane would have to have an auttomatically build security association underneath. Eg: dTLS. I have to ask some transport folks if there has already been any work in this direction. To summarize and divide and conquer: - GRASP should have a notion of non-conflicting instances. GRP messages received are associated to an instance based on how they are received from. Eg: received with the primary defined encap across ACP: assigned to ACP instance. - For ANIMA we need to define appropriately in GRASP and/or another draft that our primary GRASP instance is a Secure/ACP one, where all connections are across ACP so agents know that when sending/receiving GRASP messages into/from this instance it is secure. - Everything else is cool, but secondary. Cheers Toerless On Fri, Feb 05, 2016 at 08:06:03AM +1300, Brian E Carpenter wrote: > Hi Michael, > > On 05/02/2016 06:47, Michael Behringer (mbehring) wrote: > > Brian, > > > > I think you're asking the wrong question :-) > > I suspect it wasn't my question originally ;-) > > > I believe we had come to the conclusion that an ASA is essentially just a piece of code, which can do anything it wants; > specifically, we can't force it to use the ACP, it could also connect through the normal data plane to non-autonomic devices. > The use case was that an autonomic agent on a physical device may well control other devices which are possibly non-autonomic. > This transaction would not be inside the ACP (because the other device is non autonomic). So an ASA can have ACP based > interactions, and non-ACP based interactions. > > > > I think there are different aspects to what you're saying: > > > > - GRASP should ONLY run in the ACP. (I think this is what you mean, really, and I think we agree on that) > > Yes, or in a TLS wrapper in the absence of an ACP. > > > - But GRASP can negotiate ANYTHING. In fact, one negotiation objective could be the IP address of the peer node in the data plane. Why not? > > The IP address, or FQDN if you prefer. Certainly possible. > > > So the negotiation itself is in the ACP, the data exchanged may refer to something outside the ACP, and now *IF* you were to do a connection to that address, you wouldn't be protected by the ACP any more. > > > > But that's ok (in my opinion), as long as the context is explicit. I.e. the ASA *knows* it's dealing with something outside the ACP. I had proposed a long time ago a "context", which could then point to the data plane (or other virtual contexts). > > Right. What I think would be dangerous is for GRASP discovery to return an address which is > not secured, i.e. is not an ACP or TLS-wrappable address. That would allow an ASA to slide > gently from secure to insecure without knowing about it. > > So I think this can be resolved by some text in the Security Considerations. > > Thanks > Brian > > > > > Michael > > > > > >> -----Original Message----- > >> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Brian E > >> Carpenter > >> Sent: 04 February 2016 01:49 > >> To: Anima WG > >> Subject: [Anima] GRASP Issue 42: ACP bypass? > >> > >> o 42. Do we need a feature whereby an ASA can bypass the ACP and > >> use the data plane for efficiency/throughput? This would require > >> discovery to return non-ACP addresses and would evade ACP > >> security. > >> > >> Personal opinion: This is probably a bad idea. Trading security for efficiency in > >> network management seems like a mistake, even in autonomic > >> management. > >> > >> WG input, please... > >> > >> Regards > >> Brian > >> > >> _______________________________________________ > >> Anima mailing list > >> Anima@ietf.org > >> https://www.ietf.org/mailman/listinfo/anima > > > > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima -- --- Toerless Eckert, eckert@cisco.com From nobody Wed Mar 2 10:51:22 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2FDEF1B2D30 for ; Wed, 2 Mar 2016 10:51:21 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2 X-Spam-Level: X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id SbBbqyLSDrCB for ; Wed, 2 Mar 2016 10:51:18 -0800 (PST) Received: from mail-pf0-x232.google.com (mail-pf0-x232.google.com [IPv6:2607:f8b0:400e:c00::232]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id EE37D1B2D1D for ; Wed, 2 Mar 2016 10:51:17 -0800 (PST) Received: by mail-pf0-x232.google.com with SMTP id 124so76846062pfg.0 for ; Wed, 02 Mar 2016 10:51:17 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=lGCdJt+6ox1uQv1yugNx/ZREm+Hb+AJ34cg0zxs9eC0=; b=HNKrjAxUc9frf1N0MWotJEn9+893u7tG86b0Zn82Zwj9tSzKJKTVvpwacKFAXcuYjD LqtOhtTcd/UApK68sQ9C1V7+OuoNP/3Gs+lBt2hcPzfWUagwGZfrZc1DkKajvqd1xMoA gqTGuGc55Tgc0c/Ehe3ztzx6NLz8MuGL1e50wwvmv4N9AeqRBc+2R9veHnBvFzBADH76 AMySwsepQ6ma2MQpi68UQOhQtrSzYpxFG/q+rTwp2PXgpL4iQjlJi+L/O9aT/bxqSPII 6oJzfTlm6c0wmaBjE4syBFryJfyudeS+pUvRFLRiYB2I/GDqHkXPsu4blsW6ja9/XqRm cflA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=lGCdJt+6ox1uQv1yugNx/ZREm+Hb+AJ34cg0zxs9eC0=; b=FVpY4Dz9RaqK2+qIOgQwU7XzcC0+W5K0h/xR7ssLQ6c5uCJmuqZZ24BE5nPOlQnca4 X1ZuIwqFOuIH8+PwadWK3lxwzs0lRQ2aL7PWil5raw1dbwPA1oiiJuaVw6E/Bf1KvO92 2/AtLlWwF/I+iaqAuyhgQSi0M4G9TJ+iaz+tiyj/guQtOuMvPn+jPFEkWlM+WVp3L1JG h9NthvaPIWhLmwa4HV62fUmwU5aUAmUDk3wrAABNIAfPwvQpR6E12kZGS+3AAX9xrbpG I6AABroFvSSJwJm9c+eWCLAMGcLYqzvZZw3qPokF2vi/lSK3tPSWkg5cE9ycHWU6t3NG 4NmA== X-Gm-Message-State: AD7BkJLSIXZWODmVVkZ5cCSGkxb+/jYjfH3aqBj99d+XldD+eblkR/Qcv+NOG0yENkzejg== X-Received: by 10.98.76.194 with SMTP id e63mr40712871pfj.9.1456944677587; Wed, 02 Mar 2016 10:51:17 -0800 (PST) Received: from ?IPv6:2406:e007:451b:1:28cc:dc4c:9703:6781? ([2406:e007:451b:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id d22sm54713770pfj.32.2016.03.02.10.51.13 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 02 Mar 2016 10:51:15 -0800 (PST) To: Toerless Eckert , Michael Behringer References: <56B29E9D.5030508@gmail.com> <20160302084425.GC18386@cisco.com> From: Brian E Carpenter Organization: University of Auckland Message-ID: <56D73625.80609@gmail.com> Date: Thu, 3 Mar 2016 07:51:17 +1300 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0 MIME-Version: 1.0 In-Reply-To: <20160302084425.GC18386@cisco.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Archived-At: Cc: Anima WG Subject: Re: [Anima] GRASP Issue 28: Does ACP support secure link-local multicast? X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 02 Mar 2016 18:51:21 -0000 Hi, On 02/03/2016 21:44, Toerless Eckert wrote: > Brian, Michael: > > Once we agree on the answers, i think those answers need to be documented > somewhere. ACP draft ? Michael ? > > 1. We may have one instance of GRASP run across L2 to find candidate > ACP neighbors. This would be a minimal instance as its insecure. > And we have to figure out if we want GRASP here or use some other > potentially pre-eisting protocol like mDNS. Eg "Find ACP service on LAN". > > I am not a fan of using mDNS for this because i see a high likelyhood > of creating deployment problems when so many hosts on LANs speak mDNS. > > Once neighbor is seen in this discovery, we would start secure negotiation > with - still to pick IPsec/(DIKE)/(d)TLS. > > So, this initial instance would be insecure. Sure. GRASP is certainly an option for that. We just need IANA to assign ALL_GRASP_NEIGHBOR_6 and GRASP_LISTEN_PORT. (I am currently using ff02::114 and 1021.) > > 2. For GRAP run inside of ACP, there will of course be support for > link-local multicast. > > There is one interesting level of design that would be good to work > through and write down: > > All "simple" security associations negotiation are p2p. Eg: IKE, SSL. > so i would like to standardize for ACP first only such p2p ACP > neighborships. Doing p2mp ACP tunnels should be a later extension > if demand justifies it. I am saying this because i know how complex > GDOI etc. become. > > Every L2 interface (with ACP neighbors) would have one or more > ACP "tunnels" - one per neighbor. This leads to two possible APIs > (this is all stuff we've done in IP multicast since the 90th): > > a) L2-subnet API / "L2 pseudo broadcast": > > You send a multicast (link-local) to an L2-subnet. > ACP does send a copy to every ACP p2p tunnel on that interface. I prefer that. It simply means that when the ACP comes up, GRASP uses that L2 interface instead of the underlying physical interface that it uses in insecure mode. Presumably the necessary information will be in the ACP adjacency table that Michael proposed some time ago. > > b) per-neighbor API: > > You send multicast (link-local) to a specific ACP tunnel. > > The important thing is really just that any application that > wants to use LL multicast understands how to use these options > and uses them consistently. > > If you want to do flooding, option a) is usually easier: GRASP floods both Discovery and Flood Synchronization messages, since every AN is potentially a target, so I think we need a) > GARP on node AN1 on LANx sends LL-MC to LANx. ACP > replicates to all ACP tunnels associated with LANx. > AN2 on LANx receives the LL-MC, passes to GARP. GARP replicates/ > modifies messages. Replicates it out to all other LANs, but must > not send a copy back to LANx. Right, but Joel recently identified a slight complication there: if there's a loop in the router topology, we have to prevent a copy being sent back to LANx indirectly. That's actually a defect in the GRASP -03 specification and the prototype code, which we will fix of course. Regards Brian > > Option b) is better when you have some multicast that only > needs to go to some subset of ACP neighbors on a LAN. Then the > application, like GARP would need to take care of replicating per > neighbor. In Option b) the app like GARP could of course still > know for every neighbor/p2p-interface which underlying L2 interface > it's associated with and therefore replicate the behavior of a) > by itself if desired. > > Cheers > toerless > > On Thu, Feb 04, 2016 at 01:43:09PM +1300, Brian E Carpenter wrote: >> o 28. Does ACP support secure link-local multicast? >> >> The issue here is that GRASP has no choice but to use multicast >> for discovery or flood messages, and such messages are intrinsically >> insecure. So this isn't really an open issue for the protocol, but it >> is a question for the ACP designers. Can they provide an ACP virtual >> interface that supports secure LL multicast? >> >> For the GRASP draft, this should probably be changed from being an open >> issue into a specific paragraph in the Security Considerations. >> >> Brian >> >> _______________________________________________ >> Anima mailing list >> Anima@ietf.org >> https://www.ietf.org/mailman/listinfo/anima > From nobody Wed Mar 2 13:26:04 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BC2E71B31F1 for ; Wed, 2 Mar 2016 13:26:02 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -11.807 X-Spam-Level: X-Spam-Status: No, score=-11.807 tagged_above=-999 required=5 tests=[BAYES_50=0.8, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.006, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bK_j9bwx0S4g for ; Wed, 2 Mar 2016 13:26:00 -0800 (PST) Received: from rcdn-iport-9.cisco.com (rcdn-iport-9.cisco.com [173.37.86.80]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 7942C1B31DC for ; Wed, 2 Mar 2016 13:26:00 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=6545; q=dns/txt; s=iport; t=1456953960; x=1458163560; h=date:from:to:cc:subject:message-id:references: mime-version:in-reply-to; bh=sCWvbvAErh2AM/5igGBhDKgW1tGTHUvwskDGToDx1pA=; b=U7F10dLzfJ9lM+9jTj1EWwxynV33du8RtU+v8sySpq8VInpq1jXQXO8l RHdlqKYV0Fg/hD8xbPUaReod2NdfPJVfJ4QR86U/pX3gPwfPBAaIhJ9l9 XJF1NIqKh/bURhrFMCdMsBQP4kNyJ1jjaNlFPR5ecpba2xrYOphtQLrRH g=; X-IronPort-AV: E=Sophos;i="5.22,530,1449532800"; d="scan'208";a="76751221" Received: from alln-core-2.cisco.com ([173.36.13.135]) by rcdn-iport-9.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 02 Mar 2016 21:25:59 +0000 Received: from mcast-linux1.cisco.com (mcast-linux1.cisco.com [172.27.244.121]) by alln-core-2.cisco.com (8.14.5/8.14.5) with ESMTP id u22LPxXq015604 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Wed, 2 Mar 2016 21:25:59 GMT Received: from mcast-linux1.cisco.com (localhost.cisco.com [127.0.0.1]) by mcast-linux1.cisco.com (8.13.8/8.13.8) with ESMTP id u22LPxH0032588; Wed, 2 Mar 2016 13:25:59 -0800 Received: (from eckert@localhost) by mcast-linux1.cisco.com (8.13.8/8.13.8/Submit) id u22LPw1I032587; Wed, 2 Mar 2016 13:25:58 -0800 Date: Wed, 2 Mar 2016 13:25:58 -0800 From: Toerless Eckert To: Brian E Carpenter Message-ID: <20160302212558.GF18386@cisco.com> References: <56B29FDE.3000407@gmail.com> <38de8937546a4d67a82353cb99887804@XCH-RCD-006.cisco.com> <56B3A11B.2090304@gmail.com> <20160302090557.GD18386@cisco.com> <56D73AE0.1020701@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <56D73AE0.1020701@gmail.com> User-Agent: Mutt/1.4.2.2i Archived-At: Cc: "Michael Behringer \(mbehring\)" , Anima WG Subject: Re: [Anima] GRASP Issue 42: ACP bypass? X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 02 Mar 2016 21:26:02 -0000 On Thu, Mar 03, 2016 at 08:11:28AM +1300, Brian E Carpenter wrote: > I have been thinking of that as a *mode* not an instance, i.e. the GRASP engine > comes up by default in "insecure" mode which restricts the operations it will > accept, and then either changes to "ACP" mode when it determines that the ACP is up, > or to "TLS" mode if there is no ACP but there is a certificate infrastructure. > > It's probably correct that we need to tag individual interfaces as "insecure" > rather than assuming all interfaces are the same. If i run just GRASP across an interface insecure to discover candidate ACP neighbors, that is signaling that has nothing to do with the GRASP running inside the ACP, and "instance" to me is the best expression to indicate that there should be no interaction between them. > > -> GRASP wants p2p communication with a remote neighbor. It also knows > > it needs throughput: > > Why is that GRASP? Surely it will be an individual ASA that wants such > communication with another ASA? I don't know if GRASP should or would be used for sessions requring a good amount of bandwidth. When HTTP was invented i think it too was seen unlikely that you'd end up transporting every type of big data across it. > > -> It needs to use the secure/ACP instance of GRAP to learn of that peer > > a) data-plane address (IPv6/IPv4) > > b) ACP cert. > > -> It can then optimize communication by setting up an SSL data plane > > connection to that peer. > > Indeed. But does GRASP have any advantage for that over plain TLS? > I don't understand the advantage of that to the ASAs. Right. See above Nothing GRASP specific here. Just: IF some protocol between ASAs wants to transfer a lot of data, it would be good if that would automatically use the data plane - if data plane is available. And my description was just explaining how. > > To summarize and divide and conquer: > > - GRASP should have a notion of non-conflicting instances. > > If they don't conflict, they are either in totally orthogonal > address spaces, or on totally orthogonal interfaces, or use > a different LL multicast address and well-known port. (Or we > add a network-wide instance identifier to the protocol, which > seems like a horrible idea.) Exactly. Thats what i call "instance" ;-) Cheeers Toerless > Regards > Brian > > > GRP messages received are associated to an instance based on > > how they are received from. Eg: received with the primary defined > > encap across ACP: assigned to ACP instance. > > > > - For ANIMA we need to define appropriately in GRASP and/or another draft > > that our primary GRASP instance is a Secure/ACP one, where all > > connections are across ACP so agents know that when sending/receiving > > GRASP messages into/from this instance it is secure. > > > > - Everything else is cool, but secondary. > > > > Cheers > > Toerless > > > > On Fri, Feb 05, 2016 at 08:06:03AM +1300, Brian E Carpenter wrote: > >> Hi Michael, > >> > >> On 05/02/2016 06:47, Michael Behringer (mbehring) wrote: > >>> Brian, > >>> > >>> I think you're asking the wrong question :-) > >> > >> I suspect it wasn't my question originally ;-) > >> > >>> I believe we had come to the conclusion that an ASA is essentially just a piece of code, which can do anything it wants; > >> specifically, we can't force it to use the ACP, it could also connect through the normal data plane to non-autonomic devices. > >> The use case was that an autonomic agent on a physical device may well control other devices which are possibly non-autonomic. > >> This transaction would not be inside the ACP (because the other device is non autonomic). So an ASA can have ACP based > >> interactions, and non-ACP based interactions. > >>> > >>> I think there are different aspects to what you're saying: > >>> > >>> - GRASP should ONLY run in the ACP. (I think this is what you mean, really, and I think we agree on that) > >> > >> Yes, or in a TLS wrapper in the absence of an ACP. > >> > >>> - But GRASP can negotiate ANYTHING. In fact, one negotiation objective could be the IP address of the peer node in the data plane. Why not? > >> > >> The IP address, or FQDN if you prefer. Certainly possible. > >> > >>> So the negotiation itself is in the ACP, the data exchanged may refer to something outside the ACP, and now *IF* you were to do a connection to that address, you wouldn't be protected by the ACP any more. > >>> > >>> But that's ok (in my opinion), as long as the context is explicit. I.e. the ASA *knows* it's dealing with something outside the ACP. I had proposed a long time ago a "context", which could then point to the data plane (or other virtual contexts). > >> > >> Right. What I think would be dangerous is for GRASP discovery to return an address which is > >> not secured, i.e. is not an ACP or TLS-wrappable address. That would allow an ASA to slide > >> gently from secure to insecure without knowing about it. > >> > >> So I think this can be resolved by some text in the Security Considerations. > >> > >> Thanks > >> Brian > >> > >>> > >>> Michael > >>> > >>> > >>>> -----Original Message----- > >>>> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Brian E > >>>> Carpenter > >>>> Sent: 04 February 2016 01:49 > >>>> To: Anima WG > >>>> Subject: [Anima] GRASP Issue 42: ACP bypass? > >>>> > >>>> o 42. Do we need a feature whereby an ASA can bypass the ACP and > >>>> use the data plane for efficiency/throughput? This would require > >>>> discovery to return non-ACP addresses and would evade ACP > >>>> security. > >>>> > >>>> Personal opinion: This is probably a bad idea. Trading security for efficiency in > >>>> network management seems like a mistake, even in autonomic > >>>> management. > >>>> > >>>> WG input, please... > >>>> > >>>> Regards > >>>> Brian > >>>> > >>>> _______________________________________________ > >>>> Anima mailing list > >>>> Anima@ietf.org > >>>> https://www.ietf.org/mailman/listinfo/anima > >>> > >> > >> _______________________________________________ > >> Anima mailing list > >> Anima@ietf.org > >> https://www.ietf.org/mailman/listinfo/anima > > > > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima -- --- Toerless Eckert, eckert@cisco.com From nobody Wed Mar 2 16:03:56 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8C9E51B35B2 for ; Wed, 2 Mar 2016 16:03:54 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2 X-Spam-Level: X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id igJwf1hb8mKN for ; Wed, 2 Mar 2016 16:03:52 -0800 (PST) Received: from mail-pf0-x231.google.com (mail-pf0-x231.google.com [IPv6:2607:f8b0:400e:c00::231]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 7E3BD1B35A3 for ; Wed, 2 Mar 2016 16:03:52 -0800 (PST) Received: by mail-pf0-x231.google.com with SMTP id w128so3255680pfb.2 for ; Wed, 02 Mar 2016 16:03:52 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=D0i/BDA/IRxltLusmfdYLKufDopzZxZmMg1KbwNoXP0=; b=a/sWUMJzuMkLL70A2CmRmv0B9hsBRLAfhHphQha42g5qYwmaCdoIEET+BiAHmwbgD/ 3Ikw+S3GSY0PdpgF2c2Td2PwnTonUJPgJLiHwrDLL2Cvqdijr40aspnMDCsoc+sgyigj JL6FOsCIW9Rf0Tn4ANScuzZm98/tuf94A3vY7oD6307G4C2k9UoLN3mSLIDzJznEDy92 wlLiFg9C7N2xalqi7AKls8Wc7XTxM5VBm/wt9tYw070HgPM3WJeog1Bbvyl4t0qdNLTH 7yfVSoVa0B9kVP0N46M05zTWgkdhHrah5Y9sSvJX0XQz3nFY9qUckxkhJUUHD8O8tpOK fIpQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=D0i/BDA/IRxltLusmfdYLKufDopzZxZmMg1KbwNoXP0=; b=JHS4Vnf0FOk7Zg6Z7FKIircKr4W0Hniq/p/aWiNADOTP28qZwv3qs2CW5a6l6XnJBs Ht0k+epZlEWVvXFDE8CM8uaRjGM3bu+Oi02NTlXO89Ik56G3cLyVom/7cdG34sZFeaaF MgQDINi8WZKoayNlzcmJdQDHbWS56LGr/uXwgrjv7FG4/Vc6J7x1/CZf9uOYQoY6kOeT BeFVNY3L6GiK4CTlyjhwrbmYSVc8QfZTNwOzKzeR7hTjAqN8SPy2PtrZi8RwL7AMyEEu k5LSd/w44YGim7GmYtTzVhUcsFD61TFvDx64uQA4ndqoNXr7uDdBJ1QqPrZKhirNyGDO Q8aw== X-Gm-Message-State: AD7BkJJIUjtmQAtM+RfSvJ6+Itwjkg5V+/kiS1V0gXmozaAr7pMx5OCL7+P3FwNoXksG9w== X-Received: by 10.98.93.1 with SMTP id r1mr42298013pfb.57.1456963432121; Wed, 02 Mar 2016 16:03:52 -0800 (PST) Received: from ?IPv6:2406:e007:451b:1:28cc:dc4c:9703:6781? ([2406:e007:451b:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id ml5sm55622620pab.2.2016.03.02.16.03.48 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 02 Mar 2016 16:03:50 -0800 (PST) To: Toerless Eckert References: <56B29FDE.3000407@gmail.com> <38de8937546a4d67a82353cb99887804@XCH-RCD-006.cisco.com> <56B3A11B.2090304@gmail.com> <20160302090557.GD18386@cisco.com> <56D73AE0.1020701@gmail.com> <20160302212558.GF18386@cisco.com> From: Brian E Carpenter Organization: University of Auckland Message-ID: <56D77F68.1090904@gmail.com> Date: Thu, 3 Mar 2016 13:03:52 +1300 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0 MIME-Version: 1.0 In-Reply-To: <20160302212558.GF18386@cisco.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Archived-At: Cc: "Michael Behringer \(mbehring\)" , Anima WG Subject: Re: [Anima] GRASP Issue 42: ACP bypass? X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 03 Mar 2016 00:03:54 -0000 On 03/03/2016 10:25, Toerless Eckert wrote: > On Thu, Mar 03, 2016 at 08:11:28AM +1300, Brian E Carpenter wrote: >> I have been thinking of that as a *mode* not an instance, i.e. the GRASP engine >> comes up by default in "insecure" mode which restricts the operations it will >> accept, and then either changes to "ACP" mode when it determines that the ACP is up, >> or to "TLS" mode if there is no ACP but there is a certificate infrastructure. >> >> It's probably correct that we need to tag individual interfaces as "insecure" >> rather than assuming all interfaces are the same. > > If i run just GRASP across an interface insecure to discover candidate ACP > neighbors, that is signaling that has nothing to do with the GRASP running > inside the ACP, and "instance" to me is the best expression to indicate that there > should be no interaction between them. > >>> -> GRASP wants p2p communication with a remote neighbor. It also knows >>> it needs throughput: >> >> Why is that GRASP? Surely it will be an individual ASA that wants such >> communication with another ASA? > > I don't know if GRASP should or would be used for sessions requring a good > amount of bandwidth. When HTTP was invented i think it too was seen unlikely > that you'd end up transporting every type of big data across it. > >>> -> It needs to use the secure/ACP instance of GRAP to learn of that peer >>> a) data-plane address (IPv6/IPv4) >>> b) ACP cert. >>> -> It can then optimize communication by setting up an SSL data plane >>> connection to that peer. >> >> Indeed. But does GRASP have any advantage for that over plain TLS? >> I don't understand the advantage of that to the ASAs. > > Right. See above Nothing GRASP specific here. Just: IF some protocol > between ASAs wants to transfer a lot of data, it would be good if that > would automatically use the data plane - if data plane is available. And > my description was just explaining how. > >>> To summarize and divide and conquer: >>> - GRASP should have a notion of non-conflicting instances. >> >> If they don't conflict, they are either in totally orthogonal >> address spaces, or on totally orthogonal interfaces, or use >> a different LL multicast address and well-known port. (Or we >> add a network-wide instance identifier to the protocol, which >> seems like a horrible idea.) > > Exactly. Thats what i call "instance" ;-) So, I'm thinking that any text about this belongs in the Reference Model draft, not in the protocol spec. Brian > > Cheeers > Toerless > >> Regards >> Brian >> >>> GRP messages received are associated to an instance based on >>> how they are received from. Eg: received with the primary defined >>> encap across ACP: assigned to ACP instance. >>> >>> - For ANIMA we need to define appropriately in GRASP and/or another draft >>> that our primary GRASP instance is a Secure/ACP one, where all >>> connections are across ACP so agents know that when sending/receiving >>> GRASP messages into/from this instance it is secure. >>> >>> - Everything else is cool, but secondary. >>> >>> Cheers >>> Toerless >>> >>> On Fri, Feb 05, 2016 at 08:06:03AM +1300, Brian E Carpenter wrote: >>>> Hi Michael, >>>> >>>> On 05/02/2016 06:47, Michael Behringer (mbehring) wrote: >>>>> Brian, >>>>> >>>>> I think you're asking the wrong question :-) >>>> >>>> I suspect it wasn't my question originally ;-) >>>> >>>>> I believe we had come to the conclusion that an ASA is essentially just a piece of code, which can do anything it wants; >>>> specifically, we can't force it to use the ACP, it could also connect through the normal data plane to non-autonomic devices. >>>> The use case was that an autonomic agent on a physical device may well control other devices which are possibly non-autonomic. >>>> This transaction would not be inside the ACP (because the other device is non autonomic). So an ASA can have ACP based >>>> interactions, and non-ACP based interactions. >>>>> >>>>> I think there are different aspects to what you're saying: >>>>> >>>>> - GRASP should ONLY run in the ACP. (I think this is what you mean, really, and I think we agree on that) >>>> >>>> Yes, or in a TLS wrapper in the absence of an ACP. >>>> >>>>> - But GRASP can negotiate ANYTHING. In fact, one negotiation objective could be the IP address of the peer node in the data plane. Why not? >>>> >>>> The IP address, or FQDN if you prefer. Certainly possible. >>>> >>>>> So the negotiation itself is in the ACP, the data exchanged may refer to something outside the ACP, and now *IF* you were to do a connection to that address, you wouldn't be protected by the ACP any more. >>>>> >>>>> But that's ok (in my opinion), as long as the context is explicit. I.e. the ASA *knows* it's dealing with something outside the ACP. I had proposed a long time ago a "context", which could then point to the data plane (or other virtual contexts). >>>> >>>> Right. What I think would be dangerous is for GRASP discovery to return an address which is >>>> not secured, i.e. is not an ACP or TLS-wrappable address. That would allow an ASA to slide >>>> gently from secure to insecure without knowing about it. >>>> >>>> So I think this can be resolved by some text in the Security Considerations. >>>> >>>> Thanks >>>> Brian >>>> >>>>> >>>>> Michael >>>>> >>>>> >>>>>> -----Original Message----- >>>>>> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Brian E >>>>>> Carpenter >>>>>> Sent: 04 February 2016 01:49 >>>>>> To: Anima WG >>>>>> Subject: [Anima] GRASP Issue 42: ACP bypass? >>>>>> >>>>>> o 42. Do we need a feature whereby an ASA can bypass the ACP and >>>>>> use the data plane for efficiency/throughput? This would require >>>>>> discovery to return non-ACP addresses and would evade ACP >>>>>> security. >>>>>> >>>>>> Personal opinion: This is probably a bad idea. Trading security for efficiency in >>>>>> network management seems like a mistake, even in autonomic >>>>>> management. >>>>>> >>>>>> WG input, please... >>>>>> >>>>>> Regards >>>>>> Brian >>>>>> >>>>>> _______________________________________________ >>>>>> Anima mailing list >>>>>> Anima@ietf.org >>>>>> https://www.ietf.org/mailman/listinfo/anima >>>>> >>>> >>>> _______________________________________________ >>>> Anima mailing list >>>> Anima@ietf.org >>>> https://www.ietf.org/mailman/listinfo/anima >>> >> >> _______________________________________________ >> Anima mailing list >> Anima@ietf.org >> https://www.ietf.org/mailman/listinfo/anima > From nobody Wed Mar 2 18:16:26 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 483861B38EA for ; Wed, 2 Mar 2016 18:16:25 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.507 X-Spam-Level: X-Spam-Status: No, score=-14.507 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.006, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id g5nZwquBbiSW for ; Wed, 2 Mar 2016 18:16:23 -0800 (PST) Received: from alln-iport-3.cisco.com (alln-iport-3.cisco.com [173.37.142.90]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 087081B38E8 for ; Wed, 2 Mar 2016 18:16:22 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=3256; q=dns/txt; s=iport; t=1456971383; x=1458180983; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-id:content-transfer-encoding: mime-version; bh=IFMnDQKTnwGg2IjywOgqkfWNM1i6ZlC4q7KXbN8C31o=; b=GaxBCjtNdHVeFDs3dpzIOm7DbMsVeYYpEYWHYjSdf6uFCtj/2RG1eoKx x7YxBsUDMkjn58+e2jSTpTuWN9qoMzY5qHyV+QCkOhJbF0ldERFj4kehv PX6ekxPLkfaIecTYnafBi8EhQFnipqhpYm4ylo81oCKd9SGvoAfrlT6ij I=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0ABAgD6nddW/4cNJK1egzpSbQa6IgENg?= =?us-ascii?q?WcXCoVuAhyBJTgUAQEBAQEBAWQnhEEBAQEDAQEBARoGEToLBQsCAQgYAgImAgI?= =?us-ascii?q?CJQsUARABAQQOBQmIEAgOrQOPFgEBAQEBAQEBAQEBAQEBAQEBAQEBARV7hwOCT?= =?us-ascii?q?oQPBwEBG4MCK4EPBY0tiWoBjWKBYIREiFKFdYhWAR4BAUKDZGoBhyQJFx1+AQE?= =?us-ascii?q?B?= X-IronPort-AV: E=Sophos;i="5.22,531,1449532800"; d="scan'208";a="244281737" Received: from alln-core-2.cisco.com ([173.36.13.135]) by alln-iport-3.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 03 Mar 2016 02:16:22 +0000 Received: from XCH-RCD-011.cisco.com (xch-rcd-011.cisco.com [173.37.102.21]) by alln-core-2.cisco.com (8.14.5/8.14.5) with ESMTP id u232GLsI020304 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 3 Mar 2016 02:16:22 GMT Received: from xch-aln-013.cisco.com (173.36.7.23) by XCH-RCD-011.cisco.com (173.37.102.21) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Wed, 2 Mar 2016 20:16:21 -0600 Received: from xch-aln-013.cisco.com ([173.36.7.23]) by XCH-ALN-013.cisco.com ([173.36.7.23]) with mapi id 15.00.1104.009; Wed, 2 Mar 2016 20:16:21 -0600 From: "Max Pritikin (pritikin)" To: Sheng Jiang Thread-Topic: [Anima] Call for agenda ANIMA @ IETF 95, Buenos Aires, Argentina Thread-Index: AdFvqy/RjdZlBxVOSoyQ+hwkjsVZcQFeceMA Date: Thu, 3 Mar 2016 02:16:21 +0000 Message-ID: <5DDF4BE8-A790-4C00-825F-C686EE74664F@cisco.com> References: <5D36713D8A4E7348A7E10DF7437A4B927C62CF54@NKGEML515-MBX.china.huawei.com> In-Reply-To: <5D36713D8A4E7348A7E10DF7437A4B927C62CF54@NKGEML515-MBX.china.huawei.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-messagesentrepresentingtype: 1 x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.99.106.6] Content-Type: text/plain; charset="utf-8" Content-ID: Content-Transfer-Encoding: base64 MIME-Version: 1.0 Archived-At: Cc: "anima@ietf.org" Subject: Re: [Anima] Call for agenda ANIMA @ IETF 95, Buenos Aires, Argentina X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 03 Mar 2016 02:16:25 -0000 UGxlYXNlIGluY2x1ZGUgYSBzbG90IGZvciBib290c3RyYXBwaW5nOg0KDQpOYW1lIG9mIHRpbWUg c2xvdDogYm9vdHN0cmFwcGluZyANCk5hbWUgb2YgZHJhZnQocyk6IGRyYWZ0LWlldGYtYW5pbWEt Ym9vdHN0cmFwcGluZy1rZXlpbmZyYQ0KVGltZSByZXF1ZXN0ZWQ6IFt5b3VyIGNob2ljZV0NClBy ZXNlbnRlciBuYW1lKHMpOiBBdXRob3JzLiBJ4oCZbGwgdGFrZSBpdCBpZiBubyBvbmUgZWxzZSBk b2VzLg0KQnJpZWYgZGVzY3JpcHRpb24gb2Ygd2hhdCBpc3N1ZXMgbmVlZCBkaXNjdXNzaW5nIGFu ZCB3aGF0IHlvdSBob3BlIHRvDQphY2NvbXBsaXNoIGJ5IHByZXNlbnRpbmcgKHBsZWFzZSBmb2N1 cyBvbiBvcGVuIGlzc3VlcywgcmF0aGVyIHRoYW4gYQ0Kc3RhdHVzIHVwZGF0ZSk6DQoJDQoJU3Rh dHVzIHVwZGF0ZS4NCglFeHBsYWluIHByb3h5IG9wZXJhdGlvbnMuDQoJUHJvdG9jb2wgRGV0YWls czogcmVxdWVzdCBmb3IgZmVlZGJhY2suIA0KCURpZmZlcmVuY2UgYmV0d2VlbiDigJx0cnVzdGVk IHZlbmRvciBsb2dnaW5n4oCdIGFuZCDigJxjZXJ0aWZpZWQgbG9nZ2luZyINCg0KLSBtYXgNCg0K PiBPbiBGZWIgMjUsIDIwMTYsIGF0IDI6MDIgQU0sIFNoZW5nIEppYW5nIDxqaWFuZ3NoZW5nQGh1 YXdlaS5jb20+IHdyb3RlOg0KPiANCj4gSGksIGFsbCwNCj4gDQo+IFdlIGhhdmUgcmVxdWVzdGVk IHR3byBzZXNzaW9ucyBvZiAyIGhvdXIgKyAxLjUgaG91ciBmb3IgdGhlIEFOSU1BIFdHDQo+IG1l ZXRpbmcgZm9yIElFVEYtOTUgKEJ1ZW5vcyBBaXJlcywgQXJnZW50aW5hKSBhbmQgYXJlIHN0YXJ0 aW5nIHRvIGNvbGxlY3QNCj4gYWdlbmRhIGl0ZW1zIGZvciB0aGVzZSBzZXNzaW9ucy4gU2FtZSB3 aXRoIElFVEY5NCwgd2UgZG8gbm90IHJlc2VydmUNCj4gdGhlIHNlY29uZCBzZXNzaW9uIGZvciBu b24tY2hhcnRlciB3b3JrIGl0ZW1zLiBJdCBzYXlzIHRoZSBjaGFydGVyIHdvcmsNCj4gaXRlbSB3 b3VsZCBoYXZlIHByaW9yaXR5LiBXZSB3aWxsIG9ubHkgZGlzY3VzcyBub24tY2hhcnRlciB3b3Jr IGl0ZW1zDQo+IHRpbGwgd2UgZmluaXNoIGNoYXJ0ZXIgd29yayBpdGVtcy4gVGhlIHByaW9yaXR5 IGFtb25nIHRoZXNlIG5vbi1jaGFydGVyDQo+IHdvcmsgaXRlbXMgd291bGQgYmU6IHRoZXNlIHRo YXQgaGF2ZSBhY3RpdmUgZGlzY3Vzc2lvbiBpbiBtYWlsIGxpc3QsDQo+IHRoZW4gdGhlc2UgaGF2 ZSBzdWJtaXR0ZWQgZHJhZnRzLCBhbmQgdG9waWNzIHdpdGhvdXQgZHJhZnRzLg0KPiANCj4gUGxl YXNlIHNlbmQgdXMgKGFuaW1hLWNoYWlyc0B0b29scy5pZXRmLm9yZykgcmVxdWVzdHMgZm9yIHNl c3Npb25zIGJ5IA0KPiBNYXJjaCAzcmQsIFRodXJzZGF5IChJRVRGIHJlcXVlc3RzIHRoZSBwcmVs aW1pbmFyeSBhZ2VuZGEgYnkgTWFyY2gNCj4gNHRoLCBGcmlkYXkpIGFuZCBpbmNsdWRlOg0KPiAN Cj4gTmFtZSBvZiB0aW1lIHNsb3Q6DQo+IE5hbWUgb2YgZHJhZnQocyk6DQo+IFRpbWUgcmVxdWVz dGVkOg0KPiBQcmVzZW50ZXIgbmFtZShzKToNCj4gQnJpZWYgZGVzY3JpcHRpb24gb2Ygd2hhdCBp c3N1ZXMgbmVlZCBkaXNjdXNzaW5nIGFuZCB3aGF0IHlvdSBob3BlIHRvDQo+IGFjY29tcGxpc2gg YnkgcHJlc2VudGluZyAocGxlYXNlIGZvY3VzIG9uIG9wZW4gaXNzdWVzLCByYXRoZXIgdGhhbiBh DQo+IHN0YXR1cyB1cGRhdGUpOg0KPiANCj4gTW9yZSBkZXRhaWxzIGFib3V0IHRoZSBJRVRGIDk1 LCBCdWVub3MgQWlyZXMsIEFyZ2VudGluYSBjYW4gYmUgDQo+IGZvdW5kIGF0Og0KPiBodHRwOi8v d3d3LmlldGYub3JnL21lZXRpbmcvOTUvaW5kZXguaHRtbA0KPiANCj4gQWdhaW4sIHByZXNlbnRl cnMgYW5kIGRyYWZ0IGF1dGhvcnMgcGxlYXNlIGludm9rZSBhY3RpdmUgZGlzY3Vzc2lvbnMgaW4N Cj4gdGhlIEFOSU1BIGxpc3QuIFdlIGhhdmUgdmVyeSBsaW1pdGVkIHRpbWUgZm9yIGVhY2ggdG9w aWMgaW4gdGhlDQo+IGZhY2UtdG8tZmFjZSBtZWV0aW5nLiBNYWlsIGxpc3QgaXMgYSB2ZXJ5IGdv b2QgcGxhY2UgdG8gZGlzY3VzcyBhbmQgcmVhY2gNCj4gY29uc2Vuc3VzIG9uIHRlY2huaWNhbCBp c3N1ZXMuDQo+IA0KPiAyMDE2LTAzLTA0IChGcmlkYXkpOiBQcmVsaW1pbmFyeSBhZ2VuZGEgcHVi bGlzaGVkIGZvciBjb21tZW50Lg0KPiAyMDE2LTAzLTExIChGcmlkYXkpOiBGaW5hbCBhZ2VuZGEg dG8gYmUgcHVibGlzaGVkLg0KPiANCj4gQmVzdCByZWdhcmRzLA0KPiANCj4gVG9lcmxlc3MgJiBT aGVuZw0KPiANCj4gDQo+IF9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fDQo+IEFuaW1hIG1haWxpbmcgbGlzdA0KPiBBbmltYUBpZXRmLm9yZw0KPiBodHRwczov L3d3dy5pZXRmLm9yZy9tYWlsbWFuL2xpc3RpbmZvL2FuaW1hDQoNCg== From nobody Fri Mar 4 01:39:11 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 569721B358E for ; Fri, 4 Mar 2016 01:39:10 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.207 X-Spam-Level: X-Spam-Status: No, score=-4.207 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.006, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LlYb2IPOSWj4 for ; Fri, 4 Mar 2016 01:39:07 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 7C4CD1A3BA0 for ; Fri, 4 Mar 2016 01:39:05 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml402-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CFI17419; Fri, 04 Mar 2016 09:39:03 +0000 (GMT) Received: from NKGEML411-HUB.china.huawei.com (10.98.56.70) by lhreml402-hub.china.huawei.com (10.201.5.241) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 4 Mar 2016 09:39:01 +0000 Received: from NKGEML515-MBS.china.huawei.com ([169.254.5.102]) by nkgeml411-hub.china.huawei.com ([10.98.56.70]) with mapi id 14.03.0235.001; Fri, 4 Mar 2016 17:38:57 +0800 From: Sheng Jiang To: "Max Pritikin (pritikin)" Thread-Topic: [Anima] Call for agenda ANIMA @ IETF 95, Buenos Aires, Argentina Thread-Index: AdFvqy/RjdZlBxVOSoyQ+hwkjsVZcQFeceMAADTjdCA= Date: Fri, 4 Mar 2016 09:38:56 +0000 Message-ID: <5D36713D8A4E7348A7E10DF7437A4B927C636085@NKGEML515-MBS.china.huawei.com> References: <5D36713D8A4E7348A7E10DF7437A4B927C62CF54@NKGEML515-MBX.china.huawei.com> <5DDF4BE8-A790-4C00-825F-C686EE74664F@cisco.com> In-Reply-To: <5DDF4BE8-A790-4C00-825F-C686EE74664F@cisco.com> Accept-Language: en-GB, zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.99.197] Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A090204.56D957B7.00F4, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=169.254.5.102, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: bc0390c33b7f1f9b8e1f67958225cd5b Archived-At: Cc: "anima@ietf.org" Subject: Re: [Anima] Call for agenda ANIMA @ IETF 95, Buenos Aires, Argentina X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 04 Mar 2016 09:39:10 -0000 SGksIE1heCwNCg0KVGhhbmtzIGZvciB5b3VyIGVtYWlsLiBZb3VyIHJlcXVlc3RzIGhhdmUgYmVl biByZWNvcmRlZCBhbmQgbGlzdGVkIGluIHRoZSBhZ2VuZGEuIFdlIGhhdmUgdmVyeSBjcm93ZGVk IGFnZW5kYSByZXF1ZXN0cyBmb3IgdGhlIGNvbWluZyBtZWV0aW5nLiBJcyAxMCBtaW51dGVzIGVu b3VnaCBmb3IgeW91ciB0b3BpYz8gSW4gcHJpbmNpcGxlLCBhcyBXRyBkb2N1bWVudCwgeW91IGNh biBhbmQgc2hvdWxkIGdldCBtb3JlIHRpbWUuIEJ1dCBpZiB5b3UgY291bGQgc2F2ZSBzb21lIHRp bWUsIGl0IHdvdWxkIGJlIGhlbHBmdWwgZm9yIGNoYWlycy4gVGhhbmtzIGZvciB5b3VyIHVuZGVy c3RhbmRpbmcuDQoNClJlZ2FyZHMsDQoNClNoZW5nDQoNCj4tLS0tLU9yaWdpbmFsIE1lc3NhZ2Ut LS0tLQ0KPkZyb206IE1heCBQcml0aWtpbiAocHJpdGlraW4pIFttYWlsdG86cHJpdGlraW5AY2lz Y28uY29tXQ0KPlNlbnQ6IFRodXJzZGF5LCBNYXJjaCAwMywgMjAxNiAxMDoxNiBBTQ0KPlRvOiBT aGVuZyBKaWFuZw0KPkNjOiBhbmltYUBpZXRmLm9yZw0KPlN1YmplY3Q6IFJlOiBbQW5pbWFdIENh bGwgZm9yIGFnZW5kYSBBTklNQSBAIElFVEYgOTUsIEJ1ZW5vcyBBaXJlcywNCj5BcmdlbnRpbmEN Cj4NCj5QbGVhc2UgaW5jbHVkZSBhIHNsb3QgZm9yIGJvb3RzdHJhcHBpbmc6DQo+DQo+TmFtZSBv ZiB0aW1lIHNsb3Q6IGJvb3RzdHJhcHBpbmcNCj5OYW1lIG9mIGRyYWZ0KHMpOiBkcmFmdC1pZXRm LWFuaW1hLWJvb3RzdHJhcHBpbmcta2V5aW5mcmENCj5UaW1lIHJlcXVlc3RlZDogW3lvdXIgY2hv aWNlXQ0KPlByZXNlbnRlciBuYW1lKHMpOiBBdXRob3JzLiBJ4oCZbGwgdGFrZSBpdCBpZiBubyBv bmUgZWxzZSBkb2VzLg0KPkJyaWVmIGRlc2NyaXB0aW9uIG9mIHdoYXQgaXNzdWVzIG5lZWQgZGlz Y3Vzc2luZyBhbmQgd2hhdCB5b3UgaG9wZSB0bw0KPmFjY29tcGxpc2ggYnkgcHJlc2VudGluZyAo cGxlYXNlIGZvY3VzIG9uIG9wZW4gaXNzdWVzLCByYXRoZXIgdGhhbiBhDQo+c3RhdHVzIHVwZGF0 ZSk6DQo+DQo+CVN0YXR1cyB1cGRhdGUuDQo+CUV4cGxhaW4gcHJveHkgb3BlcmF0aW9ucy4NCj4J UHJvdG9jb2wgRGV0YWlsczogcmVxdWVzdCBmb3IgZmVlZGJhY2suDQo+CURpZmZlcmVuY2UgYmV0 d2VlbiDigJx0cnVzdGVkIHZlbmRvciBsb2dnaW5n4oCdIGFuZCDigJxjZXJ0aWZpZWQgbG9nZ2lu ZyINCj4NCj4tIG1heA0KPg0KPj4gT24gRmViIDI1LCAyMDE2LCBhdCAyOjAyIEFNLCBTaGVuZyBK aWFuZyA8amlhbmdzaGVuZ0BodWF3ZWkuY29tPg0KPndyb3RlOg0KPj4NCj4+IEhpLCBhbGwsDQo+ Pg0KPj4gV2UgaGF2ZSByZXF1ZXN0ZWQgdHdvIHNlc3Npb25zIG9mIDIgaG91ciArIDEuNSBob3Vy IGZvciB0aGUgQU5JTUEgV0cNCj4+IG1lZXRpbmcgZm9yIElFVEYtOTUgKEJ1ZW5vcyBBaXJlcywg QXJnZW50aW5hKSBhbmQgYXJlIHN0YXJ0aW5nIHRvIGNvbGxlY3QNCj4+IGFnZW5kYSBpdGVtcyBm b3IgdGhlc2Ugc2Vzc2lvbnMuIFNhbWUgd2l0aCBJRVRGOTQsIHdlIGRvIG5vdCByZXNlcnZlDQo+ PiB0aGUgc2Vjb25kIHNlc3Npb24gZm9yIG5vbi1jaGFydGVyIHdvcmsgaXRlbXMuIEl0IHNheXMg dGhlIGNoYXJ0ZXIgd29yaw0KPj4gaXRlbSB3b3VsZCBoYXZlIHByaW9yaXR5LiBXZSB3aWxsIG9u bHkgZGlzY3VzcyBub24tY2hhcnRlciB3b3JrIGl0ZW1zDQo+PiB0aWxsIHdlIGZpbmlzaCBjaGFy dGVyIHdvcmsgaXRlbXMuIFRoZSBwcmlvcml0eSBhbW9uZyB0aGVzZSBub24tY2hhcnRlcg0KPj4g d29yayBpdGVtcyB3b3VsZCBiZTogdGhlc2UgdGhhdCBoYXZlIGFjdGl2ZSBkaXNjdXNzaW9uIGlu IG1haWwgbGlzdCwNCj4+IHRoZW4gdGhlc2UgaGF2ZSBzdWJtaXR0ZWQgZHJhZnRzLCBhbmQgdG9w aWNzIHdpdGhvdXQgZHJhZnRzLg0KPj4NCj4+IFBsZWFzZSBzZW5kIHVzIChhbmltYS1jaGFpcnNA dG9vbHMuaWV0Zi5vcmcpIHJlcXVlc3RzIGZvciBzZXNzaW9ucyBieQ0KPj4gTWFyY2ggM3JkLCBU aHVyc2RheSAoSUVURiByZXF1ZXN0cyB0aGUgcHJlbGltaW5hcnkgYWdlbmRhIGJ5IE1hcmNoDQo+ PiA0dGgsIEZyaWRheSkgYW5kIGluY2x1ZGU6DQo+Pg0KPj4gTmFtZSBvZiB0aW1lIHNsb3Q6DQo+ PiBOYW1lIG9mIGRyYWZ0KHMpOg0KPj4gVGltZSByZXF1ZXN0ZWQ6DQo+PiBQcmVzZW50ZXIgbmFt ZShzKToNCj4+IEJyaWVmIGRlc2NyaXB0aW9uIG9mIHdoYXQgaXNzdWVzIG5lZWQgZGlzY3Vzc2lu ZyBhbmQgd2hhdCB5b3UgaG9wZSB0bw0KPj4gYWNjb21wbGlzaCBieSBwcmVzZW50aW5nIChwbGVh c2UgZm9jdXMgb24gb3BlbiBpc3N1ZXMsIHJhdGhlciB0aGFuIGENCj4+IHN0YXR1cyB1cGRhdGUp Og0KPj4NCj4+IE1vcmUgZGV0YWlscyBhYm91dCB0aGUgSUVURiA5NSwgQnVlbm9zIEFpcmVzLCBB cmdlbnRpbmEgY2FuIGJlDQo+PiBmb3VuZCBhdDoNCj4+IGh0dHA6Ly93d3cuaWV0Zi5vcmcvbWVl dGluZy85NS9pbmRleC5odG1sDQo+Pg0KPj4gQWdhaW4sIHByZXNlbnRlcnMgYW5kIGRyYWZ0IGF1 dGhvcnMgcGxlYXNlIGludm9rZSBhY3RpdmUgZGlzY3Vzc2lvbnMgaW4NCj4+IHRoZSBBTklNQSBs aXN0LiBXZSBoYXZlIHZlcnkgbGltaXRlZCB0aW1lIGZvciBlYWNoIHRvcGljIGluIHRoZQ0KPj4g ZmFjZS10by1mYWNlIG1lZXRpbmcuIE1haWwgbGlzdCBpcyBhIHZlcnkgZ29vZCBwbGFjZSB0byBk aXNjdXNzIGFuZCByZWFjaA0KPj4gY29uc2Vuc3VzIG9uIHRlY2huaWNhbCBpc3N1ZXMuDQo+Pg0K Pj4gMjAxNi0wMy0wNCAoRnJpZGF5KTogUHJlbGltaW5hcnkgYWdlbmRhIHB1Ymxpc2hlZCBmb3Ig Y29tbWVudC4NCj4+IDIwMTYtMDMtMTEgKEZyaWRheSk6IEZpbmFsIGFnZW5kYSB0byBiZSBwdWJs aXNoZWQuDQo+Pg0KPj4gQmVzdCByZWdhcmRzLA0KPj4NCj4+IFRvZXJsZXNzICYgU2hlbmcNCj4+ DQo+Pg0KPj4gX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18N Cj4+IEFuaW1hIG1haWxpbmcgbGlzdA0KPj4gQW5pbWFAaWV0Zi5vcmcNCj4+IGh0dHBzOi8vd3d3 LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vYW5pbWENCg0K From nobody Fri Mar 4 08:01:07 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6F4E51A6F30 for ; Fri, 4 Mar 2016 08:01:06 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.502 X-Spam-Level: X-Spam-Status: No, score=-14.502 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BxcdYRUlbtVt for ; Fri, 4 Mar 2016 08:01:04 -0800 (PST) Received: from alln-iport-6.cisco.com (alln-iport-6.cisco.com [173.37.142.93]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 51BB31A6F3F for ; Fri, 4 Mar 2016 08:01:03 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=4968; q=dns/txt; s=iport; t=1457107263; x=1458316863; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-id:content-transfer-encoding: mime-version; bh=nX5RnJ3tIX3PJyeT+e71Uyr+JJsNEhBjbBr6B2M4SeM=; b=InV6k+jmrprL31t0NzKaPiS0+RAlosUBMWUV/AkmivBG8EA92zcNk5pg zmSdhwEp01UckIqFgNYkORSIVQ6LVufpI/KVz/aLp2T/AACX3mMRrazt7 GUc3s0kfnN1dwIUId76kxwjDOPkTAESgG+Xf/RYrgKtmjb1SUfzOpzEI1 M=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AEAgBLsNlW/4gNJK1dgzpSbQa6NAENg?= =?us-ascii?q?WkXCoVuAhyBGjgUAQEBAQEBAWQnhEEBAQEDAQEBARoGEToLBQcEAgEIEQQBAQE?= =?us-ascii?q?CAiMDAgICJQsUAQgIAQEEAQ0FCYgRCA6uaY57AQEBAQEBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEBFXuHCgiCRoQJCgYCARsYgmorgQ8FjS6JcwGNZ4FihESIU4V3iFoBHgEBQoM?= =?us-ascii?q?QVWoBh3I9fgEBAQ?= X-IronPort-AV: E=Sophos;i="5.22,536,1449532800"; d="scan'208";a="245610440" Received: from alln-core-3.cisco.com ([173.36.13.136]) by alln-iport-6.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 04 Mar 2016 16:01:02 +0000 Received: from XCH-RCD-001.cisco.com (xch-rcd-001.cisco.com [173.37.102.11]) by alln-core-3.cisco.com (8.14.5/8.14.5) with ESMTP id u24G11Pe014451 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 4 Mar 2016 16:01:02 GMT Received: from xch-aln-013.cisco.com (173.36.7.23) by XCH-RCD-001.cisco.com (173.37.102.11) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 4 Mar 2016 10:01:00 -0600 Received: from xch-aln-013.cisco.com ([173.36.7.23]) by XCH-ALN-013.cisco.com ([173.36.7.23]) with mapi id 15.00.1104.009; Fri, 4 Mar 2016 10:01:00 -0600 From: "Max Pritikin (pritikin)" To: Sheng Jiang , "Toerless Eckert (eckert)" Thread-Topic: [Anima] Call for agenda ANIMA @ IETF 95, Buenos Aires, Argentina Thread-Index: AdFvqy/RjdZlBxVOSoyQ+hwkjsVZcQFeceMAADTjdCAAGjQZgA== Date: Fri, 4 Mar 2016 16:01:00 +0000 Message-ID: <76C23375-4A78-4DDC-8061-9508F101D2C1@cisco.com> References: <5D36713D8A4E7348A7E10DF7437A4B927C62CF54@NKGEML515-MBX.china.huawei.com> <5DDF4BE8-A790-4C00-825F-C686EE74664F@cisco.com> <5D36713D8A4E7348A7E10DF7437A4B927C636085@NKGEML515-MBS.china.huawei.com> In-Reply-To: <5D36713D8A4E7348A7E10DF7437A4B927C636085@NKGEML515-MBS.china.huawei.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-messagesentrepresentingtype: 1 x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.99.106.6] Content-Type: text/plain; charset="utf-8" Content-ID: <8F796F545B5E8849B172B54914768936@emea.cisco.com> Content-Transfer-Encoding: base64 MIME-Version: 1.0 Archived-At: Cc: "anima@ietf.org" Subject: Re: [Anima] Call for agenda ANIMA @ IETF 95, Buenos Aires, Argentina X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 04 Mar 2016 16:01:06 -0000 SSB0aGluayBpdCB3b3VsZCBiZSBoZWxwZnVsIHRvIGhhdmUgbW9yZSB0aW1lLiAxMG1pbiBpcyBl bm91Z2ggZm9yIGEgc3RhdHVzIHJlcG9ydCBidXQgbm90IGxpa2VseSBlbm91Z2ggZm9yIOKAnG9w ZW4gaXNzdWVz4oCdIHdoaWNoIG5lZWQgdG8gYmUgZXhwbGFpbmVkIGlmIGRpc2N1c3Npb24gaXMg dG8gaGFwcGVuKS4gDQoNClJlZ2FyZGxlc3MgeW91IGFyZSB0aGUgY2hhaXJzLCBzbyBmaWd1cmUg b3V0IHdoYXQgaXMgYXZhaWxhYmxlLiBXb3JzZSBjYXNlIEnigJlsbCBwb2ludCBvdXQgdGhlIGlz c3VlcyBhbmQgc29saWNpdGVkIGZvbGtzIHRvIGpvaW4gYm9vdHN0cmFwcGluZyBkZXNpZ24gdGVh bSBjYWxscy4gDQoNCi0gbWF4DQoNCj4gT24gTWFyIDQsIDIwMTYsIGF0IDI6MzggQU0sIFNoZW5n IEppYW5nIDxqaWFuZ3NoZW5nQGh1YXdlaS5jb20+IHdyb3RlOg0KPiANCj4gSGksIE1heCwNCj4g DQo+IFRoYW5rcyBmb3IgeW91ciBlbWFpbC4gWW91ciByZXF1ZXN0cyBoYXZlIGJlZW4gcmVjb3Jk ZWQgYW5kIGxpc3RlZCBpbiB0aGUgYWdlbmRhLiBXZSBoYXZlIHZlcnkgY3Jvd2RlZCBhZ2VuZGEg cmVxdWVzdHMgZm9yIHRoZSBjb21pbmcgbWVldGluZy4gSXMgMTAgbWludXRlcyBlbm91Z2ggZm9y IHlvdXIgdG9waWM/IEluIHByaW5jaXBsZSwgYXMgV0cgZG9jdW1lbnQsIHlvdSBjYW4gYW5kIHNo b3VsZCBnZXQgbW9yZSB0aW1lLiBCdXQgaWYgeW91IGNvdWxkIHNhdmUgc29tZSB0aW1lLCBpdCB3 b3VsZCBiZSBoZWxwZnVsIGZvciBjaGFpcnMuIFRoYW5rcyBmb3IgeW91ciB1bmRlcnN0YW5kaW5n Lg0KPiANCj4gUmVnYXJkcywNCj4gDQo+IFNoZW5nDQo+IA0KPj4gLS0tLS1PcmlnaW5hbCBNZXNz YWdlLS0tLS0NCj4+IEZyb206IE1heCBQcml0aWtpbiAocHJpdGlraW4pIFttYWlsdG86cHJpdGlr aW5AY2lzY28uY29tXQ0KPj4gU2VudDogVGh1cnNkYXksIE1hcmNoIDAzLCAyMDE2IDEwOjE2IEFN DQo+PiBUbzogU2hlbmcgSmlhbmcNCj4+IENjOiBhbmltYUBpZXRmLm9yZw0KPj4gU3ViamVjdDog UmU6IFtBbmltYV0gQ2FsbCBmb3IgYWdlbmRhIEFOSU1BIEAgSUVURiA5NSwgQnVlbm9zIEFpcmVz LA0KPj4gQXJnZW50aW5hDQo+PiANCj4+IFBsZWFzZSBpbmNsdWRlIGEgc2xvdCBmb3IgYm9vdHN0 cmFwcGluZzoNCj4+IA0KPj4gTmFtZSBvZiB0aW1lIHNsb3Q6IGJvb3RzdHJhcHBpbmcNCj4+IE5h bWUgb2YgZHJhZnQocyk6IGRyYWZ0LWlldGYtYW5pbWEtYm9vdHN0cmFwcGluZy1rZXlpbmZyYQ0K Pj4gVGltZSByZXF1ZXN0ZWQ6IFt5b3VyIGNob2ljZV0NCj4+IFByZXNlbnRlciBuYW1lKHMpOiBB dXRob3JzLiBJ4oCZbGwgdGFrZSBpdCBpZiBubyBvbmUgZWxzZSBkb2VzLg0KPj4gQnJpZWYgZGVz Y3JpcHRpb24gb2Ygd2hhdCBpc3N1ZXMgbmVlZCBkaXNjdXNzaW5nIGFuZCB3aGF0IHlvdSBob3Bl IHRvDQo+PiBhY2NvbXBsaXNoIGJ5IHByZXNlbnRpbmcgKHBsZWFzZSBmb2N1cyBvbiBvcGVuIGlz c3VlcywgcmF0aGVyIHRoYW4gYQ0KPj4gc3RhdHVzIHVwZGF0ZSk6DQo+PiANCj4+IAlTdGF0dXMg dXBkYXRlLg0KPj4gCUV4cGxhaW4gcHJveHkgb3BlcmF0aW9ucy4NCj4+IAlQcm90b2NvbCBEZXRh aWxzOiByZXF1ZXN0IGZvciBmZWVkYmFjay4NCj4+IAlEaWZmZXJlbmNlIGJldHdlZW4g4oCcdHJ1 c3RlZCB2ZW5kb3IgbG9nZ2luZ+KAnSBhbmQg4oCcY2VydGlmaWVkIGxvZ2dpbmciDQo+PiANCj4+ IC0gbWF4DQo+PiANCj4+PiBPbiBGZWIgMjUsIDIwMTYsIGF0IDI6MDIgQU0sIFNoZW5nIEppYW5n IDxqaWFuZ3NoZW5nQGh1YXdlaS5jb20+DQo+PiB3cm90ZToNCj4+PiANCj4+PiBIaSwgYWxsLA0K Pj4+IA0KPj4+IFdlIGhhdmUgcmVxdWVzdGVkIHR3byBzZXNzaW9ucyBvZiAyIGhvdXIgKyAxLjUg aG91ciBmb3IgdGhlIEFOSU1BIFdHDQo+Pj4gbWVldGluZyBmb3IgSUVURi05NSAoQnVlbm9zIEFp cmVzLCBBcmdlbnRpbmEpIGFuZCBhcmUgc3RhcnRpbmcgdG8gY29sbGVjdA0KPj4+IGFnZW5kYSBp dGVtcyBmb3IgdGhlc2Ugc2Vzc2lvbnMuIFNhbWUgd2l0aCBJRVRGOTQsIHdlIGRvIG5vdCByZXNl cnZlDQo+Pj4gdGhlIHNlY29uZCBzZXNzaW9uIGZvciBub24tY2hhcnRlciB3b3JrIGl0ZW1zLiBJ dCBzYXlzIHRoZSBjaGFydGVyIHdvcmsNCj4+PiBpdGVtIHdvdWxkIGhhdmUgcHJpb3JpdHkuIFdl IHdpbGwgb25seSBkaXNjdXNzIG5vbi1jaGFydGVyIHdvcmsgaXRlbXMNCj4+PiB0aWxsIHdlIGZp bmlzaCBjaGFydGVyIHdvcmsgaXRlbXMuIFRoZSBwcmlvcml0eSBhbW9uZyB0aGVzZSBub24tY2hh cnRlcg0KPj4+IHdvcmsgaXRlbXMgd291bGQgYmU6IHRoZXNlIHRoYXQgaGF2ZSBhY3RpdmUgZGlz Y3Vzc2lvbiBpbiBtYWlsIGxpc3QsDQo+Pj4gdGhlbiB0aGVzZSBoYXZlIHN1Ym1pdHRlZCBkcmFm dHMsIGFuZCB0b3BpY3Mgd2l0aG91dCBkcmFmdHMuDQo+Pj4gDQo+Pj4gUGxlYXNlIHNlbmQgdXMg KGFuaW1hLWNoYWlyc0B0b29scy5pZXRmLm9yZykgcmVxdWVzdHMgZm9yIHNlc3Npb25zIGJ5DQo+ Pj4gTWFyY2ggM3JkLCBUaHVyc2RheSAoSUVURiByZXF1ZXN0cyB0aGUgcHJlbGltaW5hcnkgYWdl bmRhIGJ5IE1hcmNoDQo+Pj4gNHRoLCBGcmlkYXkpIGFuZCBpbmNsdWRlOg0KPj4+IA0KPj4+IE5h bWUgb2YgdGltZSBzbG90Og0KPj4+IE5hbWUgb2YgZHJhZnQocyk6DQo+Pj4gVGltZSByZXF1ZXN0 ZWQ6DQo+Pj4gUHJlc2VudGVyIG5hbWUocyk6DQo+Pj4gQnJpZWYgZGVzY3JpcHRpb24gb2Ygd2hh dCBpc3N1ZXMgbmVlZCBkaXNjdXNzaW5nIGFuZCB3aGF0IHlvdSBob3BlIHRvDQo+Pj4gYWNjb21w bGlzaCBieSBwcmVzZW50aW5nIChwbGVhc2UgZm9jdXMgb24gb3BlbiBpc3N1ZXMsIHJhdGhlciB0 aGFuIGENCj4+PiBzdGF0dXMgdXBkYXRlKToNCj4+PiANCj4+PiBNb3JlIGRldGFpbHMgYWJvdXQg dGhlIElFVEYgOTUsIEJ1ZW5vcyBBaXJlcywgQXJnZW50aW5hIGNhbiBiZQ0KPj4+IGZvdW5kIGF0 Og0KPj4+IGh0dHA6Ly93d3cuaWV0Zi5vcmcvbWVldGluZy85NS9pbmRleC5odG1sDQo+Pj4gDQo+ Pj4gQWdhaW4sIHByZXNlbnRlcnMgYW5kIGRyYWZ0IGF1dGhvcnMgcGxlYXNlIGludm9rZSBhY3Rp dmUgZGlzY3Vzc2lvbnMgaW4NCj4+PiB0aGUgQU5JTUEgbGlzdC4gV2UgaGF2ZSB2ZXJ5IGxpbWl0 ZWQgdGltZSBmb3IgZWFjaCB0b3BpYyBpbiB0aGUNCj4+PiBmYWNlLXRvLWZhY2UgbWVldGluZy4g TWFpbCBsaXN0IGlzIGEgdmVyeSBnb29kIHBsYWNlIHRvIGRpc2N1c3MgYW5kIHJlYWNoDQo+Pj4g Y29uc2Vuc3VzIG9uIHRlY2huaWNhbCBpc3N1ZXMuDQo+Pj4gDQo+Pj4gMjAxNi0wMy0wNCAoRnJp ZGF5KTogUHJlbGltaW5hcnkgYWdlbmRhIHB1Ymxpc2hlZCBmb3IgY29tbWVudC4NCj4+PiAyMDE2 LTAzLTExIChGcmlkYXkpOiBGaW5hbCBhZ2VuZGEgdG8gYmUgcHVibGlzaGVkLg0KPj4+IA0KPj4+ IEJlc3QgcmVnYXJkcywNCj4+PiANCj4+PiBUb2VybGVzcyAmIFNoZW5nDQo+Pj4gDQo+Pj4gDQo+ Pj4gX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18NCj4+PiBB bmltYSBtYWlsaW5nIGxpc3QNCj4+PiBBbmltYUBpZXRmLm9yZw0KPj4+IGh0dHBzOi8vd3d3Lmll dGYub3JnL21haWxtYW4vbGlzdGluZm8vYW5pbWENCj4gDQoNCg== From nobody Sun Mar 6 18:26:08 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 25B361B33BA for ; Sun, 6 Mar 2016 18:26:07 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.202 X-Spam-Level: X-Spam-Status: No, score=-4.202 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IetuYiKKKzVc for ; Sun, 6 Mar 2016 18:26:03 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 160B31B33B7 for ; Sun, 6 Mar 2016 18:26:02 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml707-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CFK31117; Mon, 07 Mar 2016 02:24:25 +0000 (GMT) Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml707-cah.china.huawei.com (10.201.5.199) with Microsoft SMTP Server (TLS) id 14.3.235.1; Mon, 7 Mar 2016 02:24:24 +0000 Received: from NKGEML515-MBS.china.huawei.com ([169.254.5.102]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0235.001; Mon, 7 Mar 2016 10:24:20 +0800 From: Sheng Jiang To: "Max Pritikin (pritikin)" , "Toerless Eckert (eckert)" Thread-Topic: [Anima] Call for agenda ANIMA @ IETF 95, Buenos Aires, Argentina Thread-Index: AdFvqy/RjdZlBxVOSoyQ+hwkjsVZcQFeceMAADTjdCAAGjQZgABtdKLQ Date: Mon, 7 Mar 2016 02:24:20 +0000 Message-ID: <5D36713D8A4E7348A7E10DF7437A4B927C637D40@NKGEML515-MBS.china.huawei.com> References: <5D36713D8A4E7348A7E10DF7437A4B927C62CF54@NKGEML515-MBX.china.huawei.com> <5DDF4BE8-A790-4C00-825F-C686EE74664F@cisco.com> <5D36713D8A4E7348A7E10DF7437A4B927C636085@NKGEML515-MBS.china.huawei.com> <76C23375-4A78-4DDC-8061-9508F101D2C1@cisco.com> In-Reply-To: <76C23375-4A78-4DDC-8061-9508F101D2C1@cisco.com> Accept-Language: en-GB, zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.99.197] Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown, refid=x_failed, ip=unknown, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32, mode=multiengine Archived-At: Cc: "anima@ietf.org" Subject: Re: [Anima] Call for agenda ANIMA @ IETF 95, Buenos Aires, Argentina X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 07 Mar 2016 02:26:07 -0000 SGksIE1heCwNCg0KRm9yIG5vdywgMTUgbWlucyBoYXZlIGJlZW4gYWxsb2NhdGVkIGZvciB5b3Vy IHRvcGljLiBCdXQgd2UgZG8gZW5jb3VyYWdlIHlvdSB0byBhdCBsZWFzdCBtZW50aW9uIG9wZW4g aXNzdWVzIGluIHRoZSBtZWV0aW5nLiBZb3UgY291bGQgYWxzbyB0YWtlIHNvbWUgdGltZSB0byBk aXNjdXNzIHRoZW0sIGJ1dCB3aXRoIGxpbWl0ZWQgdGltZS4gV2UgYWxzbyBlbmNvdXJhZ2UgeW91 IHRvIHdvcmsgb2ZmbGluZSBidXQgb24tc2l0ZSB3aXRoIG90aGVyIGludGVyZXN0ZWQgcGFydGlj aXBhbnRzLiBUaGF0J3Mgb25lIG9mIHRoZSByZWFzb25zIHdlIGhhdmUgdHdvIHNlc3Npb25zLCB0 aGUgZmlyc3Qgb24gTW9uZGF5IGFuZCB0aGUgc2Vjb25kIG9uIFRodXJzZGF5LCB3aGljaCBhbGxv dyBXRyBkb2N1bWVudHMgdG8gbWFrZSBwcm9ncmVzc2VzIG9uIHNpdGUgYW5kIGJyaWVmbHkgcmVw b3J0IHRoZSBwcm9ncmVzcyBvbiBUaHVyc2RheS4NCg0KUmVnYXJkcywNCg0KU2hlbmcNCg0KPi0t LS0tT3JpZ2luYWwgTWVzc2FnZS0tLS0tDQo+RnJvbTogTWF4IFByaXRpa2luIChwcml0aWtpbikg W21haWx0bzpwcml0aWtpbkBjaXNjby5jb21dDQo+U2VudDogU2F0dXJkYXksIE1hcmNoIDA1LCAy MDE2IDEyOjAxIEFNDQo+VG86IFNoZW5nIEppYW5nOyBUb2VybGVzcyBFY2tlcnQgKGVja2VydCkN Cj5DYzogYW5pbWFAaWV0Zi5vcmcNCj5TdWJqZWN0OiBSZTogW0FuaW1hXSBDYWxsIGZvciBhZ2Vu ZGEgQU5JTUEgQCBJRVRGIDk1LCBCdWVub3MgQWlyZXMsDQo+QXJnZW50aW5hDQo+DQo+SSB0aGlu ayBpdCB3b3VsZCBiZSBoZWxwZnVsIHRvIGhhdmUgbW9yZSB0aW1lLiAxMG1pbiBpcyBlbm91Z2gg Zm9yIGEgc3RhdHVzDQo+cmVwb3J0IGJ1dCBub3QgbGlrZWx5IGVub3VnaCBmb3Ig4oCcb3BlbiBp c3N1ZXPigJ0gd2hpY2ggbmVlZCB0byBiZSBleHBsYWluZWQgaWYNCj5kaXNjdXNzaW9uIGlzIHRv IGhhcHBlbikuDQo+DQo+UmVnYXJkbGVzcyB5b3UgYXJlIHRoZSBjaGFpcnMsIHNvIGZpZ3VyZSBv dXQgd2hhdCBpcyBhdmFpbGFibGUuIFdvcnNlIGNhc2UgSeKAmWxsDQo+cG9pbnQgb3V0IHRoZSBp c3N1ZXMgYW5kIHNvbGljaXRlZCBmb2xrcyB0byBqb2luIGJvb3RzdHJhcHBpbmcgZGVzaWduIHRl YW0gY2FsbHMuDQo+DQo+LSBtYXgNCj4NCj4+IE9uIE1hciA0LCAyMDE2LCBhdCAyOjM4IEFNLCBT aGVuZyBKaWFuZyA8amlhbmdzaGVuZ0BodWF3ZWkuY29tPiB3cm90ZToNCj4+DQo+PiBIaSwgTWF4 LA0KPj4NCj4+IFRoYW5rcyBmb3IgeW91ciBlbWFpbC4gWW91ciByZXF1ZXN0cyBoYXZlIGJlZW4g cmVjb3JkZWQgYW5kIGxpc3RlZCBpbiB0aGUNCj5hZ2VuZGEuIFdlIGhhdmUgdmVyeSBjcm93ZGVk IGFnZW5kYSByZXF1ZXN0cyBmb3IgdGhlIGNvbWluZyBtZWV0aW5nLiBJcw0KPjEwIG1pbnV0ZXMg ZW5vdWdoIGZvciB5b3VyIHRvcGljPyBJbiBwcmluY2lwbGUsIGFzIFdHIGRvY3VtZW50LCB5b3Ug Y2FuIGFuZA0KPnNob3VsZCBnZXQgbW9yZSB0aW1lLiBCdXQgaWYgeW91IGNvdWxkIHNhdmUgc29t ZSB0aW1lLCBpdCB3b3VsZCBiZSBoZWxwZnVsIGZvcg0KPmNoYWlycy4gVGhhbmtzIGZvciB5b3Vy IHVuZGVyc3RhbmRpbmcuDQo+Pg0KPj4gUmVnYXJkcywNCj4+DQo+PiBTaGVuZw0KPj4NCj4+PiAt LS0tLU9yaWdpbmFsIE1lc3NhZ2UtLS0tLQ0KPj4+IEZyb206IE1heCBQcml0aWtpbiAocHJpdGlr aW4pIFttYWlsdG86cHJpdGlraW5AY2lzY28uY29tXQ0KPj4+IFNlbnQ6IFRodXJzZGF5LCBNYXJj aCAwMywgMjAxNiAxMDoxNiBBTQ0KPj4+IFRvOiBTaGVuZyBKaWFuZw0KPj4+IENjOiBhbmltYUBp ZXRmLm9yZw0KPj4+IFN1YmplY3Q6IFJlOiBbQW5pbWFdIENhbGwgZm9yIGFnZW5kYSBBTklNQSBA IElFVEYgOTUsIEJ1ZW5vcyBBaXJlcywNCj4+PiBBcmdlbnRpbmENCj4+Pg0KPj4+IFBsZWFzZSBp bmNsdWRlIGEgc2xvdCBmb3IgYm9vdHN0cmFwcGluZzoNCj4+Pg0KPj4+IE5hbWUgb2YgdGltZSBz bG90OiBib290c3RyYXBwaW5nDQo+Pj4gTmFtZSBvZiBkcmFmdChzKTogZHJhZnQtaWV0Zi1hbmlt YS1ib290c3RyYXBwaW5nLWtleWluZnJhDQo+Pj4gVGltZSByZXF1ZXN0ZWQ6IFt5b3VyIGNob2lj ZV0NCj4+PiBQcmVzZW50ZXIgbmFtZShzKTogQXV0aG9ycy4gSeKAmWxsIHRha2UgaXQgaWYgbm8g b25lIGVsc2UgZG9lcy4NCj4+PiBCcmllZiBkZXNjcmlwdGlvbiBvZiB3aGF0IGlzc3VlcyBuZWVk IGRpc2N1c3NpbmcgYW5kIHdoYXQgeW91IGhvcGUgdG8NCj4+PiBhY2NvbXBsaXNoIGJ5IHByZXNl bnRpbmcgKHBsZWFzZSBmb2N1cyBvbiBvcGVuIGlzc3VlcywgcmF0aGVyIHRoYW4gYQ0KPj4+IHN0 YXR1cyB1cGRhdGUpOg0KPj4+DQo+Pj4gCVN0YXR1cyB1cGRhdGUuDQo+Pj4gCUV4cGxhaW4gcHJv eHkgb3BlcmF0aW9ucy4NCj4+PiAJUHJvdG9jb2wgRGV0YWlsczogcmVxdWVzdCBmb3IgZmVlZGJh Y2suDQo+Pj4gCURpZmZlcmVuY2UgYmV0d2VlbiDigJx0cnVzdGVkIHZlbmRvciBsb2dnaW5n4oCd IGFuZCDigJxjZXJ0aWZpZWQgbG9nZ2luZyINCj4+Pg0KPj4+IC0gbWF4DQo+Pj4NCj4+Pj4gT24g RmViIDI1LCAyMDE2LCBhdCAyOjAyIEFNLCBTaGVuZyBKaWFuZyA8amlhbmdzaGVuZ0BodWF3ZWku Y29tPg0KPj4+IHdyb3RlOg0KPj4+Pg0KPj4+PiBIaSwgYWxsLA0KPj4+Pg0KPj4+PiBXZSBoYXZl IHJlcXVlc3RlZCB0d28gc2Vzc2lvbnMgb2YgMiBob3VyICsgMS41IGhvdXIgZm9yIHRoZSBBTklN QSBXRw0KPj4+PiBtZWV0aW5nIGZvciBJRVRGLTk1IChCdWVub3MgQWlyZXMsIEFyZ2VudGluYSkg YW5kIGFyZSBzdGFydGluZyB0byBjb2xsZWN0DQo+Pj4+IGFnZW5kYSBpdGVtcyBmb3IgdGhlc2Ug c2Vzc2lvbnMuIFNhbWUgd2l0aCBJRVRGOTQsIHdlIGRvIG5vdCByZXNlcnZlDQo+Pj4+IHRoZSBz ZWNvbmQgc2Vzc2lvbiBmb3Igbm9uLWNoYXJ0ZXIgd29yayBpdGVtcy4gSXQgc2F5cyB0aGUgY2hh cnRlciB3b3JrDQo+Pj4+IGl0ZW0gd291bGQgaGF2ZSBwcmlvcml0eS4gV2Ugd2lsbCBvbmx5IGRp c2N1c3Mgbm9uLWNoYXJ0ZXIgd29yayBpdGVtcw0KPj4+PiB0aWxsIHdlIGZpbmlzaCBjaGFydGVy IHdvcmsgaXRlbXMuIFRoZSBwcmlvcml0eSBhbW9uZyB0aGVzZSBub24tY2hhcnRlcg0KPj4+PiB3 b3JrIGl0ZW1zIHdvdWxkIGJlOiB0aGVzZSB0aGF0IGhhdmUgYWN0aXZlIGRpc2N1c3Npb24gaW4g bWFpbCBsaXN0LA0KPj4+PiB0aGVuIHRoZXNlIGhhdmUgc3VibWl0dGVkIGRyYWZ0cywgYW5kIHRv cGljcyB3aXRob3V0IGRyYWZ0cy4NCj4+Pj4NCj4+Pj4gUGxlYXNlIHNlbmQgdXMgKGFuaW1hLWNo YWlyc0B0b29scy5pZXRmLm9yZykgcmVxdWVzdHMgZm9yIHNlc3Npb25zIGJ5DQo+Pj4+IE1hcmNo IDNyZCwgVGh1cnNkYXkgKElFVEYgcmVxdWVzdHMgdGhlIHByZWxpbWluYXJ5IGFnZW5kYSBieSBN YXJjaA0KPj4+PiA0dGgsIEZyaWRheSkgYW5kIGluY2x1ZGU6DQo+Pj4+DQo+Pj4+IE5hbWUgb2Yg dGltZSBzbG90Og0KPj4+PiBOYW1lIG9mIGRyYWZ0KHMpOg0KPj4+PiBUaW1lIHJlcXVlc3RlZDoN Cj4+Pj4gUHJlc2VudGVyIG5hbWUocyk6DQo+Pj4+IEJyaWVmIGRlc2NyaXB0aW9uIG9mIHdoYXQg aXNzdWVzIG5lZWQgZGlzY3Vzc2luZyBhbmQgd2hhdCB5b3UgaG9wZSB0bw0KPj4+PiBhY2NvbXBs aXNoIGJ5IHByZXNlbnRpbmcgKHBsZWFzZSBmb2N1cyBvbiBvcGVuIGlzc3VlcywgcmF0aGVyIHRo YW4gYQ0KPj4+PiBzdGF0dXMgdXBkYXRlKToNCj4+Pj4NCj4+Pj4gTW9yZSBkZXRhaWxzIGFib3V0 IHRoZSBJRVRGIDk1LCBCdWVub3MgQWlyZXMsIEFyZ2VudGluYSBjYW4gYmUNCj4+Pj4gZm91bmQg YXQ6DQo+Pj4+IGh0dHA6Ly93d3cuaWV0Zi5vcmcvbWVldGluZy85NS9pbmRleC5odG1sDQo+Pj4+ DQo+Pj4+IEFnYWluLCBwcmVzZW50ZXJzIGFuZCBkcmFmdCBhdXRob3JzIHBsZWFzZSBpbnZva2Ug YWN0aXZlIGRpc2N1c3Npb25zIGluDQo+Pj4+IHRoZSBBTklNQSBsaXN0LiBXZSBoYXZlIHZlcnkg bGltaXRlZCB0aW1lIGZvciBlYWNoIHRvcGljIGluIHRoZQ0KPj4+PiBmYWNlLXRvLWZhY2UgbWVl dGluZy4gTWFpbCBsaXN0IGlzIGEgdmVyeSBnb29kIHBsYWNlIHRvIGRpc2N1c3MgYW5kIHJlYWNo DQo+Pj4+IGNvbnNlbnN1cyBvbiB0ZWNobmljYWwgaXNzdWVzLg0KPj4+Pg0KPj4+PiAyMDE2LTAz LTA0IChGcmlkYXkpOiBQcmVsaW1pbmFyeSBhZ2VuZGEgcHVibGlzaGVkIGZvciBjb21tZW50Lg0K Pj4+PiAyMDE2LTAzLTExIChGcmlkYXkpOiBGaW5hbCBhZ2VuZGEgdG8gYmUgcHVibGlzaGVkLg0K Pj4+Pg0KPj4+PiBCZXN0IHJlZ2FyZHMsDQo+Pj4+DQo+Pj4+IFRvZXJsZXNzICYgU2hlbmcNCj4+ Pj4NCj4+Pj4NCj4+Pj4gX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX18NCj4+Pj4gQW5pbWEgbWFpbGluZyBsaXN0DQo+Pj4+IEFuaW1hQGlldGYub3JnDQo+Pj4+ IGh0dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vYW5pbWENCj4+DQoNCg== From nobody Wed Mar 9 19:54:11 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 043EC12DC12 for ; Wed, 9 Mar 2016 19:54:10 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.222 X-Spam-Level: X-Spam-Status: No, score=-4.222 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EjU-qXi2iZDv for ; Wed, 9 Mar 2016 19:54:07 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 09B7D12D7B1 for ; Wed, 9 Mar 2016 19:54:06 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml704-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CFR19031; Thu, 10 Mar 2016 03:54:04 +0000 (GMT) Received: from NKGEML404-HUB.china.huawei.com (10.98.56.35) by lhreml704-cah.china.huawei.com (10.201.5.130) with Microsoft SMTP Server (TLS) id 14.3.235.1; Thu, 10 Mar 2016 03:54:04 +0000 Received: from NKGEML515-MBS.china.huawei.com ([169.254.5.102]) by nkgeml404-hub.china.huawei.com ([10.98.56.35]) with mapi id 14.03.0235.001; Thu, 10 Mar 2016 11:54:01 +0800 From: Sheng Jiang To: "anima@ietf.org" Thread-Topic: Primary ANIMA Agenda @ IETF 95, Buenos Aires Thread-Index: AdF6gHCGeU30KzaVTl+hPDQMRr+pqQ== Date: Thu, 10 Mar 2016 03:54:00 +0000 Message-ID: <5D36713D8A4E7348A7E10DF7437A4B927C63A135@NKGEML515-MBS.china.huawei.com> Accept-Language: en-GB, zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.99.197] Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020205.56E0EFDD.0074, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=169.254.5.102, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: 3ba27e1f6bd58d419112d1c8fe170e78 Archived-At: Cc: "anima-chairs@tools.ietf.org" Subject: [Anima] Primary ANIMA Agenda @ IETF 95, Buenos Aires X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 10 Mar 2016 03:54:10 -0000 SGksIGFsbCwNCg0KVGhlIGNoYWlycyBoYXZlIHVwbG9hZGVkIGEgcHJpbWFyeSBtZWV0aW5nIGFn ZW5kYSBmb3Igb3VyIEFOSU1BIFdHIGluIFByYWd1ZS4NCg0KaHR0cHM6Ly93d3cuaWV0Zi5vcmcv cHJvY2VlZGluZ3MvOTUvYWdlbmRhL2FnZW5kYS05NS1hbmltYQ0KDQpJZiB0aGVyZSBpcyBhbnkg cXVlc3Rpb24gb3IgcmVxdWVzdCBvbiB0aGUgYWdlbmRhLCBwbGVhc2UgbGV0IHRoZSBjaGFpcnMg a25vdy4NCg0KVGhlIHNwZWFrZXJzIHBsZWFzZSBzZW5kIHlvdXIgc2xpZGVzIHRvIHRoZSBjaGFp ciBpbiBhZHZhbmNlLCBzbyB0aGF0IHdlIGNvdWxkIHVwbG9hZCB0aGVtIGJlZm9yZSB0aGUgbWVl dGluZy4gVGhlIHNsaWRlcyBhcmUgZHVlIGJlZm9yZSA0cG0gb24gSUVURiByZWNlcHRpb24gb24g U3VuZGF5LCBBcHJpbCAzcmQuDQoNCkJlc3QgcmVnYXJkcywNCg0KU2hlbmcgKyBUb2VybGVzcw0K From nobody Thu Mar 10 22:36:14 2016 Return-Path: X-Original-To: anima@ietf.org Delivered-To: anima@ietfa.amsl.com Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id A5B8112D51F; Thu, 10 Mar 2016 22:36:11 -0800 (PST) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit From: internet-drafts@ietf.org To: X-Test-IDTracker: no X-IETF-IDTracker: 6.16.0 Auto-Submitted: auto-generated Precedence: bulk Message-ID: <20160311063611.5239.27399.idtracker@ietfa.amsl.com> Date: Thu, 10 Mar 2016 22:36:11 -0800 Archived-At: Cc: anima@ietf.org Subject: [Anima] I-D Action: draft-ietf-anima-grasp-04.txt X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Mar 2016 06:36:12 -0000 A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the Autonomic Networking Integrated Model and Approach of the IETF. Title : A Generic Autonomic Signaling Protocol (GRASP) Authors : Carsten Bormann Brian Carpenter Bing Liu Filename : draft-ietf-anima-grasp-04.txt Pages : 58 Date : 2016-03-10 Abstract: This document establishes requirements for a signaling protocol that enables autonomic devices and autonomic service agents to dynamically discover peers, to synchronize state with them, and to negotiate parameter settings mutually with them. The document then defines a general protocol for discovery, synchronization and negotiation, while the technical objectives for specific scenarios are to be described in separate documents. An Appendix briefly discusses existing protocols with comparable features. The IETF datatracker status page for this draft is: https://datatracker.ietf.org/doc/draft-ietf-anima-grasp/ There's also a htmlized version available at: https://tools.ietf.org/html/draft-ietf-anima-grasp-04 A diff from the previous version is available at: https://www.ietf.org/rfcdiff?url2=draft-ietf-anima-grasp-04 Please note that it may take a couple of minutes from the time of submission until the htmlized version and diff are available at tools.ietf.org. Internet-Drafts are also available by anonymous FTP at: ftp://ftp.ietf.org/internet-drafts/ From nobody Thu Mar 10 22:43:03 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 385A312D525 for ; Thu, 10 Mar 2016 22:43:02 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.7 X-Spam-Level: X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id lcBvbytCOJPH for ; Thu, 10 Mar 2016 22:43:00 -0800 (PST) Received: from mail-pa0-x235.google.com (mail-pa0-x235.google.com [IPv6:2607:f8b0:400e:c03::235]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4B3FC12D523 for ; Thu, 10 Mar 2016 22:43:00 -0800 (PST) Received: by mail-pa0-x235.google.com with SMTP id td3so60352495pab.2 for ; Thu, 10 Mar 2016 22:43:00 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=Pjp3liqyAkqkj79ZYjLU0jAYKssL2ANCqBBztB9UEH0=; b=a6TmY231M8spzPZxC41SnxDja9ZcCWM3HaC8qER8hDgfJEj7+KgZrHf+laqouPN8bz roNNlkEmHK2Ix85t3K0oUyAvQhGz37ZOyFTyc3GMURRUkPodTsOrEj7spciHWpAezCfu 6PIvo5KHNZyt1FgZ6vJMo3uawRVQlVxc535D5/1kVal0aBpJl3aKTXj99YAUzjLDKzE0 kg8tQxzw0q8jIU283nDys8cVhrii7uHAXKT60uTKpea5z1wX7lWweom7o2aPcH4M9CTU qAK3oyW4DP3xEGQh2eYdkPu9ziGZHEDQFBnuaunXRxaj3avLSvC0r75wS/8R8+1GiElX EJTw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=Pjp3liqyAkqkj79ZYjLU0jAYKssL2ANCqBBztB9UEH0=; b=mJt6egp4QwMvF7P78gOp9N8duslf+f8Ce4Lh+qn+UNdgv9eXR+V9f5t6PkQ/mYjVRK zR2MpTpgr9Bk4XaNgI41XbBsyBEdmODMe7wdTbZkmtCrtfNMGdyz3xAkQCRm5byo+7CX RN5RcoWtitXnz/HKF0chtR8W/fgrifLADhPUXrWfnCwn3/skzYtO3cuNGuTnw3TCICnq y9lmPRw+NfKadoS0Ma12bBGNBzGhxP2MfMUtl6LBRiDqUw6GK2v5ddyuZGoM7ZqNR1t1 8lNqeIamyOF4t1Uk8AsVnaybbKOY1J6iQ5WUT4wvVqIY5HaWqR3sKHjU9s7MJrFViuLl sMQw== X-Gm-Message-State: AD7BkJKOdzdzA8TVk1GwulVC00frIT3GLd4J2thW05sGV5tnGtLZGrVTJ8CKrNGGkpOWCw== X-Received: by 10.66.221.167 with SMTP id qf7mr11723810pac.94.1457678579678; Thu, 10 Mar 2016 22:42:59 -0800 (PST) Received: from ?IPv6:2406:e007:64b4:1:28cc:dc4c:9703:6781? ([2406:e007:64b4:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id v7sm9974372pfi.56.2016.03.10.22.42.57 for (version=TLSv1/SSLv3 cipher=OTHER); Thu, 10 Mar 2016 22:42:58 -0800 (PST) To: Anima WG References: <20160311063611.5239.27399.idtracker@ietfa.amsl.com> From: Brian E Carpenter Organization: University of Auckland Message-ID: <56E268F9.8010009@gmail.com> Date: Fri, 11 Mar 2016 19:43:05 +1300 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0 MIME-Version: 1.0 In-Reply-To: <20160311063611.5239.27399.idtracker@ietfa.amsl.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Archived-At: Subject: Re: [Anima] I-D Action: draft-ietf-anima-grasp-04.txt X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Mar 2016 06:43:02 -0000 Hi all, The changes in this version (probably the last one before the IETF): Protocol change: Restored initiator field in certain messages and adjusted relaying rules to provide complete loop detection. Joel Halpern spotted a potential amplification loop in the previous version, so we have gone two steps back (to a change introduced in the -02 version but removed in the -03 version) and one step forward (by making the loop detection stronger). The Python prototype code has been updated and tested too. The logic of the relaying mechanism is now quite complex and we'll try to explain it in a couple of slides at the IETF. Regards Brian On 11/03/2016 19:36, internet-drafts@ietf.org wrote: > > A New Internet-Draft is available from the on-line Internet-Drafts directories. > This draft is a work item of the Autonomic Networking Integrated Model and Approach of the IETF. > > Title : A Generic Autonomic Signaling Protocol (GRASP) > Authors : Carsten Bormann > Brian Carpenter > Bing Liu > Filename : draft-ietf-anima-grasp-04.txt > Pages : 58 > Date : 2016-03-10 > > Abstract: > This document establishes requirements for a signaling protocol that > enables autonomic devices and autonomic service agents to dynamically > discover peers, to synchronize state with them, and to negotiate > parameter settings mutually with them. The document then defines a > general protocol for discovery, synchronization and negotiation, > while the technical objectives for specific scenarios are to be > described in separate documents. An Appendix briefly discusses > existing protocols with comparable features. > > > The IETF datatracker status page for this draft is: > https://datatracker.ietf.org/doc/draft-ietf-anima-grasp/ > > There's also a htmlized version available at: > https://tools.ietf.org/html/draft-ietf-anima-grasp-04 > > A diff from the previous version is available at: > https://www.ietf.org/rfcdiff?url2=draft-ietf-anima-grasp-04 > > > Please note that it may take a couple of minutes from the time of submission > until the htmlized version and diff are available at tools.ietf.org. > > Internet-Drafts are also available by anonymous FTP at: > ftp://ftp.ietf.org/internet-drafts/ > > _______________________________________________ > I-D-Announce mailing list > I-D-Announce@ietf.org > https://www.ietf.org/mailman/listinfo/i-d-announce > Internet-Draft directories: http://www.ietf.org/shadow.html > or ftp://ftp.ietf.org/ietf/1shadow-sites.txt > From nobody Fri Mar 11 15:07:19 2016 Return-Path: X-Original-To: anima@ietf.org Delivered-To: anima@ietfa.amsl.com Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 7C40112DDE4; Fri, 11 Mar 2016 15:05:39 -0800 (PST) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit From: "\"IETF Secretariat\"" To: , X-Test-IDTracker: no X-IETF-IDTracker: 6.16.0 Auto-Submitted: auto-generated Precedence: bulk Message-ID: <20160311230539.15028.45869.idtracker@ietfa.amsl.com> Date: Fri, 11 Mar 2016 15:05:39 -0800 Archived-At: Cc: anima@ietf.org, terry.manderson@icann.org Subject: [Anima] anima - Requested sessions have been scheduled for IETF 95 X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Mar 2016 23:05:39 -0000 Dear Sheng Jiang, The session(s) that you have requested have been scheduled. Below is the scheduled session information followed by the original request. anima Session 1 (2:00:00) Monday, Afternoon Session III 1740-1940 Room Name: Buen Ayre B size: 125 --------------------------------------------- anima Session 2 (1:30:00) Thursday, Afternoon Session II 1620-1720 Room Name: Atlantico C size: 225 --------------------------------------------- Request Information: --------------------------------------------------------- Working Group Name: Autonomic Networking Integrated Model and Approach Area Name: Operations and Management Area Session Requester: Sheng Jiang Number of Sessions: 2 Length of Session(s): 2 Hours, 1.5 Hours Number of Attendees: 80 Conflicts to Avoid: First Priority: homenet netconf dhc v6ops 6tisch nmlrg nmrg Second Priority: opsarea 6man softwire roll 6lo supa Third Priority: cfrg saag intarea Special Requests: Please arrange the first session on Monday/Tuesday and the second session on Thursday. So the WG participators could work and make progress between the two sessions. --------------------------------------------------------- From nobody Tue Mar 15 18:38:03 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1906212D7AB for ; Tue, 15 Mar 2016 18:38:02 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.221 X-Spam-Level: X-Spam-Status: No, score=-4.221 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YSH50c-6gLYL for ; Tue, 15 Mar 2016 18:37:59 -0700 (PDT) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A8AC312D523 for ; Tue, 15 Mar 2016 18:37:56 -0700 (PDT) Received: from 172.18.7.190 (EHLO lhreml706-cah.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CKT02215; Wed, 16 Mar 2016 01:37:54 +0000 (GMT) Received: from NKGEML408-HUB.china.huawei.com (10.98.56.39) by lhreml706-cah.china.huawei.com (10.201.5.182) with Microsoft SMTP Server (TLS) id 14.3.235.1; Wed, 16 Mar 2016 01:37:54 +0000 Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml408-hub.china.huawei.com ([10.98.56.39]) with mapi id 14.03.0235.001; Wed, 16 Mar 2016 09:37:41 +0800 From: Duzongpeng To: "anima@ietf.org" Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPg== Date: Wed, 16 Mar 2016 01:37:40 +0000 Message-ID: Accept-Language: zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.149.226] Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFA7E3nkgeml514mbxchi_" MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020203.56E8B8F3.0038, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: f33a4e5ca488bcf24dbcc6594a202836 Archived-At: Cc: Laurent Ciavaglia , =?iso-8859-1?Q?J=E9ferson_Campos_Nobre?= , Sheng Jiang Subject: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Mar 2016 01:38:02 -0000 --_000_BAFEC9523F57BC48A51C20226A5589575FDFA7E3nkgeml514mbxchi_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hello, everyone I understand that Autonomic network should be able to work with as= less configuration as possible. In the ideal case, an Autonomic network sh= ould be able to work well while only intent is needed from human operators. However, I wonder whether Autonomic network may need some interven= tion from human operators through some low-level policies beyond intents. For example, in the https://tools.ietf.org/html/draft-ietf-anima-p= refix-management-00, it is suggested that the prefix lengths for the CSG, A= SG, RSG (different roles in IP RAN) can be assigned as an "intent". These c= onfiguration parameters need to be distributed in the autonomic domain to i= nfluence the detail configurations on each autonomic node. As intent is described as an abstract, declarative, high-level pol= icy used to operate an autonomic domain, such as an enterprise network. Per= haps we can rename the "intent" in the "prefix-management ID" to some other= words. Is "ASA parameters" ok here? I think some characteristics for this kind of low-level policies are 1. Network level parameters, need to be distributed by GRASP and ACP 2. Relatively static compared to intent, perhaps only needed to be con= figured once 3. Unavoidable in current network environment, mostly for establishing= the network infrastructure 4. Rarely need coordinations with others parameters, usually is closed= related to one specific autonomic function Welcome for comment. Your suggestions will be appreciated. Best Regards Zongpeng Du --_000_BAFEC9523F57BC48A51C20226A5589575FDFA7E3nkgeml514mbxchi_ Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Hello, everyone

 

     &= nbsp;   I understand that Autonomic network should be able to wor= k with as less configuration as possible. In the ideal case, an Autonomic n= etwork should be able to work well while only intent is needed from human o= perators.

 

     &= nbsp;   However, I wonder whether Autonomic network may need some= intervention from human operators through some low-level policies beyond i= ntents.

 

     &= nbsp;   For example, in the https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it i= s suggested that the prefix lengths for the CSG, ASG, RSG (different roles = in IP RAN) can be assigned as an "intent". These configuration pa= rameters need to be distributed in the autonomic domain to influence the detail configurations on each autonomic node.=

 

     &= nbsp;   As intent is described as an abstract, declarative, high-= level policy used to operate an autonomic domain, such as an enterprise net= work. Perhaps we can rename the “intent” in the “prefix-m= anagement ID” to some other words.

 

Is= “ASA parameters" ok here?

 

I = think some characteristics for this kind of low-level policies are

1= .   &= nbsp;  Network level parameter= s, need to be distributed by GRASP and ACP

2= .   &= nbsp;  Relatively static compa= red to intent, perhaps only needed to be configured once<= /p>

3= .   &= nbsp;  Unavoidable in current = network environment, mostly for establishing the network infrastructure

4= .   &= nbsp;  Rarely need coordinatio= ns with others parameters, usually is closed related to one specific autono= mic function

     &= nbsp;  

We= lcome for comment. Your suggestions will be appreciated.<= /p>

 

 

Best Regards<= /p>

Zongpeng Du

--_000_BAFEC9523F57BC48A51C20226A5589575FDFA7E3nkgeml514mbxchi_-- From nobody Wed Mar 16 01:17:07 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B55D812D795 for ; Wed, 16 Mar 2016 01:17:04 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.521 X-Spam-Level: X-Spam-Status: No, score=-14.521 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bU5bVpfivLUU for ; Wed, 16 Mar 2016 01:17:01 -0700 (PDT) Received: from alln-iport-7.cisco.com (alln-iport-7.cisco.com [173.37.142.94]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C685C12D5DF for ; Wed, 16 Mar 2016 01:17:00 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=24198; q=dns/txt; s=iport; t=1458116220; x=1459325820; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=3F/Ra/tvK/D/BXn+Glouf8u+pP8p8921dsp4yEQqAeo=; b=Gwslx8Aib3L7K8m0RlGX03rVry5RlmfP+iUNbX9hDgccNlITMxhxhbaG glDO1sRyxP0y5D8lkp/1xiBqrs2idKsBzzwIvXYVLlX5Aoi0FWCFRgGdh f5MKhMVaH1OTkJgOZp9vlkxcL8NswF1hybqRlJPxFmywGHdmOjIQilMSZ U=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0BeAgA4FulW/4gNJK1egnpMU24GugIBD?= =?us-ascii?q?YFuIYVsAoE5OBQBAQEBAQEBZCeEQQEBAQQtTBACAQgRBAEBIQcHMhQJCAEBBAE?= =?us-ascii?q?NBQiIHw6/BAEBAQEBAQEBAQEBAQEBAQEBAQEBAREEhh6ERIRpCYQEBY1yiV0Bh?= =?us-ascii?q?W6IC48Mjn4BHgEBQoNlaokqJRZ+AQEB?= X-IronPort-AV: E=Sophos;i="5.24,343,1454976000"; d="scan'208,217";a="249864147" Received: from alln-core-3.cisco.com ([173.36.13.136]) by alln-iport-7.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 16 Mar 2016 08:16:58 +0000 Received: from XCH-ALN-007.cisco.com (xch-aln-007.cisco.com [173.36.7.17]) by alln-core-3.cisco.com (8.14.5/8.14.5) with ESMTP id u2G8Gwkr003950 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Wed, 16 Mar 2016 08:16:59 GMT Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-007.cisco.com (173.36.7.17) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Wed, 16 Mar 2016 03:16:58 -0500 Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Wed, 16 Mar 2016 03:16:58 -0500 From: "Michael Behringer (mbehring)" To: Duzongpeng , "anima@ietf.org" Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rw Date: Wed, 16 Mar 2016 08:16:58 +0000 Message-ID: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> References: In-Reply-To: Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.55.238.139] Content-Type: multipart/alternative; boundary="_000_701d351267344c21b3ea053aa5b15728XCHRCD006ciscocom_" MIME-Version: 1.0 Archived-At: Cc: Laurent Ciavaglia , =?iso-8859-1?Q?J=E9ferson_Campos_Nobre?= , Sheng Jiang Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Mar 2016 08:17:04 -0000 --_000_701d351267344c21b3ea053aa5b15728XCHRCD006ciscocom_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Zongpeng, Yes, this is a recurring question and topic. Let me give my view on two lev= els: First, ANIMA isn't the only thing "influencing" or "steering" a network, at= least not initially. There is still traditional config, plus all the exist= ing and emerging SDN models. My view has always been that ANIMA sets a "def= ault" policy in a network, and all other means are more specific and can ov= erride intent. This is described in RFC7575. Main point here: ANIMA Intent = isn't the only tool in a network. Intent could perfectly well cover a high level policy such as "all nodes of= type x do this; type y does that". (Which I think is what you mention belo= w.) When we started with Intent, the vision was to keep it very high level,= more along the line "do the right thing" :) It has become clear that in t= oday's networking, people will want to push more "parameters" than "policy"= , and I think we'll just have to acknowledge that. Personally I think it'll pan out like this: Intent will have sections per a= utonomic function. An Intent parser will validate overall correctness (sign= ature, etc), then segment the Intent into its bits: One part will be the au= tonomic fabric itself, then there will be bits for each autonomic function. So the Intent parser on a node will grab intent, see there is some Intent f= or autonomic function A, will see whether this is running locally, and if s= o, it'll just pass that piece of Intent to that function. In this way, a function owner defines both the intent itself, and how it is= consumed by his/her function. The downside is that there is no "control" and people can do what they want= , in the worst case even push config, which is clearly NOT what Intent is m= eant to do. I'd hope that people don't start writing Intent like "node 17: = here is your CLI; node 23: here is your CLI"; node 37: here is your CLI". B= ut what I describe above COULD be abused in this way. RFC7575 says this is = NOT Intent. In other words: Let's not be too prescriptive, and hope that folks will do = more or less reasonable things. I've been meaning to comment on the Intent drafts and possibly contribute, = I'm just struggling with priorities at the moment (as you can see by my rel= ative silence over the past weeks). Michael From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng Sent: 16 March 2016 02:38 To: anima@ietf.org Cc: Laurent Ciavaglia ; J=E9ferson Campos Nobr= e ; Sheng Jiang Subject: [Anima] Does Autonomic network need some intervention beyond inten= t Hello, everyone I understand that Autonomic network should be able to work with as= less configuration as possible. In the ideal case, an Autonomic network sh= ould be able to work well while only intent is needed from human operators. However, I wonder whether Autonomic network may need some interven= tion from human operators through some low-level policies beyond intents. For example, in the https://tools.ietf.org/html/draft-ietf-anima-p= refix-management-00, it is suggested that the prefix lengths for the CSG, A= SG, RSG (different roles in IP RAN) can be assigned as an "intent". These c= onfiguration parameters need to be distributed in the autonomic domain to i= nfluence the detail configurations on each autonomic node. As intent is described as an abstract, declarative, high-level pol= icy used to operate an autonomic domain, such as an enterprise network. Per= haps we can rename the "intent" in the "prefix-management ID" to some other= words. Is "ASA parameters" ok here? I think some characteristics for this kind of low-level policies are 1. Network level parameters, need to be distributed by GRASP and ACP 2. Relatively static compared to intent, perhaps only needed to be co= nfigured once 3. Unavoidable in current network environment, mostly for establishin= g the network infrastructure 4. Rarely need coordinations with others parameters, usually is close= d related to one specific autonomic function Welcome for comment. Your suggestions will be appreciated. Best Regards Zongpeng Du --_000_701d351267344c21b3ea053aa5b15728XCHRCD006ciscocom_ Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

= H= i Zongpeng,

= <= o:p> 

= Y= es, this is a recurring question and topic. Let me give my view on two leve= ls:

= <= o:p> 

= F= irst, ANIMA isn’t the only thing “influencing” or “= steering” a network, at least not initially. There is still tradition= al config, plus all the existing and emerging SDN models. My view has always = been that ANIMA sets a “default” policy in a network, and all o= ther means are more specific and can override intent. This is described in = RFC7575. Main point here: ANIMA Intent isn’t the only tool in a network.

= <= o:p> 

= I= ntent could perfectly well cover a high level policy such as “all nod= es of type x do this; type y does that”. (Which I think is what you mention below.) When we started with Intent, the visio= n was to keep it very high level, more along the line “do the right t= hing” J  It has become clear that in today’s networking, people will want to = push more “parameters” than “policy”, and I think w= e’ll just have to acknowledge that.

= <= o:p> 

= P= ersonally I think it’ll pan out like this: Intent will have sections = per autonomic function. An Intent parser will validate overall correctness (signature, etc), then segment the Intent into its bit= s: One part will be the autonomic fabric itself, then there will be bits fo= r each autonomic function.

= <= o:p> 

= S= o the Intent parser on a node will grab intent, see there is some Intent fo= r autonomic function A, will see whether this is running locally, and if so, it’ll just pass that piece of Intent = to that function.

= <= o:p> 

= I= n this way, a function owner defines both the intent itself, and how it is = consumed by his/her function.

= <= o:p> 

= T= he downside is that there is no “control” and people can do wha= t they want, in the worst case even push config, which is clearly NOT what Intent is meant to do. I’d hope that people don&= #8217;t start writing Intent like “node 17: here is your CLI; node 23= : here is your CLI”; node 37: here is your CLI”. But what I des= cribe above COULD be abused in this way. RFC7575 says this is NOT Intent.

= <= o:p> 

= I= n other words: Let’s not be too prescriptive, and hope that folks wil= l do more or less reasonable things.

= <= o:p> 

= I= ’ve been meaning to comment on the Intent drafts and possibly contrib= ute, I’m just struggling with priorities at the moment (as you can see by my relative silence over the past weeks). <= /o:p>

= <= o:p> 

= M= ichael

= <= o:p> 

= <= o:p> 

= <= o:p> 

From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng
Sent: 16 March 2016 02:38 To: anima@ietf.org
Cc: Laurent Ciavaglia <la= urent.ciavaglia@nokia.com>; J=E9ferson Campos Nobre <jcnobre@inf.ufrg= s.br>; Sheng Jiang <jiangsheng@huawei.com>
Subject: [Anima] Does Autono= mic network need some intervention beyond intent

 <= /span>

Hello, everyone

 <= /span>

   =       I understand that Autonomic network should b= e able to work with as less configuration as possible. In the ideal case, a= n Autonomic network should be able to work well while only intent is needed from human= operators.

 <= /span>

   =       However, I wonder whether Autonomic network = may need some intervention from human operators through some low-level poli= cies beyond intents.

 <= /span>

   =       For example, in the https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, i= t is suggested that the prefix lengths for the CSG, ASG, RSG (different rol= es in IP RAN) can be assigned as an "intent". These configuration parameters need to be distributed = in the autonomic domain to influence the detail configurations on each auto= nomic node.

 <= /span>

   =       As intent is described as an abstract, decla= rative, high-level policy used to operate an autonomic domain, such as an e= nterprise network. Perhaps we can rename the “intent” in the “pref= ix-management ID” to some other words.

 <= /span>

Is “ASA parameters" ok here?

 

I think some characteristics for this kind of low-level polici= es are

1.       Network level parameters, need to be distribute= d by GRASP and ACP

2.       Relatively static compared to intent, perhaps o= nly needed to be configured once

3.       Unavoidable in current network environment, mos= tly for establishing the network infrastructure

4.       Rarely need coordinations with others parameter= s, usually is closed related to one specific autonomic function<= /span>

   =      

Welcome for comment. Your suggestions will be appreciated.

 

 

Best Regards<= /o:p>

Zongpeng Du

--_000_701d351267344c21b3ea053aa5b15728XCHRCD006ciscocom_-- From nobody Wed Mar 16 05:41:51 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2631B12D93A for ; Wed, 16 Mar 2016 05:41:49 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.221 X-Spam-Level: X-Spam-Status: No, score=-4.221 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gHT2HZchvlCS for ; Wed, 16 Mar 2016 05:41:45 -0700 (PDT) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CEB1A12D766 for ; Wed, 16 Mar 2016 05:41:43 -0700 (PDT) Received: from 172.18.7.190 (EHLO lhreml708-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGF42363; Wed, 16 Mar 2016 12:41:41 +0000 (GMT) Received: from NKGEML414-HUB.china.huawei.com (10.98.56.75) by lhreml708-cah.china.huawei.com (10.201.5.202) with Microsoft SMTP Server (TLS) id 14.3.235.1; Wed, 16 Mar 2016 12:41:39 +0000 Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml414-hub.china.huawei.com ([10.98.56.75]) with mapi id 14.03.0235.001; Wed, 16 Mar 2016 20:41:31 +0800 From: Duzongpeng To: "Michael Behringer (mbehring)" , "anima@ietf.org" Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAAE3aEA= Date: Wed, 16 Mar 2016 12:41:30 +0000 Message-ID: References: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> In-Reply-To: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> Accept-Language: zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.149.226] Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFA9B8nkgeml514mbxchi_" MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A090203.56E95486.0005, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: 75564e734a57cfeb0e99f48b48eaa534 Archived-At: Cc: Laurent Ciavaglia , =?iso-8859-1?Q?J=E9ferson_Campos_Nobre?= , Sheng Jiang Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Mar 2016 12:41:49 -0000 --_000_BAFEC9523F57BC48A51C20226A5589575FDFA9B8nkgeml514mbxchi_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi, Michael, Thanks for the reply. I can understand that intent is not in the c= urrent milestones, and you are working on several milestones drafts now. You are welcome to contribute to the intent draft at any time, and= thanks in advance. According to your statements, in my understanding, we can keep reg= arding the "intent" example in "prefix-management ID" as an Autonomic inten= t. It's semantics are just like "all nodes of type x do this; type y = does that", so it is a high level policy. It really helps a lot. Thanks. I generally agree with your statement, while some other personal u= nderstandings exist (shown below). But perhaps they can be discussed later. Best Regards Zongpeng Du From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com] Sent: Wednesday, March 16, 2016 4:17 PM To: Duzongpeng; anima@ietf.org Cc: Laurent Ciavaglia; J=E9ferson Campos Nobre; Sheng Jiang Subject: RE: [Anima] Does Autonomic network need some intervention beyond i= ntent Hi Zongpeng, Yes, this is a recurring question and topic. Let me give my view on two lev= els: First, ANIMA isn't the only thing "influencing" or "steering" a network, at= least not initially. There is still traditional config, plus all the exist= ing and emerging SDN models. My view has always been that ANIMA sets a "def= ault" policy in a network, and all other means are more specific and can ov= erride intent. This is described in RFC7575. Main point here: ANIMA Intent = isn't the only tool in a network. [duzongpeng] I agree that intent can coexist with traditional config. I generally agree with the policy priorities, but I have some other concern= s. Some default policies/parameters set through "intent" perhaps are fundament= al for the uplayer services. It should be prevented that human operator uses some CLIs to change them by= mistake, and the uplayer services are interrupted. At least, some warning should be reported to the human operator. [duzongpen= g] Intent could perfectly well cover a high level policy such as "all nodes of= type x do this; type y does that". (Which I think is what you mention belo= w.) When we started with Intent, the vision was to keep it very high level,= more along the line "do the right thing" :) It has become clear that in t= oday's networking, people will want to push more "parameters" than "policy"= , and I think we'll just have to acknowledge that. Personally I think it'll pan out like this: Intent will have sections per a= utonomic function. An Intent parser will validate overall correctness (sign= ature, etc), then segment the Intent into its bits: One part will be the au= tonomic fabric itself, then there will be bits for each autonomic function. So the Intent parser on a node will grab intent, see there is some Intent f= or autonomic function A, will see whether this is running locally, and if s= o, it'll just pass that piece of Intent to that function. In this way, a function owner defines both the intent itself, and how it is= consumed by his/her function. [duzongpeng]Agree.[duzongpeng] The downside is that there is no "control" an//d people can do what they wa= nt, in the worst case even push config, which is clearly NOT what Intent is= meant to do. I'd hope that people don't start writing Intent like "node 17= : here is your CLI; node 23: here is your CLI"; node 37: here is your CLI".= But what I describe above COULD be abused in this way. RFC7575 says this i= s NOT Intent. [duzongpeng]Agree. It is hard to give an exact boundary. However, I think we have two choices here. The first one is to define a policy first in every ASA, but with some open = parameters that can be configured. What we need to do is just to distribute= the network-level parameters for that function. This way is suitable to so= me relatively static functions. The second is more advanced, in the ideal case, the human operator can comm= unicate with network using natural language or something similar. This way = is more flexible, but harder to realize. [duzongpeng] In other words: Let's not be too prescriptive, and hope that folks will do = more or less reasonable things. [duzongpeng]Agree.[duzongpeng] I've been meaning to comment on the Intent drafts and possibly contribute, = I'm just struggling with priorities at the moment (as you can see by my rel= ative silence over the past weeks). Michael From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng Sent: 16 March 2016 02:38 To: anima@ietf.org Cc: Laurent Ciavaglia >; J=E9ferson Campos Nobre >; Sheng Jiang > Subject: [Anima] Does Autonomic network need some intervention beyond inten= t Hello, everyone I understand that Autonomic network should be able to work with as= less configuration as possible. In the ideal case, an Autonomic network sh= ould be able to work well while only intent is needed from human operators. However, I wonder whether Autonomic network may need some interven= tion from human operators through some low-level policies beyond intents. For example, in the https://tools.ietf.org/html/draft-ietf-anima-p= refix-management-00, it is suggested that the prefix lengths for the CSG, A= SG, RSG (different roles in IP RAN) can be assigned as an "intent". These c= onfiguration parameters need to be distributed in the autonomic domain to i= nfluence the detail configurations on each autonomic node. As intent is described as an abstract, declarative, high-level pol= icy used to operate an autonomic domain, such as an enterprise network. Per= haps we can rename the "intent" in the "prefix-management ID" to some other= words. Is "ASA parameters" ok here? I think some characteristics for this kind of low-level policies are 1. Network level parameters, need to be distributed by GRASP and ACP 2. Relatively static compared to intent, perhaps only needed to be co= nfigured once 3. Unavoidable in current network environment, mostly for establishin= g the network infrastructure 4. Rarely need coordinations with others parameters, usually is close= d related to one specific autonomic function Welcome for comment. Your suggestions will be appreciated. Best Regards Zongpeng Du --_000_BAFEC9523F57BC48A51C20226A5589575FDFA9B8nkgeml514mbxchi_ Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Hi, Mic= hael,

&n= bsp;

 &= nbsp;       Thanks for the reply. I can under= stand that intent is not in the current milestones, and you are working on = several milestones drafts now.

&n= bsp;

 &= nbsp;       You are welcome to contribute to = the intent draft at any time, and thanks in advance.

&n= bsp;

 &= nbsp;       According to your statements, in = my understanding, we can keep regarding the “intent” example in= “prefix-management ID” as an Autonomic intent.

 &= nbsp;       It’s semantics are just lik= e “= ;all nodes of type x do this; type y does that”, so it is a high leve= l policy.

&n= bsp;

 &= nbsp;       It really helps a lot. Thanks.

&n= bsp;

 &= nbsp;       I generally agree with your state= ment, while some other personal understandings exist (shown below). But per= haps they can be discussed later.

&n= bsp;

Best Re= gards           &nbs= p;            &= nbsp;           &nbs= p;            &= nbsp;           &nbs= p;     

Zongpen= g Du

&n= bsp;

From: Michael Behr= inger (mbehring) [mailto:mbehring@cisco.com]
Sent: Wednesday, March 16, 2016 4:17 PM
To: Duzongpeng; anima@ietf.org
Cc: Laurent Ciavaglia; J=E9ferson Campos Nobre; Sheng Jiang
Subject: RE: [Anima] Does Autonomic network need some intervention b= eyond intent

 

Hi Zongpeng,

 

Yes, this is a recurring question and topic. Let me give my view = on two levels:

           = ;            &n= bsp;            = ;            &n= bsp;            = ;            &n= bsp;            = ;            &n= bsp;            = ;            &n= bsp;           <= /o:p>

First, ANIMA isn’t the only thing “influencing”= or “steering” a network, at least not initially. There is stil= l traditional config, plus all the existing and emerging SDN models. My view has always been that ANIMA sets a “default” policy in = a network, and all other means are more specific and can override intent. T= his is described in RFC7575. Main point here: ANIMA Intent isn’t the = only tool in a network.

 

[duzongpeng] I agree that inten= t can coexist with traditional config.

I generally agree with the poli= cy priorities, but I have some other concerns.

Some default policies/parameters set through ̶= 0;intent” perhaps are fundamental for the uplayer services.

It should be prevented that hum= an operator uses some CLIs to change them by mistake, and the uplayer servi= ces are interrupted.

At least, some warning should b= e reported to the human operator. [duzongpeng]

&n= bsp;

Intent could perfectly well cover a high level policy such as = 220;all nodes of type x do this; type y does that”. (Which I think is= what you mention below.) When we started with Intent, the vision was to keep it very high level, more along the line “do t= he right thing” J  It has become clear that in today’s networking, peo= ple will want to push more “parameters” than “policy̶= 1;, and I think we’ll just have to acknowledge that. <= /p>

 

Personally I think it’ll pan out like this: Intent will hav= e sections per autonomic function. An Intent parser will validate overall c= orrectness (signature, etc), then segment the Intent into its bits: One part will be the autonomic fabric itself, then t= here will be bits for each autonomic function.

 

So the Intent parser on a node will grab intent, see there is som= e Intent for autonomic function A, will see whether this is running locally= , and if so, it’ll just pass that piece of Intent to that function.

 

In this way, a function owner defines both the intent itself, and= how it is consumed by his/her function.

&n= bsp;

[duzongpeng]Agree.[duzongpeng]<= o:p>

 

The downside is that there is no “control” an<= span lang=3D"EN-GB" style=3D"font-size:11.0pt;color:#1F497D">//d people can do what they want, in the worst case even push config, which is= clearly NOT what Intent is meant to do. I’d hope that people donR= 17;t start writing Intent like “node 17: here is your CLI; node 23: h= ere is your CLI”; node 37: here is your CLI”. But what I describe above COULD be abused in this way. RFC7575 says this is NO= T Intent.

 

[duzongpeng]Agree. It is hard t= o give an exact boundary.

However, I think we have two ch= oices here.

The first one is to define a po= licy first in every ASA, but with some open parameters that can be configur= ed. What we need to do is just to distribute the network-level parameters f= or that function. This way is suitable to some relatively static functions.

The second is more advanced, in= the ideal case, the human operator can communicate with network using natu= ral language or something similar. This way is more flexible, but harder to= realize.

[duzongpeng]<= /p>

&n= bsp;

In other words: Let’s not be too prescriptive, and hope tha= t folks will do more or less reasonable things.

 

[duzongpeng]Agree.[duzongpeng]<= o:p>

&n= bsp;

I’ve been meaning to comment on the Intent drafts and possi= bly contribute, I’m just struggling with priorities at the moment (as= you can see by my relative silence over the past weeks).

 

Michael

 

 

 

From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng
Sent: 16 March 2016 02:38
To: anima@ietf.org
Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; J=E9ferson Campos Nobre <jcnobre@inf.ufrgs.br>; Sheng Jian= g <jiangsheng@huawei.com>= ;
Subject: [Anima] Does Autonomic network need some intervention beyon= d intent

 

Hello, everyone

 

     &= nbsp;   I understand that Autonomic network should be able to wor= k with as less configuration as possible. In the ideal case, an Autonomic n= etwork should be able to work well while only intent is needed from human o= perators.

 

     &= nbsp;   However, I wonder whether Autonomic network may need some= intervention from human operators through some low-level policies beyond i= ntents.

 

     &= nbsp;   For example, in the https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it i= s suggested that the prefix lengths for the CSG, ASG, RSG (different roles = in IP RAN) can be assigned as an "intent". These configuration pa= rameters need to be distributed in the autonomic domain to influence the detail configurations on each autonomic node.=

 

     &= nbsp;   As intent is described as an abstract, declarative, high-= level policy used to operate an autonomic domain, such as an enterprise net= work. Perhaps we can rename the “intent” in the “prefix-m= anagement ID” to some other words.

 

Is= “ASA parameters" ok here?

 

I = think some characteristics for this kind of low-level policies are

1.  = ;     Network level parameters, need to be distribute= d by GRASP and ACP

2.  = ;     Relatively static compared to intent, perhaps o= nly needed to be configured once

3.  = ;     Unavoidable in current network environment, mos= tly for establishing the network infrastructure

4.  = ;     Rarely need coordinations with others parameter= s, usually is closed related to one specific autonomic function<= /span>

     &= nbsp;  

We= lcome for comment. Your suggestions will be appreciated.<= /p>

 

 

Best Regards<= /p>

Zongpeng Du

--_000_BAFEC9523F57BC48A51C20226A5589575FDFA9B8nkgeml514mbxchi_-- From nobody Wed Mar 16 11:40:58 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DFD7B12D5E7 for ; Wed, 16 Mar 2016 11:40:57 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.7 X-Spam-Level: X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QL78Whl3pCi8 for ; Wed, 16 Mar 2016 11:40:55 -0700 (PDT) Received: from mail-pf0-x232.google.com (mail-pf0-x232.google.com [IPv6:2607:f8b0:400e:c00::232]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C09F812DA9A for ; Wed, 16 Mar 2016 11:40:50 -0700 (PDT) Received: by mail-pf0-x232.google.com with SMTP id x3so85023637pfb.1 for ; Wed, 16 Mar 2016 11:40:50 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=vOEs3DFhgokBiE/3uo8NF2LZWshcFfWwNy59UgYXh8Y=; b=uM5+Neh9eSy9KTsAv6zGB9BzyOPWbAvbvAienPAvLpLOBFC6ou5XPel40hteUX7Dat K6weW3LBcNw8/rQZeIqUuwM6EfsGbNyC1j/UY3kDQB66AxExCM5pY6fk/Txy/jYffkVm todVa/AxKJRCZZUVs7UODf61oRPdnPMGIAi7gYZt5PsrTdsTti1X04xW/c/qOkrh9BzB ELXcLTODWLpSBhSV0YT+WZY/YSMeW1vVpvqM8iP8MiKT0uXYJzy62QybUTwuPR4qximE Ro5sPuUHzLjF9RXn9TpLr5GgdCv/ndk+On2HXnkmi8Q2Zd1VwoCW/O7ic9oLMP5hsNbe nQYg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=vOEs3DFhgokBiE/3uo8NF2LZWshcFfWwNy59UgYXh8Y=; b=Vg17hCDtyc0j4Y8ei+NuQEKUBhj0+aqo91du4zxdZVyI0jImONSszxCTMlh4ZKgrbv yYhv3+bKdtS2hiyY25Um4HjCxIwppcfVQpk4w6VhlBvqD9HIVy+v90HU2FM78FFWTZQn XKY+msFHdwxhK8+KsZWOYLoBnBA03QqjMSMAMwY6oU73Ql7gqQXoGf4fN1kYN605zv1K OP8Nl03b7LIpkrKOEa37j3twYLANY3tNqgdjYfsVMlyM+An7dFZlLQ97kya+JI2KbmE4 jNitgouuEO5jpeJDLtf+y63H4vDssKry4xxmJLrKyIpAk0LsTin3g2x4eqzZs/mllBSw LHGw== X-Gm-Message-State: AD7BkJKEY4nLrxJVBro2R/t0w5IH1ouF3IyZYA5nVWXN73mOuqocwZQ5+BO6c87+bbbDwg== X-Received: by 10.66.154.233 with SMTP id vr9mr8357531pab.66.1458153650286; Wed, 16 Mar 2016 11:40:50 -0700 (PDT) Received: from [10.1.9.199] ([103.23.18.15]) by smtp.gmail.com with ESMTPSA id lq10sm7354230pab.36.2016.03.16.11.40.45 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 16 Mar 2016 11:40:48 -0700 (PDT) To: Duzongpeng , "Michael Behringer (mbehring)" , "anima@ietf.org" References: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> From: Brian E Carpenter Organization: University of Auckland Message-ID: <56E9A8B2.8070504@gmail.com> Date: Thu, 17 Mar 2016 07:40:50 +1300 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Archived-At: Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , Sheng Jiang Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Mar 2016 18:40:58 -0000 Yes, I think that whatever the "philosophy" of intent is, we should make = sure that the implementation is pragmatic and that we can adapt to operator's requi= rements. Regards Brian On 17/03/2016 01:41, Duzongpeng wrote: > Hi, Michael, >=20 > Thanks for the reply. I can understand that intent is not in t= he current milestones, and you are working on several milestones drafts n= ow. >=20 > You are welcome to contribute to the intent draft at any time,= and thanks in advance. >=20 > According to your statements, in my understanding, we can keep= regarding the "intent" example in "prefix-management ID" as an Autonomic= intent. > It's semantics are just like "all nodes of type x do this; typ= e y does that", so it is a high level policy. >=20 > It really helps a lot. Thanks. >=20 > I generally agree with your statement, while some other person= al understandings exist (shown below). But perhaps they can be discussed = later. >=20 > Best Regards > Zongpeng Du >=20 > From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com] > Sent: Wednesday, March 16, 2016 4:17 PM > To: Duzongpeng; anima@ietf.org > Cc: Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang > Subject: RE: [Anima] Does Autonomic network need some intervention beyo= nd intent >=20 > Hi Zongpeng, >=20 > Yes, this is a recurring question and topic. Let me give my view on two= levels: >=20 > First, ANIMA isn't the only thing "influencing" or "steering" a network= , at least not initially. There is still traditional config, plus all the= existing and emerging SDN models. My view has always been that ANIMA set= s a "default" policy in a network, and all other means are more specific = and can override intent. This is described in RFC7575. Main point here: A= NIMA Intent isn't the only tool in a network. >=20 > [duzongpeng] I agree that intent can coexist with traditional config. > I generally agree with the policy priorities, but I have some other con= cerns. > Some default policies/parameters set through "intent" perhaps are funda= mental for the uplayer services. > It should be prevented that human operator uses some CLIs to change the= m by mistake, and the uplayer services are interrupted. > At least, some warning should be reported to the human operator. [duzon= gpeng] >=20 > Intent could perfectly well cover a high level policy such as "all node= s of type x do this; type y does that". (Which I think is what you mentio= n below.) When we started with Intent, the vision was to keep it very hig= h level, more along the line "do the right thing" :) It has become clear= that in today's networking, people will want to push more "parameters" t= han "policy", and I think we'll just have to acknowledge that. >=20 > Personally I think it'll pan out like this: Intent will have sections p= er autonomic function. An Intent parser will validate overall correctness= (signature, etc), then segment the Intent into its bits: One part will b= e the autonomic fabric itself, then there will be bits for each autonomic= function. >=20 > So the Intent parser on a node will grab intent, see there is some Inte= nt for autonomic function A, will see whether this is running locally, an= d if so, it'll just pass that piece of Intent to that function. >=20 > In this way, a function owner defines both the intent itself, and how i= t is consumed by his/her function. >=20 > [duzongpeng]Agree.[duzongpeng] >=20 > The downside is that there is no "control" an//d people can do what the= y want, in the worst case even push config, which is clearly NOT what Int= ent is meant to do. I'd hope that people don't start writing Intent like = "node 17: here is your CLI; node 23: here is your CLI"; node 37: here is = your CLI". But what I describe above COULD be abused in this way. RFC7575= says this is NOT Intent. >=20 > [duzongpeng]Agree. It is hard to give an exact boundary. > However, I think we have two choices here. > The first one is to define a policy first in every ASA, but with some o= pen parameters that can be configured. What we need to do is just to dist= ribute the network-level parameters for that function. This way is suitab= le to some relatively static functions. > The second is more advanced, in the ideal case, the human operator can = communicate with network using natural language or something similar. Thi= s way is more flexible, but harder to realize. > [duzongpeng] >=20 > In other words: Let's not be too prescriptive, and hope that folks will= do more or less reasonable things. >=20 > [duzongpeng]Agree.[duzongpeng] >=20 > I've been meaning to comment on the Intent drafts and possibly contribu= te, I'm just struggling with priorities at the moment (as you can see by = my relative silence over the past weeks). >=20 > Michael >=20 >=20 >=20 > From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng > Sent: 16 March 2016 02:38 > To: anima@ietf.org > Cc: Laurent Ciavaglia >; J=C3=A9ferson Campos Nobre >; Sheng Jiang > > Subject: [Anima] Does Autonomic network need some intervention beyond i= ntent >=20 > Hello, everyone >=20 > I understand that Autonomic network should be able to work wit= h as less configuration as possible. In the ideal case, an Autonomic netw= ork should be able to work well while only intent is needed from human op= erators. >=20 > However, I wonder whether Autonomic network may need some inte= rvention from human operators through some low-level policies beyond inte= nts. >=20 > For example, in the https://tools.ietf.org/html/draft-ietf-ani= ma-prefix-management-00, it is suggested that the prefix lengths for the = CSG, ASG, RSG (different roles in IP RAN) can be assigned as an "intent".= These configuration parameters need to be distributed in the autonomic d= omain to influence the detail configurations on each autonomic node. >=20 > As intent is described as an abstract, declarative, high-level= policy used to operate an autonomic domain, such as an enterprise networ= k. Perhaps we can rename the "intent" in the "prefix-management ID" to so= me other words. >=20 > Is "ASA parameters" ok here? >=20 > I think some characteristics for this kind of low-level policies are >=20 > 1. Network level parameters, need to be distributed by GRASP and = ACP >=20 > 2. Relatively static compared to intent, perhaps only needed to b= e configured once >=20 > 3. Unavoidable in current network environment, mostly for establi= shing the network infrastructure >=20 > 4. Rarely need coordinations with others parameters, usually is c= losed related to one specific autonomic function >=20 > Welcome for comment. Your suggestions will be appreciated. >=20 >=20 > Best Regards > Zongpeng Du >=20 >=20 >=20 > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima >=20 From nobody Thu Mar 17 09:05:13 2016 Return-Path: X-Original-To: anima@ietf.org Delivered-To: anima@ietfa.amsl.com Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 6145212D5B0; Thu, 17 Mar 2016 09:05:11 -0700 (PDT) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit From: internet-drafts@ietf.org To: X-Test-IDTracker: no X-IETF-IDTracker: 6.17.0 Auto-Submitted: auto-generated Precedence: bulk Message-ID: <20160317160511.13035.63108.idtracker@ietfa.amsl.com> Date: Thu, 17 Mar 2016 09:05:11 -0700 Archived-At: Cc: anima@ietf.org Subject: [Anima] I-D Action: draft-ietf-anima-bootstrapping-keyinfra-02.txt X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 17 Mar 2016 16:05:11 -0000 A New Internet-Draft is available from the on-line Internet-Drafts directories. This draft is a work item of the Autonomic Networking Integrated Model and Approach of the IETF. Title : Bootstrapping Key Infrastructures Authors : Max Pritikin Michael C. Richardson Michael H. Behringer Steinthor Bjarnason Filename : draft-ietf-anima-bootstrapping-keyinfra-02.txt Pages : 38 Date : 2016-03-17 Abstract: This document specifies automated bootstrapping of a key infrastructure (BSKI) using vendor installed IEEE 802.1AR manufacturing installed certificates, in combination with a vendor based service on the Internet. Before being authenticated, a new device has only link-local connectivity, and does not require a routable address. When a vendor provides an Internet based service, devices can be forced to join only specific domains but in limited/ disconnected networks or legacy environments we describe a variety of options that allow bootstrapping to proceed. The IETF datatracker status page for this draft is: https://datatracker.ietf.org/doc/draft-ietf-anima-bootstrapping-keyinfra/ There's also a htmlized version available at: https://tools.ietf.org/html/draft-ietf-anima-bootstrapping-keyinfra-02 A diff from the previous version is available at: https://www.ietf.org/rfcdiff?url2=draft-ietf-anima-bootstrapping-keyinfra-02 Please note that it may take a couple of minutes from the time of submission until the htmlized version and diff are available at tools.ietf.org. Internet-Drafts are also available by anonymous FTP at: ftp://ftp.ietf.org/internet-drafts/ From nobody Fri Mar 18 01:33:47 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D84F712D7B3 for ; Fri, 18 Mar 2016 01:33:45 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.222 X-Spam-Level: X-Spam-Status: No, score=-4.222 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id V_zYcx4R6ZWv for ; Fri, 18 Mar 2016 01:33:42 -0700 (PDT) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B434D12D6D6 for ; Fri, 18 Mar 2016 01:33:41 -0700 (PDT) Received: from 172.18.7.190 (EHLO lhreml701-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGH45796; Fri, 18 Mar 2016 08:33:38 +0000 (GMT) Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml701-cah.china.huawei.com (10.201.5.93) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 18 Mar 2016 08:33:36 +0000 Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0235.001; Fri, 18 Mar 2016 16:33:25 +0800 From: Duzongpeng To: Brian E Carpenter , "anima@ietf.org" Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAAE3aEAABGmDAABfNuwA Date: Fri, 18 Mar 2016 08:33:25 +0000 Message-ID: References: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> <56E9A8B2.8070504@gmail.com> In-Reply-To: <56E9A8B2.8070504@gmail.com> Accept-Language: zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.149.226] Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A090202.56EBBD64.001B, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: 75564e734a57cfeb0e99f48b48eaa534 Archived-At: Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "Michael Behringer \(mbehring\)" , Sheng Jiang Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 18 Mar 2016 08:33:46 -0000 SGksIEJyaWFuDQoNCglUaGFua3MgZm9yIHRoZSByZXNwb25zZS4gSSB0aGluayB3ZSBoYXZlIGFn cmVlZCB0aGF0ICJwcmVmaXgtbWFuYWdlbWVudCBJRCIgY2FuIGJlIGFuIEF1dG9ub21pYyBpbnRl bnQuDQoNCglIb3dldmVyLCBzb21lIEFTQSBwYXJhbWV0ZXJzIG1heSBhcHBlYXIgdG8gYmUgbW9y ZSBzdGF0aWMsIGFuZCBwZXJoYXBzIGNhbiBmdWxmaWxsIHRoZSBzdGF0ZW1lbnQgYmVsb3c6IA0K CTEuIE5ldHdvcmsgbGV2ZWwgcGFyYW1ldGVycywgbmVlZCB0byBiZSBkaXN0cmlidXRlZCBieSBH UkFTUCBhbmQgQUNQDQoJMi4gUmVsYXRpdmVseSBzdGF0aWMgY29tcGFyZWQgdG8gaW50ZW50LCBw ZXJoYXBzIG9ubHkgbmVlZGVkIHRvIGJlIGNvbmZpZ3VyZWQgb25jZQ0KCTMuIFVuYXZvaWRhYmxl IGluIGN1cnJlbnQgbmV0d29yayBlbnZpcm9ubWVudCwgbW9zdGx5IGZvciBlc3RhYmxpc2hpbmcg dGhlIG5ldHdvcmsgaW5mcmFzdHJ1Y3R1cmUNCgk0LiBSYXJlbHkgbmVlZCBjb29yZGluYXRpb25z IHdpdGggb3RoZXJzIHBhcmFtZXRlcnMsIHVzdWFsbHkgaXMgY2xvc2VkIHJlbGF0ZWQgdG8gb25l IHNwZWNpZmljIGF1dG9ub21pYyBmdW5jdGlvbg0KDQoJU29tZSBleGFtcGxlcyBjYW4gYmUgDQoJ V2hhdCBraW5kIG9mIElHUCB0aGUgQXV0b25vbWljIG5ldHdvcmsgZG9tYWluIHNob3VsZCB1c2Ug KHN1Y2ggYXMgSVMtSVMsIE9TUEYpOyANCglXaGF0IGtpbmQgb2YgdHJhbnNwb3J0IGxheWVyIG5l dHdvcmsgdGVjaG5vbG9neSB0aGUgQXV0b25vbWljIG5ldHdvcmsgZG9tYWluIHNob3VsZCB1c2Ug KHN1Y2ggYXMgTVBMUywgSVApDQoNCglTaG91bGQgdGhlc2UgQVNBIHBhcmFtZXRlcnMgYmUgcGFy dCBvZiBBTklNQSdzIHdvcmsgaXRlbXMganVzdCBhcyBpbnRlbnQ/DQoNCkJlc3QgUmVnYXJkcw0K Wm9uZ3BlbmcgRHUNCg0KLS0tLS1PcmlnaW5hbCBNZXNzYWdlLS0tLS0NCkZyb206IEJyaWFuIEUg Q2FycGVudGVyIFttYWlsdG86YnJpYW4uZS5jYXJwZW50ZXJAZ21haWwuY29tXSANClNlbnQ6IFRo dXJzZGF5LCBNYXJjaCAxNywgMjAxNiAyOjQxIEFNDQpUbzogRHV6b25ncGVuZzsgTWljaGFlbCBC ZWhyaW5nZXIgKG1iZWhyaW5nKTsgYW5pbWFAaWV0Zi5vcmcNCkNjOiBMYXVyZW50IENpYXZhZ2xp YTsgSsOpZmVyc29uIENhbXBvcyBOb2JyZTsgU2hlbmcgSmlhbmcNClN1YmplY3Q6IFJlOiBbQW5p bWFdIERvZXMgQXV0b25vbWljIG5ldHdvcmsgbmVlZCBzb21lIGludGVydmVudGlvbiBiZXlvbmQg aW50ZW50DQoNClllcywgSSB0aGluayB0aGF0IHdoYXRldmVyIHRoZSAicGhpbG9zb3BoeSIgb2Yg aW50ZW50IGlzLCB3ZSBzaG91bGQgbWFrZSBzdXJlIHRoYXQgdGhlIGltcGxlbWVudGF0aW9uIGlz IHByYWdtYXRpYyBhbmQgdGhhdCB3ZSBjYW4gYWRhcHQgdG8gb3BlcmF0b3IncyByZXF1aXJlbWVu dHMuDQoNClJlZ2FyZHMNCiAgIEJyaWFuDQoNCk9uIDE3LzAzLzIwMTYgMDE6NDEsIER1em9uZ3Bl bmcgd3JvdGU6DQo+IEhpLCBNaWNoYWVsLA0KPiANCj4gICAgICAgICAgVGhhbmtzIGZvciB0aGUg cmVwbHkuIEkgY2FuIHVuZGVyc3RhbmQgdGhhdCBpbnRlbnQgaXMgbm90IGluIHRoZSBjdXJyZW50 IG1pbGVzdG9uZXMsIGFuZCB5b3UgYXJlIHdvcmtpbmcgb24gc2V2ZXJhbCBtaWxlc3RvbmVzIGRy YWZ0cyBub3cuDQo+IA0KPiAgICAgICAgICBZb3UgYXJlIHdlbGNvbWUgdG8gY29udHJpYnV0ZSB0 byB0aGUgaW50ZW50IGRyYWZ0IGF0IGFueSB0aW1lLCBhbmQgdGhhbmtzIGluIGFkdmFuY2UuDQo+ IA0KPiAgICAgICAgICBBY2NvcmRpbmcgdG8geW91ciBzdGF0ZW1lbnRzLCBpbiBteSB1bmRlcnN0 YW5kaW5nLCB3ZSBjYW4ga2VlcCByZWdhcmRpbmcgdGhlICJpbnRlbnQiIGV4YW1wbGUgaW4gInBy ZWZpeC1tYW5hZ2VtZW50IElEIiBhcyBhbiBBdXRvbm9taWMgaW50ZW50Lg0KPiAgICAgICAgICBJ dCdzIHNlbWFudGljcyBhcmUganVzdCBsaWtlICJhbGwgbm9kZXMgb2YgdHlwZSB4IGRvIHRoaXM7 IHR5cGUgeSBkb2VzIHRoYXQiLCBzbyBpdCBpcyBhIGhpZ2ggbGV2ZWwgcG9saWN5Lg0KPiANCj4g ICAgICAgICAgSXQgcmVhbGx5IGhlbHBzIGEgbG90LiBUaGFua3MuDQo+IA0KPiAgICAgICAgICBJ IGdlbmVyYWxseSBhZ3JlZSB3aXRoIHlvdXIgc3RhdGVtZW50LCB3aGlsZSBzb21lIG90aGVyIHBl cnNvbmFsIHVuZGVyc3RhbmRpbmdzIGV4aXN0IChzaG93biBiZWxvdykuIEJ1dCBwZXJoYXBzIHRo ZXkgY2FuIGJlIGRpc2N1c3NlZCBsYXRlci4NCj4gDQo+IEJlc3QgUmVnYXJkcw0KPiBab25ncGVu ZyBEdQ0KPiANCj4gRnJvbTogTWljaGFlbCBCZWhyaW5nZXIgKG1iZWhyaW5nKSBbbWFpbHRvOm1i ZWhyaW5nQGNpc2NvLmNvbV0NCj4gU2VudDogV2VkbmVzZGF5LCBNYXJjaCAxNiwgMjAxNiA0OjE3 IFBNDQo+IFRvOiBEdXpvbmdwZW5nOyBhbmltYUBpZXRmLm9yZw0KPiBDYzogTGF1cmVudCBDaWF2 YWdsaWE7IErDqWZlcnNvbiBDYW1wb3MgTm9icmU7IFNoZW5nIEppYW5nDQo+IFN1YmplY3Q6IFJF OiBbQW5pbWFdIERvZXMgQXV0b25vbWljIG5ldHdvcmsgbmVlZCBzb21lIGludGVydmVudGlvbiAN Cj4gYmV5b25kIGludGVudA0KPiANCj4gSGkgWm9uZ3BlbmcsDQo+IA0KPiBZZXMsIHRoaXMgaXMg YSByZWN1cnJpbmcgcXVlc3Rpb24gYW5kIHRvcGljLiBMZXQgbWUgZ2l2ZSBteSB2aWV3IG9uIHR3 byBsZXZlbHM6DQo+IA0KPiBGaXJzdCwgQU5JTUEgaXNuJ3QgdGhlIG9ubHkgdGhpbmcgImluZmx1 ZW5jaW5nIiBvciAic3RlZXJpbmciIGEgbmV0d29yaywgYXQgbGVhc3Qgbm90IGluaXRpYWxseS4g VGhlcmUgaXMgc3RpbGwgdHJhZGl0aW9uYWwgY29uZmlnLCBwbHVzIGFsbCB0aGUgZXhpc3Rpbmcg YW5kIGVtZXJnaW5nIFNETiBtb2RlbHMuIE15IHZpZXcgaGFzIGFsd2F5cyBiZWVuIHRoYXQgQU5J TUEgc2V0cyBhICJkZWZhdWx0IiBwb2xpY3kgaW4gYSBuZXR3b3JrLCBhbmQgYWxsIG90aGVyIG1l YW5zIGFyZSBtb3JlIHNwZWNpZmljIGFuZCBjYW4gb3ZlcnJpZGUgaW50ZW50LiBUaGlzIGlzIGRl c2NyaWJlZCBpbiBSRkM3NTc1LiBNYWluIHBvaW50IGhlcmU6IEFOSU1BIEludGVudCBpc24ndCB0 aGUgb25seSB0b29sIGluIGEgbmV0d29yay4NCj4gDQo+IFtkdXpvbmdwZW5nXSBJIGFncmVlIHRo YXQgaW50ZW50IGNhbiBjb2V4aXN0IHdpdGggdHJhZGl0aW9uYWwgY29uZmlnLg0KPiBJIGdlbmVy YWxseSBhZ3JlZSB3aXRoIHRoZSBwb2xpY3kgcHJpb3JpdGllcywgYnV0IEkgaGF2ZSBzb21lIG90 aGVyIGNvbmNlcm5zLg0KPiBTb21lIGRlZmF1bHQgcG9saWNpZXMvcGFyYW1ldGVycyBzZXQgdGhy b3VnaCAiaW50ZW50IiBwZXJoYXBzIGFyZSBmdW5kYW1lbnRhbCBmb3IgdGhlIHVwbGF5ZXIgc2Vy dmljZXMuDQo+IEl0IHNob3VsZCBiZSBwcmV2ZW50ZWQgdGhhdCBodW1hbiBvcGVyYXRvciB1c2Vz IHNvbWUgQ0xJcyB0byBjaGFuZ2UgdGhlbSBieSBtaXN0YWtlLCBhbmQgdGhlIHVwbGF5ZXIgc2Vy dmljZXMgYXJlIGludGVycnVwdGVkLg0KPiBBdCBsZWFzdCwgc29tZSB3YXJuaW5nIHNob3VsZCBi ZSByZXBvcnRlZCB0byB0aGUgaHVtYW4gb3BlcmF0b3IuIA0KPiBbZHV6b25ncGVuZ10NCj4gDQo+ IEludGVudCBjb3VsZCBwZXJmZWN0bHkgd2VsbCBjb3ZlciBhIGhpZ2ggbGV2ZWwgcG9saWN5IHN1 Y2ggYXMgImFsbCBub2RlcyBvZiB0eXBlIHggZG8gdGhpczsgdHlwZSB5IGRvZXMgdGhhdCIuIChX aGljaCBJIHRoaW5rIGlzIHdoYXQgeW91IG1lbnRpb24gYmVsb3cuKSBXaGVuIHdlIHN0YXJ0ZWQg d2l0aCBJbnRlbnQsIHRoZSB2aXNpb24gd2FzIHRvIGtlZXAgaXQgdmVyeSBoaWdoIGxldmVsLCBt b3JlIGFsb25nIHRoZSBsaW5lICJkbyB0aGUgcmlnaHQgdGhpbmciIDopICBJdCBoYXMgYmVjb21l IGNsZWFyIHRoYXQgaW4gdG9kYXkncyBuZXR3b3JraW5nLCBwZW9wbGUgd2lsbCB3YW50IHRvIHB1 c2ggbW9yZSAicGFyYW1ldGVycyIgdGhhbiAicG9saWN5IiwgYW5kIEkgdGhpbmsgd2UnbGwganVz dCBoYXZlIHRvIGFja25vd2xlZGdlIHRoYXQuDQo+IA0KPiBQZXJzb25hbGx5IEkgdGhpbmsgaXQn bGwgcGFuIG91dCBsaWtlIHRoaXM6IEludGVudCB3aWxsIGhhdmUgc2VjdGlvbnMgcGVyIGF1dG9u b21pYyBmdW5jdGlvbi4gQW4gSW50ZW50IHBhcnNlciB3aWxsIHZhbGlkYXRlIG92ZXJhbGwgY29y cmVjdG5lc3MgKHNpZ25hdHVyZSwgZXRjKSwgdGhlbiBzZWdtZW50IHRoZSBJbnRlbnQgaW50byBp dHMgYml0czogT25lIHBhcnQgd2lsbCBiZSB0aGUgYXV0b25vbWljIGZhYnJpYyBpdHNlbGYsIHRo ZW4gdGhlcmUgd2lsbCBiZSBiaXRzIGZvciBlYWNoIGF1dG9ub21pYyBmdW5jdGlvbi4NCj4gDQo+ IFNvIHRoZSBJbnRlbnQgcGFyc2VyIG9uIGEgbm9kZSB3aWxsIGdyYWIgaW50ZW50LCBzZWUgdGhl cmUgaXMgc29tZSBJbnRlbnQgZm9yIGF1dG9ub21pYyBmdW5jdGlvbiBBLCB3aWxsIHNlZSB3aGV0 aGVyIHRoaXMgaXMgcnVubmluZyBsb2NhbGx5LCBhbmQgaWYgc28sIGl0J2xsIGp1c3QgcGFzcyB0 aGF0IHBpZWNlIG9mIEludGVudCB0byB0aGF0IGZ1bmN0aW9uLg0KPiANCj4gSW4gdGhpcyB3YXks IGEgZnVuY3Rpb24gb3duZXIgZGVmaW5lcyBib3RoIHRoZSBpbnRlbnQgaXRzZWxmLCBhbmQgaG93 IGl0IGlzIGNvbnN1bWVkIGJ5IGhpcy9oZXIgZnVuY3Rpb24uDQo+IA0KPiBbZHV6b25ncGVuZ11B Z3JlZS5bZHV6b25ncGVuZ10NCj4gDQo+IFRoZSBkb3duc2lkZSBpcyB0aGF0IHRoZXJlIGlzIG5v ICJjb250cm9sIiBhbi8vZCBwZW9wbGUgY2FuIGRvIHdoYXQgdGhleSB3YW50LCBpbiB0aGUgd29y c3QgY2FzZSBldmVuIHB1c2ggY29uZmlnLCB3aGljaCBpcyBjbGVhcmx5IE5PVCB3aGF0IEludGVu dCBpcyBtZWFudCB0byBkby4gSSdkIGhvcGUgdGhhdCBwZW9wbGUgZG9uJ3Qgc3RhcnQgd3JpdGlu ZyBJbnRlbnQgbGlrZSAibm9kZSAxNzogaGVyZSBpcyB5b3VyIENMSTsgbm9kZSAyMzogaGVyZSBp cyB5b3VyIENMSSI7IG5vZGUgMzc6IGhlcmUgaXMgeW91ciBDTEkiLiBCdXQgd2hhdCBJIGRlc2Ny aWJlIGFib3ZlIENPVUxEIGJlIGFidXNlZCBpbiB0aGlzIHdheS4gUkZDNzU3NSBzYXlzIHRoaXMg aXMgTk9UIEludGVudC4NCj4gDQo+IFtkdXpvbmdwZW5nXUFncmVlLiBJdCBpcyBoYXJkIHRvIGdp dmUgYW4gZXhhY3QgYm91bmRhcnkuDQo+IEhvd2V2ZXIsIEkgdGhpbmsgd2UgaGF2ZSB0d28gY2hv aWNlcyBoZXJlLg0KPiBUaGUgZmlyc3Qgb25lIGlzIHRvIGRlZmluZSBhIHBvbGljeSBmaXJzdCBp biBldmVyeSBBU0EsIGJ1dCB3aXRoIHNvbWUgb3BlbiBwYXJhbWV0ZXJzIHRoYXQgY2FuIGJlIGNv bmZpZ3VyZWQuIFdoYXQgd2UgbmVlZCB0byBkbyBpcyBqdXN0IHRvIGRpc3RyaWJ1dGUgdGhlIG5l dHdvcmstbGV2ZWwgcGFyYW1ldGVycyBmb3IgdGhhdCBmdW5jdGlvbi4gVGhpcyB3YXkgaXMgc3Vp dGFibGUgdG8gc29tZSByZWxhdGl2ZWx5IHN0YXRpYyBmdW5jdGlvbnMuDQo+IFRoZSBzZWNvbmQg aXMgbW9yZSBhZHZhbmNlZCwgaW4gdGhlIGlkZWFsIGNhc2UsIHRoZSBodW1hbiBvcGVyYXRvciBj YW4gY29tbXVuaWNhdGUgd2l0aCBuZXR3b3JrIHVzaW5nIG5hdHVyYWwgbGFuZ3VhZ2Ugb3Igc29t ZXRoaW5nIHNpbWlsYXIuIFRoaXMgd2F5IGlzIG1vcmUgZmxleGlibGUsIGJ1dCBoYXJkZXIgdG8g cmVhbGl6ZS4NCj4gW2R1em9uZ3BlbmddDQo+IA0KPiBJbiBvdGhlciB3b3JkczogTGV0J3Mgbm90 IGJlIHRvbyBwcmVzY3JpcHRpdmUsIGFuZCBob3BlIHRoYXQgZm9sa3Mgd2lsbCBkbyBtb3JlIG9y IGxlc3MgcmVhc29uYWJsZSB0aGluZ3MuDQo+IA0KPiBbZHV6b25ncGVuZ11BZ3JlZS5bZHV6b25n cGVuZ10NCj4gDQo+IEkndmUgYmVlbiBtZWFuaW5nIHRvIGNvbW1lbnQgb24gdGhlIEludGVudCBk cmFmdHMgYW5kIHBvc3NpYmx5IGNvbnRyaWJ1dGUsIEknbSBqdXN0IHN0cnVnZ2xpbmcgd2l0aCBw cmlvcml0aWVzIGF0IHRoZSBtb21lbnQgKGFzIHlvdSBjYW4gc2VlIGJ5IG15IHJlbGF0aXZlIHNp bGVuY2Ugb3ZlciB0aGUgcGFzdCB3ZWVrcykuDQo+IA0KPiBNaWNoYWVsDQo+IA0KPiANCj4gDQo+ IEZyb206IEFuaW1hIFttYWlsdG86YW5pbWEtYm91bmNlc0BpZXRmLm9yZ10gT24gQmVoYWxmIE9m IER1em9uZ3BlbmcNCj4gU2VudDogMTYgTWFyY2ggMjAxNiAwMjozOA0KPiBUbzogYW5pbWFAaWV0 Zi5vcmc8bWFpbHRvOmFuaW1hQGlldGYub3JnPg0KPiBDYzogTGF1cmVudCBDaWF2YWdsaWEgDQo+ IDxsYXVyZW50LmNpYXZhZ2xpYUBub2tpYS5jb208bWFpbHRvOmxhdXJlbnQuY2lhdmFnbGlhQG5v a2lhLmNvbT4+OyANCj4gSsOpZmVyc29uIENhbXBvcyBOb2JyZSANCj4gPGpjbm9icmVAaW5mLnVm cmdzLmJyPG1haWx0bzpqY25vYnJlQGluZi51ZnJncy5icj4+OyBTaGVuZyBKaWFuZyANCj4gPGpp YW5nc2hlbmdAaHVhd2VpLmNvbTxtYWlsdG86amlhbmdzaGVuZ0BodWF3ZWkuY29tPj4NCj4gU3Vi amVjdDogW0FuaW1hXSBEb2VzIEF1dG9ub21pYyBuZXR3b3JrIG5lZWQgc29tZSBpbnRlcnZlbnRp b24gYmV5b25kIA0KPiBpbnRlbnQNCj4gDQo+IEhlbGxvLCBldmVyeW9uZQ0KPiANCj4gICAgICAg ICAgSSB1bmRlcnN0YW5kIHRoYXQgQXV0b25vbWljIG5ldHdvcmsgc2hvdWxkIGJlIGFibGUgdG8g d29yayB3aXRoIGFzIGxlc3MgY29uZmlndXJhdGlvbiBhcyBwb3NzaWJsZS4gSW4gdGhlIGlkZWFs IGNhc2UsIGFuIEF1dG9ub21pYyBuZXR3b3JrIHNob3VsZCBiZSBhYmxlIHRvIHdvcmsgd2VsbCB3 aGlsZSBvbmx5IGludGVudCBpcyBuZWVkZWQgZnJvbSBodW1hbiBvcGVyYXRvcnMuDQo+IA0KPiAg ICAgICAgICBIb3dldmVyLCBJIHdvbmRlciB3aGV0aGVyIEF1dG9ub21pYyBuZXR3b3JrIG1heSBu ZWVkIHNvbWUgaW50ZXJ2ZW50aW9uIGZyb20gaHVtYW4gb3BlcmF0b3JzIHRocm91Z2ggc29tZSBs b3ctbGV2ZWwgcG9saWNpZXMgYmV5b25kIGludGVudHMuDQo+IA0KPiAgICAgICAgICBGb3IgZXhh bXBsZSwgaW4gdGhlIGh0dHBzOi8vdG9vbHMuaWV0Zi5vcmcvaHRtbC9kcmFmdC1pZXRmLWFuaW1h LXByZWZpeC1tYW5hZ2VtZW50LTAwLCBpdCBpcyBzdWdnZXN0ZWQgdGhhdCB0aGUgcHJlZml4IGxl bmd0aHMgZm9yIHRoZSBDU0csIEFTRywgUlNHIChkaWZmZXJlbnQgcm9sZXMgaW4gSVAgUkFOKSBj YW4gYmUgYXNzaWduZWQgYXMgYW4gImludGVudCIuIFRoZXNlIGNvbmZpZ3VyYXRpb24gcGFyYW1l dGVycyBuZWVkIHRvIGJlIGRpc3RyaWJ1dGVkIGluIHRoZSBhdXRvbm9taWMgZG9tYWluIHRvIGlu Zmx1ZW5jZSB0aGUgZGV0YWlsIGNvbmZpZ3VyYXRpb25zIG9uIGVhY2ggYXV0b25vbWljIG5vZGUu DQo+IA0KPiAgICAgICAgICBBcyBpbnRlbnQgaXMgZGVzY3JpYmVkIGFzIGFuIGFic3RyYWN0LCBk ZWNsYXJhdGl2ZSwgaGlnaC1sZXZlbCBwb2xpY3kgdXNlZCB0byBvcGVyYXRlIGFuIGF1dG9ub21p YyBkb21haW4sIHN1Y2ggYXMgYW4gZW50ZXJwcmlzZSBuZXR3b3JrLiBQZXJoYXBzIHdlIGNhbiBy ZW5hbWUgdGhlICJpbnRlbnQiIGluIHRoZSAicHJlZml4LW1hbmFnZW1lbnQgSUQiIHRvIHNvbWUg b3RoZXIgd29yZHMuDQo+IA0KPiBJcyAiQVNBIHBhcmFtZXRlcnMiIG9rIGhlcmU/DQo+IA0KPiBJ IHRoaW5rIHNvbWUgY2hhcmFjdGVyaXN0aWNzIGZvciB0aGlzIGtpbmQgb2YgbG93LWxldmVsIHBv bGljaWVzIGFyZQ0KPiANCj4gMS4gICAgICAgTmV0d29yayBsZXZlbCBwYXJhbWV0ZXJzLCBuZWVk IHRvIGJlIGRpc3RyaWJ1dGVkIGJ5IEdSQVNQIGFuZCBBQ1ANCj4gDQo+IDIuICAgICAgIFJlbGF0 aXZlbHkgc3RhdGljIGNvbXBhcmVkIHRvIGludGVudCwgcGVyaGFwcyBvbmx5IG5lZWRlZCB0byBi ZSBjb25maWd1cmVkIG9uY2UNCj4gDQo+IDMuICAgICAgIFVuYXZvaWRhYmxlIGluIGN1cnJlbnQg bmV0d29yayBlbnZpcm9ubWVudCwgbW9zdGx5IGZvciBlc3RhYmxpc2hpbmcgdGhlIG5ldHdvcmsg aW5mcmFzdHJ1Y3R1cmUNCj4gDQo+IDQuICAgICAgIFJhcmVseSBuZWVkIGNvb3JkaW5hdGlvbnMg d2l0aCBvdGhlcnMgcGFyYW1ldGVycywgdXN1YWxseSBpcyBjbG9zZWQgcmVsYXRlZCB0byBvbmUg c3BlY2lmaWMgYXV0b25vbWljIGZ1bmN0aW9uDQo+IA0KPiBXZWxjb21lIGZvciBjb21tZW50LiBZ b3VyIHN1Z2dlc3Rpb25zIHdpbGwgYmUgYXBwcmVjaWF0ZWQuDQo+IA0KPiANCj4gQmVzdCBSZWdh cmRzDQo+IFpvbmdwZW5nIER1DQo+IA0KPiANCj4gDQo+IF9fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fDQo+IEFuaW1hIG1haWxpbmcgbGlzdA0KPiBBbmltYUBp ZXRmLm9yZw0KPiBodHRwczovL3d3dy5pZXRmLm9yZy9tYWlsbWFuL2xpc3RpbmZvL2FuaW1hDQo+ IA0KDQo= From nobody Fri Mar 18 02:52:37 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1AE6812DAFC for ; Fri, 18 Mar 2016 02:52:35 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.522 X-Spam-Level: X-Spam-Status: No, score=-14.522 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H4=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id FIE8Ursci5NH for ; Fri, 18 Mar 2016 02:52:29 -0700 (PDT) Received: from alln-iport-3.cisco.com (alln-iport-3.cisco.com [173.37.142.90]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D4CD312DA28 for ; Fri, 18 Mar 2016 02:52:28 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=4288; q=dns/txt; s=iport; t=1458294748; x=1459504348; h=from:to:subject:date:message-id: content-transfer-encoding:mime-version; bh=V8DCbmHaQTEEftbnOpUF4va3p4jfPKJC+oBG9zMHyz0=; b=P3085GUe7RmJ5JOyXolGGyqXTRbKJU6WK2nyaE/Ls0KPy6N7kzuQypEJ CHLeMtbeedJ/aj/XBQ1/MzTDorJviyy3SmmznpsvmjOET6OzRcGZqZENh sTG5KSV1Js5KKKBGsyx+tU5v11Vm4Kzb8SZ4t7waLnNObfg031FAtvJb3 I=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D0AQB/z+tW/4oNJK1eg0WBSLoTAQ2Bb?= =?us-ascii?q?4c9OBQBAQEBAQEBZBwLhEg6MSABPkImAQQbiB+hcJ83AQEIAQEBARyGHo5WBY0?= =?us-ascii?q?7ihsBgSqMUoFshEqIWI8EAR4BAUKBfoFnik9+AQEB?= X-IronPort-AV: E=Sophos;i="5.24,354,1454976000"; d="scan'208";a="250152935" Received: from alln-core-5.cisco.com ([173.36.13.138]) by alln-iport-3.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 18 Mar 2016 09:52:27 +0000 Received: from XCH-ALN-008.cisco.com (xch-aln-008.cisco.com [173.36.7.18]) by alln-core-5.cisco.com (8.14.5/8.14.5) with ESMTP id u2I9qRcn011932 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL) for ; Fri, 18 Mar 2016 09:52:27 GMT Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-008.cisco.com (173.36.7.18) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 18 Mar 2016 04:52:26 -0500 Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 18 Mar 2016 04:52:27 -0500 From: "Michael Behringer (mbehring)" To: "anima@ietf.org" Thread-Topic: Intent yang model - some thoughts Thread-Index: AdGA84Aiy7EwHtVtTUWSCghLdxIQCw== Date: Fri, 18 Mar 2016 09:52:26 +0000 Message-ID: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.55.238.139] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Archived-At: Subject: [Anima] Intent yang model - some thoughts X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 18 Mar 2016 09:52:35 -0000 WRT the current discussion on what should be in Intent and what shouldn't, = as mentioned before, I think we should leave this to the owner of an autono= mic function to decide. This risks "abuse" in a way that wasn't foreseen, b= ut it gives maximum agility, which is what we need to get ANIMA off the gro= und, IMO. If we are in the nice position to find that it gets adopted a lot= , and if there are things that need fixing then, we can work on version 2.= =20 Several months ago I started with a yang model for Intent, but never finish= ed. Maybe this helps with this discussion. Note I'm not at all a yang exper= t, so please look more at the general structure for now. Not sure if it hel= ps the discussion...=20 +--rw intent +--rw meta-data=20 | +--rw intent-version uint32 // the version of the intent specifica= tion | +--rw timestamp uint32 // latest time stamp wins | +--rw administrator string // who created that intent | +--rw device? string // on which node was it created | +--rw execution-time uint32 // plus other parameters on when to exe= cute | +--rw expiration uint32 // when does this intent cease to be va= lid | +--rw feedback // how to let the operator know about = success/failure | +--rw signature ?? // intent must be signed | +--rw scope=20 +--rw domains=20 +--rw domain* [name]=20 +--rw name string // is there a fqdn type defined? +--rw meta-data // general about the domain | +--rw include-subdomains boolean // should sub-domains foll= ow this policy? | +--rw fail-open boolean // fail-open or fail-closed policy | +--rw maintenance-window // when can critical operations ta= ke place | +--rw ... // other general policies on domai= n level | ... | +--rw ani-intent +-- (contains all policies which the ANI should follow) | ex: ACP policies, coordination between AN functions, securit= y policies, etc | for ani-intent, scope is always "all" (ie all nodes must pro= cess) | ... ... All of the above MUST be standardised, because it refers to the AN infrastr= ucture itself. Of the below we need to standardise the structure, but not content for each= function Conceptually, we define Intent for a domain.=20 ... | +--rw intent-for-autonomic-functions +--rw function_1 [name] | +--rw scope // which devices in the do= main need to listen.=20 | | +--rw all boolean // all devices | | +--rw role string // as defined in certifica= te | | +--rw dynamic_role string // as determined dynamical= ly | | +--rw tag string // can be used for locatio= n=20 | +-- function_intent // to be defined by functi= on owner | +--rw function_2 [name] | +--rw scope // which devices in the do= main need to listen.=20 | | +--rw all boolean // all devices | | +--rw role string // as defined in certifica= te | | +--rw dynamic_role string // as determined dynamical= ly | | +--rw tag string // can be used for locatio= n=20 | +-- function_intent // to be defined by functi= on owner | ... =20 If a node receives Intent, it checks:=20 1) meta data: Mostly for forwarding and processing of Intent 2) domain info: domain specific policy 3) ANI Intent: Intent for the ANI itself; every node must be able to proces= s that 4) For each of the "function" blocks:=20 - should I run this function?=20 - do I run this function?=20 - if so, forward the rest of that block to the corresponding handler.=20 =20 I think we want a clean architectural separation between the ANI and the fu= nctions. This way every function can do what it wants. Thoughts?=20 Michael From nobody Fri Mar 18 06:57:46 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4458D12D968 for ; Fri, 18 Mar 2016 06:57:44 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.222 X-Spam-Level: X-Spam-Status: No, score=-4.222 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id h2zru7gueH9t for ; Fri, 18 Mar 2016 06:57:42 -0700 (PDT) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 03F9D12DA15 for ; Fri, 18 Mar 2016 06:56:14 -0700 (PDT) Received: from 172.18.7.190 (EHLO lhreml703-cah.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CKY94738; Fri, 18 Mar 2016 13:56:12 +0000 (GMT) Received: from NKGEML408-HUB.china.huawei.com (10.98.56.39) by lhreml703-cah.china.huawei.com (10.201.5.104) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 18 Mar 2016 13:56:12 +0000 Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml408-hub.china.huawei.com ([10.98.56.39]) with mapi id 14.03.0235.001; Fri, 18 Mar 2016 21:56:05 +0800 From: "Liubing (Leo)" To: "Michael Behringer (mbehring)" Thread-Topic: Updated Naming section in Reference Model draft Thread-Index: AQHRgR3ql7jQaa9H8ECIrxNf53jvGg== Date: Fri, 18 Mar 2016 13:56:05 +0000 Message-ID: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5BD20@nkgeml514-mbx.china.huawei.com> References: <5fb0758540334e3482096303f3ac4074@XCH-RCD-006.cisco.com> In-Reply-To: <5fb0758540334e3482096303f3ac4074@XCH-RCD-006.cisco.com> Accept-Language: en-US, zh-CN Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.98.117] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A090203.56EC08FD.0158, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: 2ab489fa71e0a9f2f2f847238dfd4343 Archived-At: Cc: "anima@ietf.org" Subject: [Anima] Updated Naming section in Reference Model draft X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 18 Mar 2016 13:57:44 -0000 Hi Michael and all, I've prepared some update texts for the naming section. Due to previous discussion, I think it's good to take the specific naming s= emantics/pattern out of the RM document. Then the RM document only keeps so= me general consideration of naming. If me or somebody want to propose a specific naming mechanism, we can alway= s write a dedicated draft for discussion. Following is the new text. It is basically the previous Section 4.1.1 with = minor revision. "4.1 Naming Inside a domain, each autonomic device should be assigned a name. Basic considerations for forming the names is as the following: - Uniqueness The names must not collide within one autonomic domain. It is acceptable th= at the names in different domains collide, since they could be distinguishe= d by domains. - Consistency The devices' naming should follow the same pattern within a domain. - Autonomous The names must be assigned automatically without any human intervene. It is recommended that the names are generated by the autonomic nodes thems= elves. Specific naming convention is out of the scope of this document." Any comments? Best regards, Bing > -----Original Message----- > From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com] > Sent: Thursday, March 17, 2016 12:36 AM > To: Liubing (Leo) > Subject: Naming section >=20 > Hi Bing, >=20 > It's time to update drafts... ;-) You did the naming section in the ANI= MA > reference doc. >=20 > We decided to take out all the semantics, and "uplevel" the naming > discussion significantly, without specifying a particular naming conventi= on. >=20 > Did you have time to think about this section? Can you provide updated te= xt? > Monday is the deadline... If you have updated text, I suggest to copy the > mailing list, to enable discussion. >=20 > Thoughts? >=20 > Michael From nobody Fri Mar 18 09:19:30 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 47D6D12D9AA for ; Fri, 18 Mar 2016 09:19:30 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.522 X-Spam-Level: X-Spam-Status: No, score=-14.522 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id O4Fpnwjanpf4 for ; Fri, 18 Mar 2016 09:19:29 -0700 (PDT) Received: from rcdn-iport-4.cisco.com (rcdn-iport-4.cisco.com [173.37.86.75]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D32E412D571 for ; Fri, 18 Mar 2016 09:19:28 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=2545; q=dns/txt; s=iport; t=1458317968; x=1459527568; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=Nspjlx5s42kDBiRMNnOheo/tf5RmJp0t7JFANGtv4qc=; b=CHnttjt8OM1t46QMrpfdlxVLvP29r45Gi6MqV7IWXkABu+AHTAm+IRzq iVlJdGlCR492RY7cAqmDFsv1qORKdmcCCu5XfEfb8cFtPMmnWo3BitXEy myk+pjheENT3w+9uzZtCrsJOcIqoEbeunZUd4pIpDkDaUj+mHsj/2O7Xf I=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D1AQA1KexW/4YNJK1eg0SBRQa6FgENg?= =?us-ascii?q?W+GDQKBLjgUAQEBAQEBAWQnhEEBAQEEOjQLDAQCAQgRBAEBHwkHMhQJCAIEDgU?= =?us-ascii?q?IiB/AdQEBAQEBAQEBAQEBAQEBAQEBAQEBARWGHoREihIFjTuFSYRTAY18gWyNI?= =?us-ascii?q?oYOiHcBHgEBQoIDGYFJaollfgEBAQ?= X-IronPort-AV: E=Sophos;i="5.24,355,1454976000"; d="scan'208";a="84262770" Received: from alln-core-12.cisco.com ([173.36.13.134]) by rcdn-iport-4.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 18 Mar 2016 16:19:28 +0000 Received: from XCH-RCD-009.cisco.com (xch-rcd-009.cisco.com [173.37.102.19]) by alln-core-12.cisco.com (8.14.5/8.14.5) with ESMTP id u2IGJRRB009219 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 18 Mar 2016 16:19:27 GMT Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-009.cisco.com (173.37.102.19) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 18 Mar 2016 11:19:27 -0500 Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 18 Mar 2016 11:19:26 -0500 From: "Michael Behringer (mbehring)" To: "Liubing (Leo)" Thread-Topic: Updated Naming section in Reference Model draft Thread-Index: AQHRgR30Ymxq0WGo1Uaj/vtkuG7QoJ9fYUjA Date: Fri, 18 Mar 2016 16:19:26 +0000 Message-ID: <84f3f4f01c974670a2c8608f6a12e7f4@XCH-RCD-006.cisco.com> References: <5fb0758540334e3482096303f3ac4074@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5BD20@nkgeml514-mbx.china.huawei.com> In-Reply-To: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5BD20@nkgeml514-mbx.china.huawei.com> Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.55.238.139] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Archived-At: Cc: "anima@ietf.org" Subject: Re: [Anima] Updated Naming section in Reference Model draft X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 18 Mar 2016 16:19:30 -0000 I've added this text in the draft for now. We need to discuss this in more = detail, in Buenos Aires.=20 Thanks Bing! Other co-authors: I'm compiling the next version of the reference draft, th= ere are quite a few contributions still outstanding... Michael > -----Original Message----- > From: Liubing (Leo) [mailto:leo.liubing@huawei.com] > Sent: 18 March 2016 14:56 > To: Michael Behringer (mbehring) > Cc: anima@ietf.org > Subject: Updated Naming section in Reference Model draft >=20 > Hi Michael and all, >=20 > I've prepared some update texts for the naming section. >=20 > Due to previous discussion, I think it's good to take the specific naming > semantics/pattern out of the RM document. Then the RM document only > keeps some general consideration of naming. > If me or somebody want to propose a specific naming mechanism, we can > always write a dedicated draft for discussion. >=20 > Following is the new text. It is basically the previous Section 4.1.1 wit= h minor > revision. >=20 > "4.1 Naming > Inside a domain, each autonomic device should be assigned a name. >=20 > Basic considerations for forming the names is as the following: > - Uniqueness > The names must not collide within one autonomic domain. It is acceptable > that the names in different domains collide, since they could be distingu= ished > by domains. > - Consistency > The devices' naming should follow the same pattern within a domain. > - Autonomous > The names must be assigned automatically without any human intervene. > It is recommended that the names are generated by the autonomic nodes > themselves. >=20 > Specific naming convention is out of the scope of this document." >=20 >=20 > Any comments? >=20 > Best regards, > Bing >=20 > > -----Original Message----- > > From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com] > > Sent: Thursday, March 17, 2016 12:36 AM > > To: Liubing (Leo) > > Subject: Naming section > > > > Hi Bing, > > > > It's time to update drafts... ;-) You did the naming section in the A= NIMA > > reference doc. > > > > We decided to take out all the semantics, and "uplevel" the naming > > discussion significantly, without specifying a particular naming conven= tion. > > > > Did you have time to think about this section? Can you provide updated > text? > > Monday is the deadline... If you have updated text, I suggest to copy > > the mailing list, to enable discussion. > > > > Thoughts? > > > > Michael From nobody Fri Mar 18 09:45:45 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D9AF112DAC7 for ; Fri, 18 Mar 2016 09:45:43 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.522 X-Spam-Level: X-Spam-Status: No, score=-14.522 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id TBful2NxmCKV for ; Fri, 18 Mar 2016 09:45:42 -0700 (PDT) Received: from rcdn-iport-6.cisco.com (rcdn-iport-6.cisco.com [173.37.86.77]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8919912D94C for ; Fri, 18 Mar 2016 09:45:36 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=758; q=dns/txt; s=iport; t=1458319536; x=1459529136; h=from:to:subject:date:message-id: content-transfer-encoding:mime-version; bh=TOJUzLZwEDjnBpSZP8Hu4XdPFpsIv/k2taKnJh0hHIg=; b=XN3Rwen/sVWvHs03Wrba/6N45AyA0DdrphjvGIhmcF9kUzYO1I5l7hlz 4v74hE5oqadtoeICByxfRibBevdeqBnIyLz7c2ZMy4Ymj6427iqhUDCS6 vgYAoVUTpJFvpZqFwDPRJ+O3b0FtKjNWyrK8Qh4gOT36MKg/WToOEYcnU s=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D1AQAjMOxW/4ENJK1eg0RTcga6FgENg?= =?us-ascii?q?W8lhxk4FAEBAQEBAQFkHAuESDpRAT5CJgEEG4gfDqE7nzIBAQEHAQEBAQEXBIY?= =?us-ascii?q?ehESKEgWXVwGFcIgMjw6PBQEeAQFCg2VqiWV+AQEB?= X-IronPort-AV: E=Sophos;i="5.24,356,1454976000"; d="scan'208";a="84258566" Received: from alln-core-9.cisco.com ([173.36.13.129]) by rcdn-iport-6.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 18 Mar 2016 16:45:34 +0000 Received: from XCH-RCD-009.cisco.com (xch-rcd-009.cisco.com [173.37.102.19]) by alln-core-9.cisco.com (8.14.5/8.14.5) with ESMTP id u2IGjYbq014526 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL) for ; Fri, 18 Mar 2016 16:45:34 GMT Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-009.cisco.com (173.37.102.19) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 18 Mar 2016 11:45:34 -0500 Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 18 Mar 2016 11:45:33 -0500 From: "Michael Behringer (mbehring)" To: "anima@ietf.org" Thread-Topic: New Reference draft Thread-Index: AdGBNR4yVj+XspWfT0WUwaGDZsnYsQ== Date: Fri, 18 Mar 2016 16:45:33 +0000 Message-ID: Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.55.238.139] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Archived-At: Subject: [Anima] New Reference draft X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 18 Mar 2016 16:45:44 -0000 I've received a couple of edits from co-authors, but many are still outstan= ding.=20 But the structure of the document is getting solid; I've re-ordered some bi= ts and edited to make things clearer, inserted references to the documents = that actually explain the protocols and mechanisms. The current draft is at= :=20 https://github.com/mbehring/ANIMA-Reference-Model/blob/master/draft-ietf-an= ima-reference-model-01-draft.txt Co-authors, especially folks who have committed to shortening the (*) secti= ons (everything not in scope of the current charter): It would be great if = you could provide a much shorter version of those sections, as requested by= our chairs.=20 I'll compile everything on Monday and submit.=20 Michael From nobody Sun Mar 20 12:17:08 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EAB7012D547 for ; Sun, 20 Mar 2016 12:17:06 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.7 X-Spam-Level: X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gETeamFMcYYf for ; Sun, 20 Mar 2016 12:17:04 -0700 (PDT) Received: from mail-pf0-x22f.google.com (mail-pf0-x22f.google.com [IPv6:2607:f8b0:400e:c00::22f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 7F9E712D5AD for ; Sun, 20 Mar 2016 12:17:04 -0700 (PDT) Received: by mail-pf0-x22f.google.com with SMTP id 4so107156527pfd.0 for ; Sun, 20 Mar 2016 12:17:04 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=zS3CFoHCehPx9nlRvWUPcKK7sb7yVVTmR2S8mL5eA+Y=; b=zXVSO43GvsXgN1v9QMFZz4954jtBhMKusFoR6QbeSb+tJZBNhnkxJZKeml3DGVb2n+ QVMputBGNAHMq0y07CFBFSrjhBhhSLAnKPZivPQK1jFwObYW407VTk9isaSfpEXXcsHg 61NnqgI8cFdmJUpir+VUBwoyUaPFx48MaCT0zj900NcPhJVOX1X+nDCFiWogw0Slc1CB Y3MMPlmPG4I/CDdpQSbbzPEGMuMaHLrM8pTVrnvZDlrDi9jXS98QZE5u0JUhcafgcSO8 3F+zuf3+v3RsrKS/0kZQrRzg70Qgc28dHsmPHs1lduMaCl51EWXqLo/hA55uk0d/1Lzc hvcg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=zS3CFoHCehPx9nlRvWUPcKK7sb7yVVTmR2S8mL5eA+Y=; b=gDFrn6DQjZyab7MB95BBajw00Ke+4Z1WAlHpLNOmdoBKEnOegzSuHXD+UrxvnYWiE/ UTYycLUT4zRCdG2m+XYlP41XcYm8a7/+sJIxhiScLKwSEg5J9B3urGK2UYiS9dbvWamy 1PhvZHzac7EUo7NivPdyXhlEkNfCaIdMk59IwHGDtxGd1aM/6SZ/d55EAonUVdcDBZMn il4gzVieWiGtMxWmfzQimFlvKW9QqXvJAr8sVm8yp5GpHn6783xiAd8kGDOIAYZs32Ow D/+PRF++ZnGzVeEaFMpjWuJ8VjaSsHOUNtlWRIVrHCLua/JCOunNTVfLe0kTqk/+EtH3 W+Zg== X-Gm-Message-State: AD7BkJKd27gDlowJhY9QlYgNMTerMcKBNtrcWB2nxnG0R9mpXmWtT1kaKZoeFTFNnrhYWw== X-Received: by 10.66.248.198 with SMTP id yo6mr39378327pac.54.1458501423913; Sun, 20 Mar 2016 12:17:03 -0700 (PDT) Received: from ?IPv6:2406:e007:7b9a:1:28cc:dc4c:9703:6781? ([2406:e007:7b9a:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id qy7sm35084841pab.34.2016.03.20.12.16.59 (version=TLSv1/SSLv3 cipher=OTHER); Sun, 20 Mar 2016 12:17:02 -0700 (PDT) To: Duzongpeng , "anima@ietf.org" References: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> <56E9A8B2.8070504@gmail.com> From: Brian E Carpenter Organization: University of Auckland Message-ID: <56EEF72A.5050309@gmail.com> Date: Mon, 21 Mar 2016 08:16:58 +1300 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Archived-At: Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "Michael Behringer \(mbehring\)" , Sheng Jiang Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 20 Mar 2016 19:17:07 -0000 Hi Zongpeng, > Should these ASA parameters be part of ANIMA's work items just as inten= t? As Michael's yang outline showed, there will be network-wide aspects to intent and also ASA-specific aspects. I think that in practice there will= be simple parameters in the ASA-specific parts, so (whether we use yang o= r some other syntax) I see no advantage in having a separate mechanism for the parameters - in a sense we never need to answer the question, we simp= ly treat all the information in the same way, intent or parameter. Maybe we should think about levels of specialization. Top level, network intent. Next level, intent for all ASAs supporting a given objective. Those two levels would be flooded by GRASP. Next level, intent for a sing= le ASA. That would be transmitted by unicast GRASP synchronization. Regards Brian On 18/03/2016 21:33, Duzongpeng wrote: > Hi, Brian >=20 > Thanks for the response. I think we have agreed that "prefix-managemen= t ID" can be an Autonomic intent. >=20 > However, some ASA parameters may appear to be more static, and perhaps= can fulfill the statement below:=20 > 1. Network level parameters, need to be distributed by GRASP and ACP > 2. Relatively static compared to intent, perhaps only needed to be con= figured once > 3. Unavoidable in current network environment, mostly for establishing= the network infrastructure > 4. Rarely need coordinations with others parameters, usually is closed= related to one specific autonomic function >=20 > Some examples can be=20 > What kind of IGP the Autonomic network domain should use (such as IS-I= S, OSPF);=20 > What kind of transport layer network technology the Autonomic network = domain should use (such as MPLS, IP) >=20 > Should these ASA parameters be part of ANIMA's work items just as inte= nt? >=20 > Best Regards > Zongpeng Du >=20 > -----Original Message----- > From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]=20 > Sent: Thursday, March 17, 2016 2:41 AM > To: Duzongpeng; Michael Behringer (mbehring); anima@ietf.org > Cc: Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang > Subject: Re: [Anima] Does Autonomic network need some intervention beyo= nd intent >=20 > Yes, I think that whatever the "philosophy" of intent is, we should mak= e sure that the implementation is pragmatic and that we can adapt to oper= ator's requirements. >=20 > Regards > Brian >=20 > On 17/03/2016 01:41, Duzongpeng wrote: >> Hi, Michael, >> >> Thanks for the reply. I can understand that intent is not in = the current milestones, and you are working on several milestones drafts = now. >> >> You are welcome to contribute to the intent draft at any time= , and thanks in advance. >> >> According to your statements, in my understanding, we can kee= p regarding the "intent" example in "prefix-management ID" as an Autonomi= c intent. >> It's semantics are just like "all nodes of type x do this; ty= pe y does that", so it is a high level policy. >> >> It really helps a lot. Thanks. >> >> I generally agree with your statement, while some other perso= nal understandings exist (shown below). But perhaps they can be discussed= later. >> >> Best Regards >> Zongpeng Du >> >> From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com] >> Sent: Wednesday, March 16, 2016 4:17 PM >> To: Duzongpeng; anima@ietf.org >> Cc: Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang >> Subject: RE: [Anima] Does Autonomic network need some intervention=20 >> beyond intent >> >> Hi Zongpeng, >> >> Yes, this is a recurring question and topic. Let me give my view on tw= o levels: >> >> First, ANIMA isn't the only thing "influencing" or "steering" a networ= k, at least not initially. There is still traditional config, plus all th= e existing and emerging SDN models. My view has always been that ANIMA se= ts a "default" policy in a network, and all other means are more specific= and can override intent. This is described in RFC7575. Main point here: = ANIMA Intent isn't the only tool in a network. >> >> [duzongpeng] I agree that intent can coexist with traditional config. >> I generally agree with the policy priorities, but I have some other co= ncerns. >> Some default policies/parameters set through "intent" perhaps are fund= amental for the uplayer services. >> It should be prevented that human operator uses some CLIs to change th= em by mistake, and the uplayer services are interrupted. >> At least, some warning should be reported to the human operator.=20 >> [duzongpeng] >> >> Intent could perfectly well cover a high level policy such as "all nod= es of type x do this; type y does that". (Which I think is what you menti= on below.) When we started with Intent, the vision was to keep it very hi= gh level, more along the line "do the right thing" :) It has become clea= r that in today's networking, people will want to push more "parameters" = than "policy", and I think we'll just have to acknowledge that. >> >> Personally I think it'll pan out like this: Intent will have sections = per autonomic function. An Intent parser will validate overall correctnes= s (signature, etc), then segment the Intent into its bits: One part will = be the autonomic fabric itself, then there will be bits for each autonomi= c function. >> >> So the Intent parser on a node will grab intent, see there is some Int= ent for autonomic function A, will see whether this is running locally, a= nd if so, it'll just pass that piece of Intent to that function. >> >> In this way, a function owner defines both the intent itself, and how = it is consumed by his/her function. >> >> [duzongpeng]Agree.[duzongpeng] >> >> The downside is that there is no "control" an//d people can do what th= ey want, in the worst case even push config, which is clearly NOT what In= tent is meant to do. I'd hope that people don't start writing Intent like= "node 17: here is your CLI; node 23: here is your CLI"; node 37: here is= your CLI". But what I describe above COULD be abused in this way. RFC757= 5 says this is NOT Intent. >> >> [duzongpeng]Agree. It is hard to give an exact boundary. >> However, I think we have two choices here. >> The first one is to define a policy first in every ASA, but with some = open parameters that can be configured. What we need to do is just to dis= tribute the network-level parameters for that function. This way is suita= ble to some relatively static functions. >> The second is more advanced, in the ideal case, the human operator can= communicate with network using natural language or something similar. Th= is way is more flexible, but harder to realize. >> [duzongpeng] >> >> In other words: Let's not be too prescriptive, and hope that folks wil= l do more or less reasonable things. >> >> [duzongpeng]Agree.[duzongpeng] >> >> I've been meaning to comment on the Intent drafts and possibly contrib= ute, I'm just struggling with priorities at the moment (as you can see by= my relative silence over the past weeks). >> >> Michael >> >> >> >> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng >> Sent: 16 March 2016 02:38 >> To: anima@ietf.org >> Cc: Laurent Ciavaglia=20 >> >;=20 >> J=C3=A9ferson Campos Nobre=20 >> >; Sheng Jiang=20 >> > >> Subject: [Anima] Does Autonomic network need some intervention beyond = >> intent >> >> Hello, everyone >> >> I understand that Autonomic network should be able to work wi= th as less configuration as possible. In the ideal case, an Autonomic net= work should be able to work well while only intent is needed from human o= perators. >> >> However, I wonder whether Autonomic network may need some int= ervention from human operators through some low-level policies beyond int= ents. >> >> For example, in the https://tools.ietf.org/html/draft-ietf-an= ima-prefix-management-00, it is suggested that the prefix lengths for the= CSG, ASG, RSG (different roles in IP RAN) can be assigned as an "intent"= =2E These configuration parameters need to be distributed in the autonomi= c domain to influence the detail configurations on each autonomic node. >> >> As intent is described as an abstract, declarative, high-leve= l policy used to operate an autonomic domain, such as an enterprise netwo= rk. Perhaps we can rename the "intent" in the "prefix-management ID" to s= ome other words. >> >> Is "ASA parameters" ok here? >> >> I think some characteristics for this kind of low-level policies are >> >> 1. Network level parameters, need to be distributed by GRASP and= ACP >> >> 2. Relatively static compared to intent, perhaps only needed to = be configured once >> >> 3. Unavoidable in current network environment, mostly for establ= ishing the network infrastructure >> >> 4. Rarely need coordinations with others parameters, usually is = closed related to one specific autonomic function >> >> Welcome for comment. Your suggestions will be appreciated. >> >> >> Best Regards >> Zongpeng Du >> >> >> >> _______________________________________________ >> Anima mailing list >> Anima@ietf.org >> https://www.ietf.org/mailman/listinfo/anima >> >=20 From nobody Sun Mar 20 12:27:45 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 32D0212D61E for ; Sun, 20 Mar 2016 12:27:43 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.7 X-Spam-Level: X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5Z3RCpObMkAS for ; Sun, 20 Mar 2016 12:27:41 -0700 (PDT) Received: from mail-pf0-x233.google.com (mail-pf0-x233.google.com [IPv6:2607:f8b0:400e:c00::233]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A425512D55C for ; Sun, 20 Mar 2016 12:27:39 -0700 (PDT) Received: by mail-pf0-x233.google.com with SMTP id u190so239021513pfb.3 for ; Sun, 20 Mar 2016 12:27:39 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=spCsDijeEam0/BrGPEAV77cdIeTuKFCk0RI5ErWyeg4=; b=R+TVfK9RgwCBKqBhgBFrH4+20hCMrnfkmkQnhB5kdeam+KdBoJ1Uk/YxfRJ4t7eefW UNIwshih8TPwPck8HJAeHfars3dkngpo8xGJemiSqE2Z4KV8jz05orpdosHOG+59sMbM sRBtR3fUu9wS3l8Dt3yP8NuPUZ/XsK/o/MMogrf8nyCgecJ2CwAnCcvt1JLryfGNV0NC +JWYTOQ0hzBQYDoCVJEPllEN8/eBvlXo+gNyADO4YaqKkPUaqdx52657tvdBcMDW9YqV Pw3nqYB2MNVbqw1KPoZqyovHfmZCr6JQkigxByfJFN3ovHfdKskHNwnvjpkgix47WyI7 S3Zw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=spCsDijeEam0/BrGPEAV77cdIeTuKFCk0RI5ErWyeg4=; b=LFX83OTvUUKd7z49g5kB7QqWcsaoXPyebV2h0eLiEqtwfYbA7Qv9Y32F/5HlpaK5Ge k8a91tOhOgjoynPFV9fvbhdsxLn68xYNP2xw44c/QqfIhkD8Wd071TZtTnPhEropxH9g YBL9vrKaBPoQUKf7DmT1nqmU7MMXAq7GpeOwx0XmByZsipWJvSnNEyCkwdUOD8fwWitH howDEUUtPOdFlJBCpJiW6Zwli5gW538MRHDysJGvSPLmQrkzWmMmhOdvc89xV+ip2L6A IQhgYX3QLaT/TI0qV4qMfxY7qJ8XaiaIKtAs230ADWzXf1bNvoZxgRV6/Nmft6SO2DY+ sBuQ== X-Gm-Message-State: AD7BkJK3Fn6MlOaWZ2Dbypaa4M3cihAWnJrHErwIv9dZkFUYsnUh5dMdyoaYeOTxbNKRvA== X-Received: by 10.66.220.162 with SMTP id px2mr40036692pac.15.1458502059196; Sun, 20 Mar 2016 12:27:39 -0700 (PDT) Received: from ?IPv6:2406:e007:7b9a:1:28cc:dc4c:9703:6781? ([2406:e007:7b9a:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id f65sm35095793pfd.47.2016.03.20.12.27.35 (version=TLSv1/SSLv3 cipher=OTHER); Sun, 20 Mar 2016 12:27:37 -0700 (PDT) To: "Michael Behringer (mbehring)" , "anima@ietf.org" References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> From: Brian E Carpenter Organization: University of Auckland Message-ID: <56EEF9A7.90709@gmail.com> Date: Mon, 21 Mar 2016 08:27:35 +1300 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0 MIME-Version: 1.0 In-Reply-To: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Archived-At: Subject: Re: [Anima] Intent yang model - some thoughts X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 20 Mar 2016 19:27:43 -0000 Hi Michael, At a glance, from someone who is a yang beginner, this looks plausible, although I think we need some analysis of the purpose of many of the items. Just to pick a random example, consider "maintenance-window". - is this fixed or does it change? - if it changes, how long in advance does the new intent get sent? - what are the semantics (must all ASAs stop operation, etc.)? - what happens if maintenance ends early? - what happens if maintenance ends late? In other words these items are not self-defining; each one needs work. A beginner's question: how is yang serialised on the wire? And another: can we send partial yang, i.e. only send the bits that changed? Regards Brian On 18/03/2016 22:52, Michael Behringer (mbehring) wrote: > WRT the current discussion on what should be in Intent and what shouldn't, as mentioned before, I think we should leave this to the owner of an autonomic function to decide. This risks "abuse" in a way that wasn't foreseen, but it gives maximum agility, which is what we need to get ANIMA off the ground, IMO. If we are in the nice position to find that it gets adopted a lot, and if there are things that need fixing then, we can work on version 2. > > Several months ago I started with a yang model for Intent, but never finished. Maybe this helps with this discussion. Note I'm not at all a yang expert, so please look more at the general structure for now. Not sure if it helps the discussion... > > +--rw intent > +--rw meta-data > | +--rw intent-version uint32 // the version of the intent specification > | +--rw timestamp uint32 // latest time stamp wins > | +--rw administrator string // who created that intent > | +--rw device? string // on which node was it created > | +--rw execution-time uint32 // plus other parameters on when to execute > | +--rw expiration uint32 // when does this intent cease to be valid > | +--rw feedback // how to let the operator know about success/failure > | +--rw signature ?? // intent must be signed > | > +--rw scope > +--rw domains > +--rw domain* [name] > +--rw name string // is there a fqdn type defined? > +--rw meta-data // general about the domain > | +--rw include-subdomains boolean // should sub-domains follow this policy? > | +--rw fail-open boolean // fail-open or fail-closed policy > | +--rw maintenance-window // when can critical operations take place > | +--rw ... // other general policies on domain level > | ... > | > +--rw ani-intent > +-- (contains all policies which the ANI should follow) > | ex: ACP policies, coordination between AN functions, security policies, etc > | for ani-intent, scope is always "all" (ie all nodes must process) > | ... > ... > All of the above MUST be standardised, because it refers to the AN infrastructure itself. > Of the below we need to standardise the structure, but not content for each function > Conceptually, we define Intent for a domain. > ... > | > +--rw intent-for-autonomic-functions > +--rw function_1 [name] > | +--rw scope // which devices in the domain need to listen. > | | +--rw all boolean // all devices > | | +--rw role string // as defined in certificate > | | +--rw dynamic_role string // as determined dynamically > | | +--rw tag string // can be used for location > | +-- function_intent // to be defined by function owner > | > +--rw function_2 [name] > | +--rw scope // which devices in the domain need to listen. > | | +--rw all boolean // all devices > | | +--rw role string // as defined in certificate > | | +--rw dynamic_role string // as determined dynamically > | | +--rw tag string // can be used for location > | +-- function_intent // to be defined by function owner > | > ... > > If a node receives Intent, it checks: > 1) meta data: Mostly for forwarding and processing of Intent > 2) domain info: domain specific policy > 3) ANI Intent: Intent for the ANI itself; every node must be able to process that > 4) For each of the "function" blocks: > - should I run this function? > - do I run this function? > - if so, forward the rest of that block to the corresponding handler. > > I think we want a clean architectural separation between the ANI and the functions. This way every function can do what it wants. > > Thoughts? > Michael > > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima > From nobody Sun Mar 20 12:50:47 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EF7EC12D59E for ; Sun, 20 Mar 2016 12:50:45 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.702 X-Spam-Level: X-Spam-Status: No, score=-2.702 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id faRr79Z6AWq9 for ; Sun, 20 Mar 2016 12:50:44 -0700 (PDT) Received: from maila2.tigertech.net (maila2.tigertech.net [208.80.4.152]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 04BA512D5A2 for ; Sun, 20 Mar 2016 12:50:44 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by maila2.tigertech.net (Postfix) with ESMTP id C7911240BE0; Sun, 20 Mar 2016 12:50:43 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1458503443; bh=3YjrFLN1uo+0MW3uqx3ULO6+gmOrjaZP6SAQahy3ryM=; h=Subject:To:References:From:Date:In-Reply-To:From; b=ncwrLpmEoTJQzH5VRPs1P0SuvqYRLgWXbtPh/3CITjIh1arVujPmIAfBPQbnOkU5d h7sv7KGcwczSgmf7BDLeYQQlfTkZ841hpR0AAf8XHd5WU9gJJI/wyKOamGWmyz+wmm vw7vwdfyhQ9Zix+M7QSvNyRKbcoOZxjIRA037n/U= X-Virus-Scanned: Debian amavisd-new at maila2.tigertech.net Received: from Joels-MacBook-Pro.local (unknown [88.131.67.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by maila2.tigertech.net (Postfix) with ESMTPSA id AB0E9240141; Sun, 20 Mar 2016 12:50:42 -0700 (PDT) To: Brian E Carpenter , "Michael Behringer (mbehring)" , "anima@ietf.org" References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <56EEF9A7.90709@gmail.com> From: "Joel M. Halpern" Message-ID: <56EEFF10.1050808@joelhalpern.com> Date: Sun, 20 Mar 2016 15:50:40 -0400 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.0 MIME-Version: 1.0 In-Reply-To: <56EEF9A7.90709@gmail.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Archived-At: Subject: Re: [Anima] Intent yang model - some thoughts X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 20 Mar 2016 19:50:46 -0000 There are two aspects of this model that have me very confused. First, in most policy work I have seen one specifies a set of policies, and then specifies what they apply to. This model seems to say "Here are a bunch of domains", and then for each domain it lists the policies. Structuring it this way means that reuse is harder. Somewhat related to that, I find the given metadata confusing: 1) There apparently is some notion that a domain has sub-domains, but that there may be things in the higher domain? And that you may or may not want the policy to recurse downwards. But if you want it to recurse downwards, you want it to recurse forever? (This looks like an optimization for sharing policies, but only across all sub-domains of a domain.) 2) Fail-open, maintenance-window, and other such items seem to be treated as being applicable and relevant to all policies in the domain. That is not at all obvious. 2a) For a given ASA / function, fail-open may be meaningful, and for an individual policy the goal may be fail-open or fail-closed. Or we may not even know what "fail" means. 2b) Even for policies where a maintenance window makes sense, it is unlikely that all policies will have the same maintenance window. Usually, since maintenance can be unpredictable in practice, policies have on-off switches individually. Second, am confused by the modeling for "functions". I will assume that these functions are ASAs. So they are network-wide targets of policy. But then I am confused as to what is being modeled here to help the function modeling. I see a boolean for whether the function applies to all devices. It seems unlikely that the choices are all / one, so a boolean seems very wrong. I think the idea is taht you can specify "all" or a single role. I would think a list of roles would actually be more useful. And then I see dynamic-role and tag. I have no idea what those are supposed to mean in an ASA-independent fashion. I do think that there are some ASA-policy attributes that are useful and common. Things like whether the policy is active. As a minor confusion, on the scale of things, I presume that function_intent is actually the placeholder for the function parameters? Yours, Joel On 3/20/16 3:27 PM, Brian E Carpenter wrote: > Hi Michael, > > At a glance, from someone who is a yang beginner, this looks plausible, > although I think we need some analysis of the purpose of many of the > items. Just to pick a random example, consider "maintenance-window". > - is this fixed or does it change? > - if it changes, how long in advance does the new intent get sent? > - what are the semantics (must all ASAs stop operation, etc.)? > - what happens if maintenance ends early? > - what happens if maintenance ends late? > In other words these items are not self-defining; each one needs work. > > A beginner's question: how is yang serialised on the wire? > > And another: can we send partial yang, i.e. only send the bits that changed? > > Regards > Brian > > > On 18/03/2016 22:52, Michael Behringer (mbehring) wrote: >> WRT the current discussion on what should be in Intent and what shouldn't, as mentioned before, I think we should leave this to the owner of an autonomic function to decide. This risks "abuse" in a way that wasn't foreseen, but it gives maximum agility, which is what we need to get ANIMA off the ground, IMO. If we are in the nice position to find that it gets adopted a lot, and if there are things that need fixing then, we can work on version 2. >> >> Several months ago I started with a yang model for Intent, but never finished. Maybe this helps with this discussion. Note I'm not at all a yang expert, so please look more at the general structure for now. Not sure if it helps the discussion... >> >> +--rw intent >> +--rw meta-data >> | +--rw intent-version uint32 // the version of the intent specification >> | +--rw timestamp uint32 // latest time stamp wins >> | +--rw administrator string // who created that intent >> | +--rw device? string // on which node was it created >> | +--rw execution-time uint32 // plus other parameters on when to execute >> | +--rw expiration uint32 // when does this intent cease to be valid >> | +--rw feedback // how to let the operator know about success/failure >> | +--rw signature ?? // intent must be signed >> | >> +--rw scope >> +--rw domains >> +--rw domain* [name] >> +--rw name string // is there a fqdn type defined? >> +--rw meta-data // general about the domain >> | +--rw include-subdomains boolean // should sub-domains follow this policy? >> | +--rw fail-open boolean // fail-open or fail-closed policy >> | +--rw maintenance-window // when can critical operations take place >> | +--rw ... // other general policies on domain level >> | ... >> | >> +--rw ani-intent >> +-- (contains all policies which the ANI should follow) >> | ex: ACP policies, coordination between AN functions, security policies, etc >> | for ani-intent, scope is always "all" (ie all nodes must process) >> | ... >> ... >> All of the above MUST be standardised, because it refers to the AN infrastructure itself. >> Of the below we need to standardise the structure, but not content for each function >> Conceptually, we define Intent for a domain. >> ... >> | >> +--rw intent-for-autonomic-functions >> +--rw function_1 [name] >> | +--rw scope // which devices in the domain need to listen. >> | | +--rw all boolean // all devices >> | | +--rw role string // as defined in certificate >> | | +--rw dynamic_role string // as determined dynamically >> | | +--rw tag string // can be used for location >> | +-- function_intent // to be defined by function owner >> | >> +--rw function_2 [name] >> | +--rw scope // which devices in the domain need to listen. >> | | +--rw all boolean // all devices >> | | +--rw role string // as defined in certificate >> | | +--rw dynamic_role string // as determined dynamically >> | | +--rw tag string // can be used for location >> | +-- function_intent // to be defined by function owner >> | >> ... >> >> If a node receives Intent, it checks: >> 1) meta data: Mostly for forwarding and processing of Intent >> 2) domain info: domain specific policy >> 3) ANI Intent: Intent for the ANI itself; every node must be able to process that >> 4) For each of the "function" blocks: >> - should I run this function? >> - do I run this function? >> - if so, forward the rest of that block to the corresponding handler. >> >> I think we want a clean architectural separation between the ANI and the functions. This way every function can do what it wants. >> >> Thoughts? >> Michael >> >> _______________________________________________ >> Anima mailing list >> Anima@ietf.org >> https://www.ietf.org/mailman/listinfo/anima >> > > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima > From nobody Sun Mar 20 18:25:57 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E0CDD12D56D for ; Sun, 20 Mar 2016 18:25:55 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.35 X-Spam-Level: X-Spam-Status: No, score=-2.35 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FREEMAIL_FORGED_FROMDOMAIN=0.249, FREEMAIL_FROM=0.001, HEADER_FROM_DIFFERENT_DOMAINS=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H2=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Txtsn6udW14q for ; Sun, 20 Mar 2016 18:25:53 -0700 (PDT) Received: from mail-io0-f172.google.com (mail-io0-f172.google.com [209.85.223.172]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A453A12D52E for ; Sun, 20 Mar 2016 18:25:53 -0700 (PDT) Received: by mail-io0-f172.google.com with SMTP id m184so195346953iof.1 for ; Sun, 20 Mar 2016 18:25:53 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to; bh=d558TRoYFJiEjMzbR1k3LamAM6wFjV6JNM9eL+rrR9w=; b=dTTgvc1nDSZc/SpqQoINahm/VlCeGulDg0i78S/rFG5jrgB/UlerLWqcnHtbbD7jk3 AaxlrFRFXvd6u0PVg/t7fjnR0hwVIYKng9FWsJVf6dtfg70+p65znL4myD7k6uzIa8Q6 Hagfu6ywsStJQUJ2OvdOX+RYOyT84628fGK6UNJWyXzAaDRIVXxiHzTQS8tAk+7OG3EI i4VZC7dX57qlHxdgwunlRM1gtyQneSP7dc0kr9R7JLpvl5tiqqwc9o9V/cXZ8D6HNnYJ lzT6bfjiG8LdwNKErl8KyO3W6+bqTaogbXkZtog1qVpys3+opoclfxW95OHGQtiWeNbB DckA== X-Gm-Message-State: AD7BkJLFhMG1Dqk75QUIMJP+zWMWjYwm47zZ3L5NoJ55ia4ADYYjvtXM3kZ+6ZVJdld7RINv+8ZYtyp2hvdCYQ== X-Received: by 10.107.26.201 with SMTP id a192mr27049097ioa.13.1458523553033; Sun, 20 Mar 2016 18:25:53 -0700 (PDT) MIME-Version: 1.0 References: In-Reply-To: From: =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= Date: Mon, 21 Mar 2016 01:25:43 +0000 Message-ID: To: "Michael Behringer (mbehring)" , "anima@ietf.org" Content-Type: multipart/alternative; boundary=001a113ff0bc467379052e84fa89 Archived-At: Subject: Re: [Anima] New Reference draft X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 01:25:56 -0000 --001a113ff0bc467379052e84fa89 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable > > Hi Michael. > I updated Sections Intent and Aggregated Reporting to make them shorter. > If it is not enough, just ping me. > Best. > > J=C3=A9ferson > > > Em sex, 18 de mar de 2016 =C3=A0s 13:45, Michael Behringer (mbehring) < > mbehring@cisco.com> escreveu: > >> I've received a couple of edits from co-authors, but many are still >> outstanding. >> >> But the structure of the document is getting solid; I've re-ordered some >> bits and edited to make things clearer, inserted references to the >> documents that actually explain the protocols and mechanisms. The curren= t >> draft is at: >> >> >> https://github.com/mbehring/ANIMA-Reference-Model/blob/master/draft-ietf= -anima-reference-model-01-draft.txt >> >> Co-authors, especially folks who have committed to shortening the (*) >> sections (everything not in scope of the current charter): It would be >> great if you could provide a much shorter version of those sections, as >> requested by our chairs. >> >> I'll compile everything on Monday and submit. >> Michael >> >> _______________________________________________ >> Anima mailing list >> Anima@ietf.org >> https://www.ietf.org/mailman/listinfo/anima >> > --001a113ff0bc467379052e84fa89 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
=
Hi Michael.
I updated Sections Intent and Aggregat= ed Reporting to make them shorter.
If it is not enough, just ping me.Best.

J=C3=A9ferson


Em sex, 18 de mar de 2016 =C3=A0s 13:45, Michael Behringer (mbehri= ng) <mbehring@ci= sco.com> escreveu:
I've= received a couple of edits from co-authors, but many are still outstanding= .

But the structure of the document is getting solid; I've re-ordered som= e bits and edited to make things clearer, inserted references to the docume= nts that actually explain the protocols and mechanisms. The current draft i= s at:

https://github.com/mbehring/ANIMA-Reference-Model/blob/master/draft-= ietf-anima-reference-model-01-draft.txt

Co-authors, especially folks who have committed to shortening the (*) secti= ons (everything not in scope of the current charter): It would be great if = you could provide a much shorter version of those sections, as requested by= our chairs.

I'll compile everything on Monday and submit.
Michael

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima
--001a113ff0bc467379052e84fa89-- From jeferson.nobre@ufrgs.br Sun Mar 20 15:45:00 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 899CF12D561 for ; Sun, 20 Mar 2016 15:45:00 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.302 X-Spam-Level: X-Spam-Status: No, score=-4.302 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=ufrgs.br Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eRE9e9Pc96AK for ; Sun, 20 Mar 2016 15:44:57 -0700 (PDT) Received: from delivery2.ufrgs.br (delivery2.ufrgs.br [143.54.2.212]) by ietfa.amsl.com (Postfix) with ESMTP id 3002912D555 for ; Sun, 20 Mar 2016 15:44:55 -0700 (PDT) Received: from delivery2.ufrgs.br (localhost [127.0.0.1]) by delivery2.ufrgs.br (Postfix) with ESMTP id A01DE20A0A4 for ; Sun, 20 Mar 2016 19:44:52 -0300 (BRT) DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=ufrgs.br; s=mail; t=1458513893; bh=Mev8nCoc5TRZF8cucCbq/gta0VCpb/svnN82KvBwU9w=; h=MIME-Version:References:In-Reply-To:From:Date:Message-ID:Subject: To:Content-Type; b=6mHh8zZSZQt9xmXm1gvFrwzFdwXq+52fPA+mtZfBGT5wQQ155Ax8Q1FZLcQzPreyn hXrFqTCan4dGN4IexxAllM1gCh99BOTQY1A1Wgs9IXXguP0lI8NwtF70uUKqfL6527 quHu27G4FPZZdSEK7WBTDg9fxDd9k9DGx9Imio9I= Received: from msa2.ufrgs.br (msa2.ufrgs.br [143.54.2.209]) by delivery2.ufrgs.br (Postfix) with ESMTP id 8D0833001B6 for ; Sun, 20 Mar 2016 19:44:52 -0300 (BRT) Received: by mail-io0-f172.google.com with SMTP id m184so193118752iof.1 for ; Sun, 20 Mar 2016 15:44:51 -0700 (PDT) X-Gm-Message-State: AD7BkJKDtzJlAeRIARPLfY5YxUvfDJ5WO/sfZoXkVkupeXPY4WwU94QRs7HxvYcr8+35dAxVENU/f58Npdd5Tg== X-Received: by 10.107.132.18 with SMTP id g18mr27945498iod.48.1458513889643; Sun, 20 Mar 2016 15:44:49 -0700 (PDT) MIME-Version: 1.0 References: In-Reply-To: From: =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= Date: Sun, 20 Mar 2016 22:44:40 +0000 X-Gmail-Original-Message-ID: Message-ID: To: "Michael Behringer (mbehring)" , "anima@ietf.org" Content-Type: multipart/alternative; boundary=001a113ecbe44af7ce052e82ba36 X-AV-Checked: ClamAV using ClamSMTP Archived-At: X-Mailman-Approved-At: Sun, 20 Mar 2016 20:01:49 -0700 Subject: Re: [Anima] New Reference draft X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 00:01:30 -0000 --001a113ecbe44af7ce052e82ba36 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi Michael. I updated Sections Intent and Aggregated Reporting to make them shorter. If it is not enough, just ping me. Best. J=C3=A9ferson Em sex, 18 de mar de 2016 =C3=A0s 13:45, Michael Behringer (mbehring) < mbehring@cisco.com> escreveu: > I've received a couple of edits from co-authors, but many are still > outstanding. > > But the structure of the document is getting solid; I've re-ordered some > bits and edited to make things clearer, inserted references to the > documents that actually explain the protocols and mechanisms. The current > draft is at: > > > https://github.com/mbehring/ANIMA-Reference-Model/blob/master/draft-ietf-= anima-reference-model-01-draft.txt > > Co-authors, especially folks who have committed to shortening the (*) > sections (everything not in scope of the current charter): It would be > great if you could provide a much shorter version of those sections, as > requested by our chairs. > > I'll compile everything on Monday and submit. > Michael > > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima > --001a113ecbe44af7ce052e82ba36 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Hi Michael.
I updated Sections Intent and Aggregated R= eporting to make them shorter.
If it is not enough, just ping me.=
Best.
J=C3=A9ferson

=
Em sex, 18 de mar de 2016 =C3=A0s 13:45, Michael Behringer= (mbehring) <mbehring@cisco.com> escreveu:
I've received a= couple of edits from co-authors, but many are still outstanding.

But the structure of the document is getting solid; I've re-ordered som= e bits and edited to make things clearer, inserted references to the docume= nts that actually explain the protocols and mechanisms. The current draft i= s at:

https://github.com/mbehring/ANIMA-Reference-Model/blob/master/draft-= ietf-anima-reference-model-01-draft.txt

Co-authors, especially folks who have committed to shortening the (*) secti= ons (everything not in scope of the current charter): It would be great if = you could provide a much shorter version of those sections, as requested by= our chairs.

I'll compile everything on Monday and submit.
Michael

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima
--001a113ecbe44af7ce052e82ba36-- From nobody Sun Mar 20 20:48:37 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 517A512D569 for ; Sun, 20 Mar 2016 20:48:35 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.222 X-Spam-Level: X-Spam-Status: No, score=-4.222 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bT9-74ISmvl5 for ; Sun, 20 Mar 2016 20:48:32 -0700 (PDT) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BF36B12D518 for ; Sun, 20 Mar 2016 20:48:31 -0700 (PDT) Received: from 172.18.7.190 (EHLO lhreml704-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGJ55117; Mon, 21 Mar 2016 03:48:29 +0000 (GMT) Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml704-cah.china.huawei.com (10.201.5.130) with Microsoft SMTP Server (TLS) id 14.3.235.1; Mon, 21 Mar 2016 03:48:28 +0000 Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0235.001; Mon, 21 Mar 2016 11:48:17 +0800 From: Duzongpeng To: Brian E Carpenter , "anima@ietf.org" Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAAE3aEAABGmDAABfNuwAAGs2jQAAIUCJgA== Date: Mon, 21 Mar 2016 03:48:16 +0000 Message-ID: References: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> <56E9A8B2.8070504@gmail.com> <56EEF72A.5050309@gmail.com> In-Reply-To: <56EEF72A.5050309@gmail.com> Accept-Language: zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.149.226] Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A090203.56EF6F0E.0046, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: 75564e734a57cfeb0e99f48b48eaa534 Archived-At: Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "Michael Behringer \(mbehring\)" , Sheng Jiang Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 03:48:35 -0000 SGksIEJyaWFuDQoNClRoYW5rcyBmb3IgeW91ciByZXNwb25zZS4gSSBhZ3JlZSB3aXRoIHlvdXIg aWRlYS4NCg0KQWNjb3JkaW5nIHRvIHJlY2VudGx5IGRpc2N1c3Npb24sIEkgdGhpbmsgcGVyaGFw cyB3ZSBjYW4gZ2l2ZSBBTklNQSBpbnRlbnQgYSBtb3JlIGdlbmVyYWwgc2NvcGUuDQoNCk9uIG9u ZSBzaWRlLCB3ZSBzaG91bGQgc3VwcG9ydCB0aGUgd29yayBvZiAiSW50ZW50IiBkZWZpbmVkIGlu IG90aGVyIHdvcmsgZ3JvdXBzLCB3aGljaCBtZWFucyB0aGUgImludGVudCIgc2hvdWxkIGJlIGFi bGUgdG8gYmUgaW5wdXR0ZWQgaW50byB0aGUgQXV0b25vbWljIG5ldHdvcmsgYW5kIHdvcmsgcHJv cGVybHk7IA0KDQpPbiB0aGUgb3RoZXIgc2lkZSwgc29tZSBwYXJhbWV0ZXJzIHRoYXQgYXJlIG5l ZWRlZCBieSBBdXRvbm9taWMgTmV0d29yaywgZXZlbiBtYXkgbm90IGJlIGNvbnNpZGVyZWQgYXMg ImludGVudCIgaW4gb3RoZXIgd29yayBncm91cHMsIHNob3VsZCBhbHNvIGJlIGluY2x1ZGVkIGlu IHRoZSBBTklNQSBpbnRlbnQgc2NvcGUsIHdoaWNoIG1heSBhbHNvIG5lZWQgdG8gYmUgZmxvb2Rl ZCBieSBHUkFTUCBhbmQgQUNQLg0KDQpPZiBjb3Vyc2UsIGFzIHNhaWQgYmUgTWljaGFlbCwgd2Ug YXJlIHN0aWxsIGF0IHByZWxpbWluYXJ5IHN0YWdlLCBhbmQgdGhlIG93bmVyIG9mIGFuIGF1dG9u b21pYyBmdW5jdGlvbiBzaG91bGQgZGVjaWRlIHdoYXQgaXMgbmVlZGVkLg0KSWYgdGhpbmdzIGRp ZCBub3QgZ28gcmlnaHRseSBpbiB0aGUgZGV2ZWxvcG1lbnQgb2YgQVNBcyBhbmQgbW9kaWZpY2F0 aW9ucyB3YXMgcHJvdmVkIHRvIGJlIG5lZWRlZCwgbWF5YmUgd2UgY2FuIHdvcmsgb24gdmVyc2lv biAyLg0KDQpCZXN0IFJlZ2FyZHMNClpvbmdwZW5nIER1DQoNCi0tLS0tT3JpZ2luYWwgTWVzc2Fn ZS0tLS0tDQpGcm9tOiBBbmltYSBbbWFpbHRvOmFuaW1hLWJvdW5jZXNAaWV0Zi5vcmddIE9uIEJl aGFsZiBPZiBCcmlhbiBFIENhcnBlbnRlcg0KU2VudDogTW9uZGF5LCBNYXJjaCAyMSwgMjAxNiAz OjE3IEFNDQpUbzogRHV6b25ncGVuZzsgYW5pbWFAaWV0Zi5vcmcNCkNjOiBMYXVyZW50IENpYXZh Z2xpYTsgSsOpZmVyc29uIENhbXBvcyBOb2JyZTsgTWljaGFlbCBCZWhyaW5nZXIgKG1iZWhyaW5n KTsgU2hlbmcgSmlhbmcNClN1YmplY3Q6IFJlOiBbQW5pbWFdIERvZXMgQXV0b25vbWljIG5ldHdv cmsgbmVlZCBzb21lIGludGVydmVudGlvbiBiZXlvbmQgaW50ZW50DQoNCkhpIFpvbmdwZW5nLA0K DQo+IFNob3VsZCB0aGVzZSBBU0EgcGFyYW1ldGVycyBiZSBwYXJ0IG9mIEFOSU1BJ3Mgd29yayBp dGVtcyBqdXN0IGFzIGludGVudD8NCg0KQXMgTWljaGFlbCdzIHlhbmcgb3V0bGluZSBzaG93ZWQs IHRoZXJlIHdpbGwgYmUgbmV0d29yay13aWRlIGFzcGVjdHMgdG8gaW50ZW50IGFuZCBhbHNvIEFT QS1zcGVjaWZpYyBhc3BlY3RzLiBJIHRoaW5rIHRoYXQgaW4gcHJhY3RpY2UgdGhlcmUgd2lsbCBi ZSBzaW1wbGUgcGFyYW1ldGVycyBpbiB0aGUgQVNBLXNwZWNpZmljIHBhcnRzLCBzbyAod2hldGhl ciB3ZSB1c2UgeWFuZyBvciBzb21lIG90aGVyIHN5bnRheCkgSSBzZWUgbm8gYWR2YW50YWdlIGlu IGhhdmluZyBhIHNlcGFyYXRlIG1lY2hhbmlzbSBmb3IgdGhlIHBhcmFtZXRlcnMgLSBpbiBhIHNl bnNlIHdlIG5ldmVyIG5lZWQgdG8gYW5zd2VyIHRoZSBxdWVzdGlvbiwgd2Ugc2ltcGx5IHRyZWF0 IGFsbCB0aGUgaW5mb3JtYXRpb24gaW4gdGhlIHNhbWUgd2F5LCBpbnRlbnQgb3IgcGFyYW1ldGVy Lg0KDQpNYXliZSB3ZSBzaG91bGQgdGhpbmsgYWJvdXQgbGV2ZWxzIG9mIHNwZWNpYWxpemF0aW9u LiBUb3AgbGV2ZWwsIG5ldHdvcmsgaW50ZW50LiBOZXh0IGxldmVsLCBpbnRlbnQgZm9yIGFsbCBB U0FzIHN1cHBvcnRpbmcgYSBnaXZlbiBvYmplY3RpdmUuDQpUaG9zZSB0d28gbGV2ZWxzIHdvdWxk IGJlIGZsb29kZWQgYnkgR1JBU1AuIE5leHQgbGV2ZWwsIGludGVudCBmb3IgYSBzaW5nbGUgQVNB LiBUaGF0IHdvdWxkIGJlIHRyYW5zbWl0dGVkIGJ5IHVuaWNhc3QgR1JBU1Agc3luY2hyb25pemF0 aW9uLg0KDQpSZWdhcmRzDQogICBCcmlhbg0KDQpPbiAxOC8wMy8yMDE2IDIxOjMzLCBEdXpvbmdw ZW5nIHdyb3RlOg0KPiBIaSwgQnJpYW4NCj4gDQo+IAlUaGFua3MgZm9yIHRoZSByZXNwb25zZS4g SSB0aGluayB3ZSBoYXZlIGFncmVlZCB0aGF0ICJwcmVmaXgtbWFuYWdlbWVudCBJRCIgY2FuIGJl IGFuIEF1dG9ub21pYyBpbnRlbnQuDQo+IA0KPiAJSG93ZXZlciwgc29tZSBBU0EgcGFyYW1ldGVy cyBtYXkgYXBwZWFyIHRvIGJlIG1vcmUgc3RhdGljLCBhbmQgcGVyaGFwcyBjYW4gZnVsZmlsbCB0 aGUgc3RhdGVtZW50IGJlbG93OiANCj4gCTEuIE5ldHdvcmsgbGV2ZWwgcGFyYW1ldGVycywgbmVl ZCB0byBiZSBkaXN0cmlidXRlZCBieSBHUkFTUCBhbmQgQUNQDQo+IAkyLiBSZWxhdGl2ZWx5IHN0 YXRpYyBjb21wYXJlZCB0byBpbnRlbnQsIHBlcmhhcHMgb25seSBuZWVkZWQgdG8gYmUgY29uZmln dXJlZCBvbmNlDQo+IAkzLiBVbmF2b2lkYWJsZSBpbiBjdXJyZW50IG5ldHdvcmsgZW52aXJvbm1l bnQsIG1vc3RseSBmb3IgZXN0YWJsaXNoaW5nIHRoZSBuZXR3b3JrIGluZnJhc3RydWN0dXJlDQo+ IAk0LiBSYXJlbHkgbmVlZCBjb29yZGluYXRpb25zIHdpdGggb3RoZXJzIHBhcmFtZXRlcnMsIHVz dWFsbHkgaXMgDQo+IGNsb3NlZCByZWxhdGVkIHRvIG9uZSBzcGVjaWZpYyBhdXRvbm9taWMgZnVu Y3Rpb24NCj4gDQo+IAlTb21lIGV4YW1wbGVzIGNhbiBiZSANCj4gCVdoYXQga2luZCBvZiBJR1Ag dGhlIEF1dG9ub21pYyBuZXR3b3JrIGRvbWFpbiBzaG91bGQgdXNlIChzdWNoIGFzIElTLUlTLCBP U1BGKTsgDQo+IAlXaGF0IGtpbmQgb2YgdHJhbnNwb3J0IGxheWVyIG5ldHdvcmsgdGVjaG5vbG9n eSB0aGUgQXV0b25vbWljIG5ldHdvcmsgDQo+IGRvbWFpbiBzaG91bGQgdXNlIChzdWNoIGFzIE1Q TFMsIElQKQ0KPiANCj4gCVNob3VsZCB0aGVzZSBBU0EgcGFyYW1ldGVycyBiZSBwYXJ0IG9mIEFO SU1BJ3Mgd29yayBpdGVtcyBqdXN0IGFzIGludGVudD8NCj4gDQo+IEJlc3QgUmVnYXJkcw0KPiBa b25ncGVuZyBEdQ0KPiANCj4gLS0tLS1PcmlnaW5hbCBNZXNzYWdlLS0tLS0NCj4gRnJvbTogQnJp YW4gRSBDYXJwZW50ZXIgW21haWx0bzpicmlhbi5lLmNhcnBlbnRlckBnbWFpbC5jb21dDQo+IFNl bnQ6IFRodXJzZGF5LCBNYXJjaCAxNywgMjAxNiAyOjQxIEFNDQo+IFRvOiBEdXpvbmdwZW5nOyBN aWNoYWVsIEJlaHJpbmdlciAobWJlaHJpbmcpOyBhbmltYUBpZXRmLm9yZw0KPiBDYzogTGF1cmVu dCBDaWF2YWdsaWE7IErDqWZlcnNvbiBDYW1wb3MgTm9icmU7IFNoZW5nIEppYW5nDQo+IFN1Ympl Y3Q6IFJlOiBbQW5pbWFdIERvZXMgQXV0b25vbWljIG5ldHdvcmsgbmVlZCBzb21lIGludGVydmVu dGlvbiANCj4gYmV5b25kIGludGVudA0KPiANCj4gWWVzLCBJIHRoaW5rIHRoYXQgd2hhdGV2ZXIg dGhlICJwaGlsb3NvcGh5IiBvZiBpbnRlbnQgaXMsIHdlIHNob3VsZCBtYWtlIHN1cmUgdGhhdCB0 aGUgaW1wbGVtZW50YXRpb24gaXMgcHJhZ21hdGljIGFuZCB0aGF0IHdlIGNhbiBhZGFwdCB0byBv cGVyYXRvcidzIHJlcXVpcmVtZW50cy4NCj4gDQo+IFJlZ2FyZHMNCj4gICAgQnJpYW4NCj4gDQo+ IE9uIDE3LzAzLzIwMTYgMDE6NDEsIER1em9uZ3Blbmcgd3JvdGU6DQo+PiBIaSwgTWljaGFlbCwN Cj4+DQo+PiAgICAgICAgICBUaGFua3MgZm9yIHRoZSByZXBseS4gSSBjYW4gdW5kZXJzdGFuZCB0 aGF0IGludGVudCBpcyBub3QgaW4gdGhlIGN1cnJlbnQgbWlsZXN0b25lcywgYW5kIHlvdSBhcmUg d29ya2luZyBvbiBzZXZlcmFsIG1pbGVzdG9uZXMgZHJhZnRzIG5vdy4NCj4+DQo+PiAgICAgICAg ICBZb3UgYXJlIHdlbGNvbWUgdG8gY29udHJpYnV0ZSB0byB0aGUgaW50ZW50IGRyYWZ0IGF0IGFu eSB0aW1lLCBhbmQgdGhhbmtzIGluIGFkdmFuY2UuDQo+Pg0KPj4gICAgICAgICAgQWNjb3JkaW5n IHRvIHlvdXIgc3RhdGVtZW50cywgaW4gbXkgdW5kZXJzdGFuZGluZywgd2UgY2FuIGtlZXAgcmVn YXJkaW5nIHRoZSAiaW50ZW50IiBleGFtcGxlIGluICJwcmVmaXgtbWFuYWdlbWVudCBJRCIgYXMg YW4gQXV0b25vbWljIGludGVudC4NCj4+ICAgICAgICAgIEl0J3Mgc2VtYW50aWNzIGFyZSBqdXN0 IGxpa2UgImFsbCBub2RlcyBvZiB0eXBlIHggZG8gdGhpczsgdHlwZSB5IGRvZXMgdGhhdCIsIHNv IGl0IGlzIGEgaGlnaCBsZXZlbCBwb2xpY3kuDQo+Pg0KPj4gICAgICAgICAgSXQgcmVhbGx5IGhl bHBzIGEgbG90LiBUaGFua3MuDQo+Pg0KPj4gICAgICAgICAgSSBnZW5lcmFsbHkgYWdyZWUgd2l0 aCB5b3VyIHN0YXRlbWVudCwgd2hpbGUgc29tZSBvdGhlciBwZXJzb25hbCB1bmRlcnN0YW5kaW5n cyBleGlzdCAoc2hvd24gYmVsb3cpLiBCdXQgcGVyaGFwcyB0aGV5IGNhbiBiZSBkaXNjdXNzZWQg bGF0ZXIuDQo+Pg0KPj4gQmVzdCBSZWdhcmRzDQo+PiBab25ncGVuZyBEdQ0KPj4NCj4+IEZyb206 IE1pY2hhZWwgQmVocmluZ2VyIChtYmVocmluZykgW21haWx0bzptYmVocmluZ0BjaXNjby5jb21d DQo+PiBTZW50OiBXZWRuZXNkYXksIE1hcmNoIDE2LCAyMDE2IDQ6MTcgUE0NCj4+IFRvOiBEdXpv bmdwZW5nOyBhbmltYUBpZXRmLm9yZw0KPj4gQ2M6IExhdXJlbnQgQ2lhdmFnbGlhOyBKw6lmZXJz b24gQ2FtcG9zIE5vYnJlOyBTaGVuZyBKaWFuZw0KPj4gU3ViamVjdDogUkU6IFtBbmltYV0gRG9l cyBBdXRvbm9taWMgbmV0d29yayBuZWVkIHNvbWUgaW50ZXJ2ZW50aW9uIA0KPj4gYmV5b25kIGlu dGVudA0KPj4NCj4+IEhpIFpvbmdwZW5nLA0KPj4NCj4+IFllcywgdGhpcyBpcyBhIHJlY3Vycmlu ZyBxdWVzdGlvbiBhbmQgdG9waWMuIExldCBtZSBnaXZlIG15IHZpZXcgb24gdHdvIGxldmVsczoN Cj4+DQo+PiBGaXJzdCwgQU5JTUEgaXNuJ3QgdGhlIG9ubHkgdGhpbmcgImluZmx1ZW5jaW5nIiBv ciAic3RlZXJpbmciIGEgbmV0d29yaywgYXQgbGVhc3Qgbm90IGluaXRpYWxseS4gVGhlcmUgaXMg c3RpbGwgdHJhZGl0aW9uYWwgY29uZmlnLCBwbHVzIGFsbCB0aGUgZXhpc3RpbmcgYW5kIGVtZXJn aW5nIFNETiBtb2RlbHMuIE15IHZpZXcgaGFzIGFsd2F5cyBiZWVuIHRoYXQgQU5JTUEgc2V0cyBh ICJkZWZhdWx0IiBwb2xpY3kgaW4gYSBuZXR3b3JrLCBhbmQgYWxsIG90aGVyIG1lYW5zIGFyZSBt b3JlIHNwZWNpZmljIGFuZCBjYW4gb3ZlcnJpZGUgaW50ZW50LiBUaGlzIGlzIGRlc2NyaWJlZCBp biBSRkM3NTc1LiBNYWluIHBvaW50IGhlcmU6IEFOSU1BIEludGVudCBpc24ndCB0aGUgb25seSB0 b29sIGluIGEgbmV0d29yay4NCj4+DQo+PiBbZHV6b25ncGVuZ10gSSBhZ3JlZSB0aGF0IGludGVu dCBjYW4gY29leGlzdCB3aXRoIHRyYWRpdGlvbmFsIGNvbmZpZy4NCj4+IEkgZ2VuZXJhbGx5IGFn cmVlIHdpdGggdGhlIHBvbGljeSBwcmlvcml0aWVzLCBidXQgSSBoYXZlIHNvbWUgb3RoZXIgY29u Y2VybnMuDQo+PiBTb21lIGRlZmF1bHQgcG9saWNpZXMvcGFyYW1ldGVycyBzZXQgdGhyb3VnaCAi aW50ZW50IiBwZXJoYXBzIGFyZSBmdW5kYW1lbnRhbCBmb3IgdGhlIHVwbGF5ZXIgc2VydmljZXMu DQo+PiBJdCBzaG91bGQgYmUgcHJldmVudGVkIHRoYXQgaHVtYW4gb3BlcmF0b3IgdXNlcyBzb21l IENMSXMgdG8gY2hhbmdlIHRoZW0gYnkgbWlzdGFrZSwgYW5kIHRoZSB1cGxheWVyIHNlcnZpY2Vz IGFyZSBpbnRlcnJ1cHRlZC4NCj4+IEF0IGxlYXN0LCBzb21lIHdhcm5pbmcgc2hvdWxkIGJlIHJl cG9ydGVkIHRvIHRoZSBodW1hbiBvcGVyYXRvci4gDQo+PiBbZHV6b25ncGVuZ10NCj4+DQo+PiBJ bnRlbnQgY291bGQgcGVyZmVjdGx5IHdlbGwgY292ZXIgYSBoaWdoIGxldmVsIHBvbGljeSBzdWNo IGFzICJhbGwgbm9kZXMgb2YgdHlwZSB4IGRvIHRoaXM7IHR5cGUgeSBkb2VzIHRoYXQiLiAoV2hp Y2ggSSB0aGluayBpcyB3aGF0IHlvdSBtZW50aW9uIGJlbG93LikgV2hlbiB3ZSBzdGFydGVkIHdp dGggSW50ZW50LCB0aGUgdmlzaW9uIHdhcyB0byBrZWVwIGl0IHZlcnkgaGlnaCBsZXZlbCwgbW9y ZSBhbG9uZyB0aGUgbGluZSAiZG8gdGhlIHJpZ2h0IHRoaW5nIiA6KSAgSXQgaGFzIGJlY29tZSBj bGVhciB0aGF0IGluIHRvZGF5J3MgbmV0d29ya2luZywgcGVvcGxlIHdpbGwgd2FudCB0byBwdXNo IG1vcmUgInBhcmFtZXRlcnMiIHRoYW4gInBvbGljeSIsIGFuZCBJIHRoaW5rIHdlJ2xsIGp1c3Qg aGF2ZSB0byBhY2tub3dsZWRnZSB0aGF0Lg0KPj4NCj4+IFBlcnNvbmFsbHkgSSB0aGluayBpdCds bCBwYW4gb3V0IGxpa2UgdGhpczogSW50ZW50IHdpbGwgaGF2ZSBzZWN0aW9ucyBwZXIgYXV0b25v bWljIGZ1bmN0aW9uLiBBbiBJbnRlbnQgcGFyc2VyIHdpbGwgdmFsaWRhdGUgb3ZlcmFsbCBjb3Jy ZWN0bmVzcyAoc2lnbmF0dXJlLCBldGMpLCB0aGVuIHNlZ21lbnQgdGhlIEludGVudCBpbnRvIGl0 cyBiaXRzOiBPbmUgcGFydCB3aWxsIGJlIHRoZSBhdXRvbm9taWMgZmFicmljIGl0c2VsZiwgdGhl biB0aGVyZSB3aWxsIGJlIGJpdHMgZm9yIGVhY2ggYXV0b25vbWljIGZ1bmN0aW9uLg0KPj4NCj4+ IFNvIHRoZSBJbnRlbnQgcGFyc2VyIG9uIGEgbm9kZSB3aWxsIGdyYWIgaW50ZW50LCBzZWUgdGhl cmUgaXMgc29tZSBJbnRlbnQgZm9yIGF1dG9ub21pYyBmdW5jdGlvbiBBLCB3aWxsIHNlZSB3aGV0 aGVyIHRoaXMgaXMgcnVubmluZyBsb2NhbGx5LCBhbmQgaWYgc28sIGl0J2xsIGp1c3QgcGFzcyB0 aGF0IHBpZWNlIG9mIEludGVudCB0byB0aGF0IGZ1bmN0aW9uLg0KPj4NCj4+IEluIHRoaXMgd2F5 LCBhIGZ1bmN0aW9uIG93bmVyIGRlZmluZXMgYm90aCB0aGUgaW50ZW50IGl0c2VsZiwgYW5kIGhv dyBpdCBpcyBjb25zdW1lZCBieSBoaXMvaGVyIGZ1bmN0aW9uLg0KPj4NCj4+IFtkdXpvbmdwZW5n XUFncmVlLltkdXpvbmdwZW5nXQ0KPj4NCj4+IFRoZSBkb3duc2lkZSBpcyB0aGF0IHRoZXJlIGlz IG5vICJjb250cm9sIiBhbi8vZCBwZW9wbGUgY2FuIGRvIHdoYXQgdGhleSB3YW50LCBpbiB0aGUg d29yc3QgY2FzZSBldmVuIHB1c2ggY29uZmlnLCB3aGljaCBpcyBjbGVhcmx5IE5PVCB3aGF0IElu dGVudCBpcyBtZWFudCB0byBkby4gSSdkIGhvcGUgdGhhdCBwZW9wbGUgZG9uJ3Qgc3RhcnQgd3Jp dGluZyBJbnRlbnQgbGlrZSAibm9kZSAxNzogaGVyZSBpcyB5b3VyIENMSTsgbm9kZSAyMzogaGVy ZSBpcyB5b3VyIENMSSI7IG5vZGUgMzc6IGhlcmUgaXMgeW91ciBDTEkiLiBCdXQgd2hhdCBJIGRl c2NyaWJlIGFib3ZlIENPVUxEIGJlIGFidXNlZCBpbiB0aGlzIHdheS4gUkZDNzU3NSBzYXlzIHRo aXMgaXMgTk9UIEludGVudC4NCj4+DQo+PiBbZHV6b25ncGVuZ11BZ3JlZS4gSXQgaXMgaGFyZCB0 byBnaXZlIGFuIGV4YWN0IGJvdW5kYXJ5Lg0KPj4gSG93ZXZlciwgSSB0aGluayB3ZSBoYXZlIHR3 byBjaG9pY2VzIGhlcmUuDQo+PiBUaGUgZmlyc3Qgb25lIGlzIHRvIGRlZmluZSBhIHBvbGljeSBm aXJzdCBpbiBldmVyeSBBU0EsIGJ1dCB3aXRoIHNvbWUgb3BlbiBwYXJhbWV0ZXJzIHRoYXQgY2Fu IGJlIGNvbmZpZ3VyZWQuIFdoYXQgd2UgbmVlZCB0byBkbyBpcyBqdXN0IHRvIGRpc3RyaWJ1dGUg dGhlIG5ldHdvcmstbGV2ZWwgcGFyYW1ldGVycyBmb3IgdGhhdCBmdW5jdGlvbi4gVGhpcyB3YXkg aXMgc3VpdGFibGUgdG8gc29tZSByZWxhdGl2ZWx5IHN0YXRpYyBmdW5jdGlvbnMuDQo+PiBUaGUg c2Vjb25kIGlzIG1vcmUgYWR2YW5jZWQsIGluIHRoZSBpZGVhbCBjYXNlLCB0aGUgaHVtYW4gb3Bl cmF0b3IgY2FuIGNvbW11bmljYXRlIHdpdGggbmV0d29yayB1c2luZyBuYXR1cmFsIGxhbmd1YWdl IG9yIHNvbWV0aGluZyBzaW1pbGFyLiBUaGlzIHdheSBpcyBtb3JlIGZsZXhpYmxlLCBidXQgaGFy ZGVyIHRvIHJlYWxpemUuDQo+PiBbZHV6b25ncGVuZ10NCj4+DQo+PiBJbiBvdGhlciB3b3Jkczog TGV0J3Mgbm90IGJlIHRvbyBwcmVzY3JpcHRpdmUsIGFuZCBob3BlIHRoYXQgZm9sa3Mgd2lsbCBk byBtb3JlIG9yIGxlc3MgcmVhc29uYWJsZSB0aGluZ3MuDQo+Pg0KPj4gW2R1em9uZ3BlbmddQWdy ZWUuW2R1em9uZ3BlbmddDQo+Pg0KPj4gSSd2ZSBiZWVuIG1lYW5pbmcgdG8gY29tbWVudCBvbiB0 aGUgSW50ZW50IGRyYWZ0cyBhbmQgcG9zc2libHkgY29udHJpYnV0ZSwgSSdtIGp1c3Qgc3RydWdn bGluZyB3aXRoIHByaW9yaXRpZXMgYXQgdGhlIG1vbWVudCAoYXMgeW91IGNhbiBzZWUgYnkgbXkg cmVsYXRpdmUgc2lsZW5jZSBvdmVyIHRoZSBwYXN0IHdlZWtzKS4NCj4+DQo+PiBNaWNoYWVsDQo+ Pg0KPj4NCj4+DQo+PiBGcm9tOiBBbmltYSBbbWFpbHRvOmFuaW1hLWJvdW5jZXNAaWV0Zi5vcmdd IE9uIEJlaGFsZiBPZiBEdXpvbmdwZW5nDQo+PiBTZW50OiAxNiBNYXJjaCAyMDE2IDAyOjM4DQo+ PiBUbzogYW5pbWFAaWV0Zi5vcmc8bWFpbHRvOmFuaW1hQGlldGYub3JnPg0KPj4gQ2M6IExhdXJl bnQgQ2lhdmFnbGlhDQo+PiA8bGF1cmVudC5jaWF2YWdsaWFAbm9raWEuY29tPG1haWx0bzpsYXVy ZW50LmNpYXZhZ2xpYUBub2tpYS5jb20+PjsNCj4+IErDqWZlcnNvbiBDYW1wb3MgTm9icmUNCj4+ IDxqY25vYnJlQGluZi51ZnJncy5icjxtYWlsdG86amNub2JyZUBpbmYudWZyZ3MuYnI+PjsgU2hl bmcgSmlhbmcgDQo+PiA8amlhbmdzaGVuZ0BodWF3ZWkuY29tPG1haWx0bzpqaWFuZ3NoZW5nQGh1 YXdlaS5jb20+Pg0KPj4gU3ViamVjdDogW0FuaW1hXSBEb2VzIEF1dG9ub21pYyBuZXR3b3JrIG5l ZWQgc29tZSBpbnRlcnZlbnRpb24gYmV5b25kIA0KPj4gaW50ZW50DQo+Pg0KPj4gSGVsbG8sIGV2 ZXJ5b25lDQo+Pg0KPj4gICAgICAgICAgSSB1bmRlcnN0YW5kIHRoYXQgQXV0b25vbWljIG5ldHdv cmsgc2hvdWxkIGJlIGFibGUgdG8gd29yayB3aXRoIGFzIGxlc3MgY29uZmlndXJhdGlvbiBhcyBw b3NzaWJsZS4gSW4gdGhlIGlkZWFsIGNhc2UsIGFuIEF1dG9ub21pYyBuZXR3b3JrIHNob3VsZCBi ZSBhYmxlIHRvIHdvcmsgd2VsbCB3aGlsZSBvbmx5IGludGVudCBpcyBuZWVkZWQgZnJvbSBodW1h biBvcGVyYXRvcnMuDQo+Pg0KPj4gICAgICAgICAgSG93ZXZlciwgSSB3b25kZXIgd2hldGhlciBB dXRvbm9taWMgbmV0d29yayBtYXkgbmVlZCBzb21lIGludGVydmVudGlvbiBmcm9tIGh1bWFuIG9w ZXJhdG9ycyB0aHJvdWdoIHNvbWUgbG93LWxldmVsIHBvbGljaWVzIGJleW9uZCBpbnRlbnRzLg0K Pj4NCj4+ICAgICAgICAgIEZvciBleGFtcGxlLCBpbiB0aGUgaHR0cHM6Ly90b29scy5pZXRmLm9y Zy9odG1sL2RyYWZ0LWlldGYtYW5pbWEtcHJlZml4LW1hbmFnZW1lbnQtMDAsIGl0IGlzIHN1Z2dl c3RlZCB0aGF0IHRoZSBwcmVmaXggbGVuZ3RocyBmb3IgdGhlIENTRywgQVNHLCBSU0cgKGRpZmZl cmVudCByb2xlcyBpbiBJUCBSQU4pIGNhbiBiZSBhc3NpZ25lZCBhcyBhbiAiaW50ZW50Ii4gVGhl c2UgY29uZmlndXJhdGlvbiBwYXJhbWV0ZXJzIG5lZWQgdG8gYmUgZGlzdHJpYnV0ZWQgaW4gdGhl IGF1dG9ub21pYyBkb21haW4gdG8gaW5mbHVlbmNlIHRoZSBkZXRhaWwgY29uZmlndXJhdGlvbnMg b24gZWFjaCBhdXRvbm9taWMgbm9kZS4NCj4+DQo+PiAgICAgICAgICBBcyBpbnRlbnQgaXMgZGVz Y3JpYmVkIGFzIGFuIGFic3RyYWN0LCBkZWNsYXJhdGl2ZSwgaGlnaC1sZXZlbCBwb2xpY3kgdXNl ZCB0byBvcGVyYXRlIGFuIGF1dG9ub21pYyBkb21haW4sIHN1Y2ggYXMgYW4gZW50ZXJwcmlzZSBu ZXR3b3JrLiBQZXJoYXBzIHdlIGNhbiByZW5hbWUgdGhlICJpbnRlbnQiIGluIHRoZSAicHJlZml4 LW1hbmFnZW1lbnQgSUQiIHRvIHNvbWUgb3RoZXIgd29yZHMuDQo+Pg0KPj4gSXMgIkFTQSBwYXJh bWV0ZXJzIiBvayBoZXJlPw0KPj4NCj4+IEkgdGhpbmsgc29tZSBjaGFyYWN0ZXJpc3RpY3MgZm9y IHRoaXMga2luZCBvZiBsb3ctbGV2ZWwgcG9saWNpZXMgYXJlDQo+Pg0KPj4gMS4gICAgICAgTmV0 d29yayBsZXZlbCBwYXJhbWV0ZXJzLCBuZWVkIHRvIGJlIGRpc3RyaWJ1dGVkIGJ5IEdSQVNQIGFu ZCBBQ1ANCj4+DQo+PiAyLiAgICAgICBSZWxhdGl2ZWx5IHN0YXRpYyBjb21wYXJlZCB0byBpbnRl bnQsIHBlcmhhcHMgb25seSBuZWVkZWQgdG8gYmUgY29uZmlndXJlZCBvbmNlDQo+Pg0KPj4gMy4g ICAgICAgVW5hdm9pZGFibGUgaW4gY3VycmVudCBuZXR3b3JrIGVudmlyb25tZW50LCBtb3N0bHkg Zm9yIGVzdGFibGlzaGluZyB0aGUgbmV0d29yayBpbmZyYXN0cnVjdHVyZQ0KPj4NCj4+IDQuICAg ICAgIFJhcmVseSBuZWVkIGNvb3JkaW5hdGlvbnMgd2l0aCBvdGhlcnMgcGFyYW1ldGVycywgdXN1 YWxseSBpcyBjbG9zZWQgcmVsYXRlZCB0byBvbmUgc3BlY2lmaWMgYXV0b25vbWljIGZ1bmN0aW9u DQo+Pg0KPj4gV2VsY29tZSBmb3IgY29tbWVudC4gWW91ciBzdWdnZXN0aW9ucyB3aWxsIGJlIGFw cHJlY2lhdGVkLg0KPj4NCj4+DQo+PiBCZXN0IFJlZ2FyZHMNCj4+IFpvbmdwZW5nIER1DQo+Pg0K Pj4NCj4+DQo+PiBfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f Xw0KPj4gQW5pbWEgbWFpbGluZyBsaXN0DQo+PiBBbmltYUBpZXRmLm9yZw0KPj4gaHR0cHM6Ly93 d3cuaWV0Zi5vcmcvbWFpbG1hbi9saXN0aW5mby9hbmltYQ0KPj4NCj4gDQoNCl9fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fDQpBbmltYSBtYWlsaW5nIGxpc3QN CkFuaW1hQGlldGYub3JnDQpodHRwczovL3d3dy5pZXRmLm9yZy9tYWlsbWFuL2xpc3RpbmZvL2Fu aW1hDQo= From nobody Mon Mar 21 02:31:38 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0CA8D12D664 for ; Mon, 21 Mar 2016 02:31:36 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.522 X-Spam-Level: X-Spam-Status: No, score=-14.522 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Ezm5LnBSRgZn for ; Mon, 21 Mar 2016 02:31:33 -0700 (PDT) Received: from rcdn-iport-6.cisco.com (rcdn-iport-6.cisco.com [173.37.86.77]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E412B12D53F for ; Mon, 21 Mar 2016 02:31:32 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=16176; q=dns/txt; s=iport; t=1458552692; x=1459762292; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=T5qfDBE9WrmjXxq5J9kIVaBV8tVXgpQGsoxcwLh66QY=; b=A6JOz31cKfG1O7kDUbQbYv3//rgxhydRRIWRuWpRhBtdCx8bIh3hj1Ig UPh1v9r7agtz2XtTMQWMRSnOwXFsXo1f6VfhgpgaNcsiSeRWugvYDTXMK QBJZWJ9w1CnVQ5siD7BD4acyxyYuPPq+pa/KoSKWkMkwdEYx2EK4yTV7O 8=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D+AQC0vu9W/5BdJa1egzNTcgauUYtNA?= =?us-ascii?q?Q2BbxcKhWwCHIEGOBQBAQEBAQEBZCeEQQEBAQMBAQEBIBE6CwUHBAIBCBEEAQE?= =?us-ascii?q?BAgIjAwICAh8GCxQBCAgCBAENBQiICgMKCA6vAYodDYRiAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEBAQEBAQEBEQR8hSKDRX+CPoIUgmqCVgWXJjEBhXCGHoFujw6HMYdUAR4BAUK?= =?us-ascii?q?CAxmBSWoBiFslFn4BAQE?= X-IronPort-AV: E=Sophos;i="5.24,370,1454976000"; d="scan'208";a="84926946" Received: from rcdn-core-8.cisco.com ([173.37.93.144]) by rcdn-iport-6.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 21 Mar 2016 09:31:31 +0000 Received: from XCH-RCD-008.cisco.com (xch-rcd-008.cisco.com [173.37.102.18]) by rcdn-core-8.cisco.com (8.14.5/8.14.5) with ESMTP id u2L9VV8c030719 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Mon, 21 Mar 2016 09:31:31 GMT Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-008.cisco.com (173.37.102.18) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Mon, 21 Mar 2016 04:31:30 -0500 Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Mon, 21 Mar 2016 04:31:30 -0500 From: "Michael Behringer (mbehring)" To: Brian E Carpenter , Duzongpeng , "anima@ietf.org" Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAAE3aEAAH6dCAABPXnqAAHsO/gAAEuC0cA== Date: Mon, 21 Mar 2016 09:31:30 +0000 Message-ID: <782ad410ef844630a942c27f6279a49f@XCH-RCD-006.cisco.com> References: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> <56E9A8B2.8070504@gmail.com> <56EEF72A.5050309@gmail.com> In-Reply-To: <56EEF72A.5050309@gmail.com> Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.55.238.139] Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 MIME-Version: 1.0 Archived-At: Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , Sheng Jiang Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 09:31:36 -0000 PiAtLS0tLU9yaWdpbmFsIE1lc3NhZ2UtLS0tLQ0KPiBGcm9tOiBCcmlhbiBFIENhcnBlbnRlciBb bWFpbHRvOmJyaWFuLmUuY2FycGVudGVyQGdtYWlsLmNvbV0NCj4gU2VudDogMjAgTWFyY2ggMjAx NiAyMDoxNw0KPiBUbzogRHV6b25ncGVuZyA8ZHV6b25ncGVuZ0BodWF3ZWkuY29tPjsgYW5pbWFA aWV0Zi5vcmcNCj4gQ2M6IExhdXJlbnQgQ2lhdmFnbGlhIDxsYXVyZW50LmNpYXZhZ2xpYUBub2tp YS5jb20+OyBKw6lmZXJzb24gQ2FtcG9zIE5vYnJlDQo+IDxqY25vYnJlQGluZi51ZnJncy5icj47 IFNoZW5nIEppYW5nIDxqaWFuZ3NoZW5nQGh1YXdlaS5jb20+OyBNaWNoYWVsDQo+IEJlaHJpbmdl ciAobWJlaHJpbmcpIDxtYmVocmluZ0BjaXNjby5jb20+DQo+IFN1YmplY3Q6IFJlOiBbQW5pbWFd IERvZXMgQXV0b25vbWljIG5ldHdvcmsgbmVlZCBzb21lIGludGVydmVudGlvbg0KPiBiZXlvbmQg aW50ZW50DQo+IA0KPiBIaSBab25ncGVuZywNCj4gDQo+ID4gU2hvdWxkIHRoZXNlIEFTQSBwYXJh bWV0ZXJzIGJlIHBhcnQgb2YgQU5JTUEncyB3b3JrIGl0ZW1zIGp1c3QgYXMNCj4gaW50ZW50Pw0K PiANCj4gQXMgTWljaGFlbCdzIHlhbmcgb3V0bGluZSBzaG93ZWQsIHRoZXJlIHdpbGwgYmUgbmV0 d29yay13aWRlIGFzcGVjdHMgdG8NCj4gaW50ZW50IGFuZCBhbHNvIEFTQS1zcGVjaWZpYyBhc3Bl Y3RzLiBJIHRoaW5rIHRoYXQgaW4gcHJhY3RpY2UgdGhlcmUgd2lsbCBiZQ0KPiBzaW1wbGUgcGFy YW1ldGVycyBpbiB0aGUgQVNBLXNwZWNpZmljIHBhcnRzLCBzbyAod2hldGhlciB3ZSB1c2UgeWFu ZyBvcg0KPiBzb21lIG90aGVyIHN5bnRheCkgSSBzZWUgbm8gYWR2YW50YWdlIGluIGhhdmluZyBh IHNlcGFyYXRlIG1lY2hhbmlzbSBmb3INCj4gdGhlIHBhcmFtZXRlcnMgLSBpbiBhIHNlbnNlIHdl IG5ldmVyIG5lZWQgdG8gYW5zd2VyIHRoZSBxdWVzdGlvbiwgd2UNCj4gc2ltcGx5IHRyZWF0IGFs bCB0aGUgaW5mb3JtYXRpb24gaW4gdGhlIHNhbWUgd2F5LCBpbnRlbnQgb3IgcGFyYW1ldGVyLg0K DQpBZ3JlZSBzbyBmYXIuIA0KIA0KPiBNYXliZSB3ZSBzaG91bGQgdGhpbmsgYWJvdXQgbGV2ZWxz IG9mIHNwZWNpYWxpemF0aW9uLiBUb3AgbGV2ZWwsIG5ldHdvcmsNCj4gaW50ZW50LiBOZXh0IGxl dmVsLCBpbnRlbnQgZm9yIGFsbCBBU0FzIHN1cHBvcnRpbmcgYSBnaXZlbiBvYmplY3RpdmUuDQo+ IFRob3NlIHR3byBsZXZlbHMgd291bGQgYmUgZmxvb2RlZCBieSBHUkFTUC4gTmV4dCBsZXZlbCwg aW50ZW50IGZvciBhIHNpbmdsZQ0KPiBBU0EuIFRoYXQgd291bGQgYmUgdHJhbnNtaXR0ZWQgYnkg dW5pY2FzdCBHUkFTUCBzeW5jaHJvbml6YXRpb24uDQoNCkhlcmUgSSBkb24ndCBoYXZlIHRoZSBz YW1lIHZpZXcuIFRoZSBmaXJzdCB0d28gbGV2ZWxzIHdlIGFncmVlIG9uLiBBbnl0aGluZyB0aGF0 J3MgZm9yIGEgc3BlY2lmaWMgbm9kZSBkb2Vzbid0IHF1YWxpZnkgYXMgIkludGVudCIgaW4gbXkg dmlldy4gKElnbm9yZSB0aGF0IHRoZSBmbG9vZGluZyBtZWNoYW5pc20gbWF5IGNvbnRhaW4gdW5p Y2FzdCB0cmFuc21pc3Npb25zIGFzIHBhcnQgb2YgdGhlIG92ZXJhbGwgc2NoZW1lKS4gDQoNCklm IEludGVudCBpcyBhIGhpZ2gtbGV2ZWwgcG9saWN5ICh3aGljaCBJIHRoaW5rIHdlIGFsbCBhZ3Jl ZSBvbj8pIHRoZW4gSSBjYW4gdGhpbmsgb2YgdHdvIG1vcmUgInRoaW5ncyIgdGhhdCBhcmUgTk9U IEludGVudCwgdG8gYWRkIHRvIG15IGxpc3Qgb2YgIm5vdC1JbnRlbnQiOiANCg0KLSBhbnl0aGlu ZyB0aGF0J3MgZm9yIGEgc3BlY2lmaWMgbm9kZQ0KLSBhbnl0aGluZyB0aGF0J3MgbmVnb3RpYXRl ZCBkeW5hbWljYWxseS4gKEludGVudCBpcyBhbHdheXMgb3JpZ2luYXRpbmcgZnJvbSBzb21ldGhp bmcgbm9uLXRlY2huaWNhbCAtIGl0J3MgYSBwb2xpY3kpLg0KLSBhbnl0aGluZyB0aGF0IG5lZWRz IHRvIGNoYW5nZSBmcmVxdWVudGx5IChoaWdoLWxldmVsIHBvbGljeSBpcyBub3JtYWxseSBwcmV0 dHkgc3RhYmxlKQ0KLSBhbnl0aGluZyB0aGF0J3Mgbm90IGEgInBvbGljeSIsIGllIGRvZXNuJ3Qg Y29tZSBmcm9tIGEgbm9uLXRlY2huaWNhbCByZWFzb25pbmcuICh0aGF0J3Mgbm90IGEgZ29vZCB3 YXkgdG8gZGVmaW5lIGl0LCBidXQgY2FuJ3QgdGhpbmsgb2YgYSBiZXR0ZXIgb25lIHJpZ2h0IG5v dykNCg0KSSB0aGluayB5b3UncmUgc3VnZ2VzdGluZyBhbiBvcHRpbWlzYXRpb24gb2YgSW50ZW50 ICpkaXN0cmlidXRpb24qLCBCcmlhbiwgcmlnaHQ/IFdoZXJlLCBpZiBJIGtub3cgb25seSBhIGZl dyBub2RlcyBydW4gYSBjZXJ0YWluIGZ1bmN0aW9uLCBJIGp1c3Qgc2VuZCB0aGF0IHBhcnRpY3Vs YXIgSW50ZW50IHRvIHRob3NlIG5vZGVzIG9ubHk/IChpcyB0aGF0IHdoYXQgeW91J3JlIHRoaW5r aW5nIG9mPykgDQoNClRoaXMgd291bGQgY2VydGFpbmx5IGJlIHRlbXB0aW5nLCBidXQgSSB0aGlu ayBpdCdzIGEgc2xpcHBlcnkgc2xvcGUsIGFuZCB3b3VsZCBxdWlja2x5IGdldCBpbnRvIGNvbmZs aWN0IHdpdGggY3VycmVudCBwb2xpY3kgbW9kZWxzLiBJJ2QgcHJlZmVyIGlmIHdlIGtlZXAgIm91 ciIgSW50ZW50IGFzICJtdXN0IGJlIGZsb29kZWQgZXZlcnl3aGVyZSIuIElmIGl0IGlzIHJlYWxs eSBoaWdoIGxldmVsIGd1aWRhbmNlLCB0aGUgb3B0aW1pemF0aW9uIHNob3VsZG4ndCBiZSBuZWVk ZWQuIFNhaWQgZGlmZmVyZW50bHk6IElmIHdlIHN0YXJ0IHB1c2hpbmcgaHVnZSBkYXRhIG1vZGVs cywgbWF5YmUgd2UncmUgb24gdGhlIHdyb25nIHRyYWNrIGluIHRoZSBmaXJzdCBwbGFjZT8gDQoN Ck1pY2hhZWwNCg0KPiBSZWdhcmRzDQo+ICAgIEJyaWFuDQo+IA0KPiBPbiAxOC8wMy8yMDE2IDIx OjMzLCBEdXpvbmdwZW5nIHdyb3RlOg0KPiA+IEhpLCBCcmlhbg0KPiA+DQo+ID4gCVRoYW5rcyBm b3IgdGhlIHJlc3BvbnNlLiBJIHRoaW5rIHdlIGhhdmUgYWdyZWVkIHRoYXQgInByZWZpeC0NCj4g bWFuYWdlbWVudCBJRCIgY2FuIGJlIGFuIEF1dG9ub21pYyBpbnRlbnQuDQo+ID4NCj4gPiAJSG93 ZXZlciwgc29tZSBBU0EgcGFyYW1ldGVycyBtYXkgYXBwZWFyIHRvIGJlIG1vcmUgc3RhdGljLCBh bmQNCj4gcGVyaGFwcyBjYW4gZnVsZmlsbCB0aGUgc3RhdGVtZW50IGJlbG93Og0KPiA+IAkxLiBO ZXR3b3JrIGxldmVsIHBhcmFtZXRlcnMsIG5lZWQgdG8gYmUgZGlzdHJpYnV0ZWQgYnkgR1JBU1Ag YW5kDQo+IEFDUA0KPiA+IAkyLiBSZWxhdGl2ZWx5IHN0YXRpYyBjb21wYXJlZCB0byBpbnRlbnQs IHBlcmhhcHMgb25seSBuZWVkZWQgdG8gYmUNCj4gY29uZmlndXJlZCBvbmNlDQo+ID4gCTMuIFVu YXZvaWRhYmxlIGluIGN1cnJlbnQgbmV0d29yayBlbnZpcm9ubWVudCwgbW9zdGx5IGZvcg0KPiBl c3RhYmxpc2hpbmcgdGhlIG5ldHdvcmsgaW5mcmFzdHJ1Y3R1cmUNCj4gPiAJNC4gUmFyZWx5IG5l ZWQgY29vcmRpbmF0aW9ucyB3aXRoIG90aGVycyBwYXJhbWV0ZXJzLCB1c3VhbGx5IGlzDQo+ID4g Y2xvc2VkIHJlbGF0ZWQgdG8gb25lIHNwZWNpZmljIGF1dG9ub21pYyBmdW5jdGlvbg0KPiA+DQo+ ID4gCVNvbWUgZXhhbXBsZXMgY2FuIGJlDQo+ID4gCVdoYXQga2luZCBvZiBJR1AgdGhlIEF1dG9u b21pYyBuZXR3b3JrIGRvbWFpbiBzaG91bGQgdXNlIChzdWNoIGFzDQo+IElTLUlTLCBPU1BGKTsN Cj4gPiAJV2hhdCBraW5kIG9mIHRyYW5zcG9ydCBsYXllciBuZXR3b3JrIHRlY2hub2xvZ3kgdGhl IEF1dG9ub21pYw0KPiBuZXR3b3JrDQo+ID4gZG9tYWluIHNob3VsZCB1c2UgKHN1Y2ggYXMgTVBM UywgSVApDQo+ID4NCj4gPiAJU2hvdWxkIHRoZXNlIEFTQSBwYXJhbWV0ZXJzIGJlIHBhcnQgb2Yg QU5JTUEncyB3b3JrIGl0ZW1zIGp1c3QgYXMNCj4gaW50ZW50Pw0KPiA+DQo+ID4gQmVzdCBSZWdh cmRzDQo+ID4gWm9uZ3BlbmcgRHUNCj4gPg0KPiA+IC0tLS0tT3JpZ2luYWwgTWVzc2FnZS0tLS0t DQo+ID4gRnJvbTogQnJpYW4gRSBDYXJwZW50ZXIgW21haWx0bzpicmlhbi5lLmNhcnBlbnRlckBn bWFpbC5jb21dDQo+ID4gU2VudDogVGh1cnNkYXksIE1hcmNoIDE3LCAyMDE2IDI6NDEgQU0NCj4g PiBUbzogRHV6b25ncGVuZzsgTWljaGFlbCBCZWhyaW5nZXIgKG1iZWhyaW5nKTsgYW5pbWFAaWV0 Zi5vcmcNCj4gPiBDYzogTGF1cmVudCBDaWF2YWdsaWE7IErDqWZlcnNvbiBDYW1wb3MgTm9icmU7 IFNoZW5nIEppYW5nDQo+ID4gU3ViamVjdDogUmU6IFtBbmltYV0gRG9lcyBBdXRvbm9taWMgbmV0 d29yayBuZWVkIHNvbWUgaW50ZXJ2ZW50aW9uDQo+ID4gYmV5b25kIGludGVudA0KPiA+DQo+ID4g WWVzLCBJIHRoaW5rIHRoYXQgd2hhdGV2ZXIgdGhlICJwaGlsb3NvcGh5IiBvZiBpbnRlbnQgaXMs IHdlIHNob3VsZCBtYWtlDQo+IHN1cmUgdGhhdCB0aGUgaW1wbGVtZW50YXRpb24gaXMgcHJhZ21h dGljIGFuZCB0aGF0IHdlIGNhbiBhZGFwdCB0bw0KPiBvcGVyYXRvcidzIHJlcXVpcmVtZW50cy4N Cj4gPg0KPiA+IFJlZ2FyZHMNCj4gPiAgICBCcmlhbg0KPiA+DQo+ID4gT24gMTcvMDMvMjAxNiAw MTo0MSwgRHV6b25ncGVuZyB3cm90ZToNCj4gPj4gSGksIE1pY2hhZWwsDQo+ID4+DQo+ID4+ICAg ICAgICAgIFRoYW5rcyBmb3IgdGhlIHJlcGx5LiBJIGNhbiB1bmRlcnN0YW5kIHRoYXQgaW50ZW50 IGlzIG5vdCBpbiB0aGUgY3VycmVudA0KPiBtaWxlc3RvbmVzLCBhbmQgeW91IGFyZSB3b3JraW5n IG9uIHNldmVyYWwgbWlsZXN0b25lcyBkcmFmdHMgbm93Lg0KPiA+Pg0KPiA+PiAgICAgICAgICBZ b3UgYXJlIHdlbGNvbWUgdG8gY29udHJpYnV0ZSB0byB0aGUgaW50ZW50IGRyYWZ0IGF0IGFueSB0 aW1lLCBhbmQNCj4gdGhhbmtzIGluIGFkdmFuY2UuDQo+ID4+DQo+ID4+ICAgICAgICAgIEFjY29y ZGluZyB0byB5b3VyIHN0YXRlbWVudHMsIGluIG15IHVuZGVyc3RhbmRpbmcsIHdlIGNhbiBrZWVw DQo+IHJlZ2FyZGluZyB0aGUgImludGVudCIgZXhhbXBsZSBpbiAicHJlZml4LW1hbmFnZW1lbnQg SUQiIGFzIGFuIEF1dG9ub21pYw0KPiBpbnRlbnQuDQo+ID4+ICAgICAgICAgIEl0J3Mgc2VtYW50 aWNzIGFyZSBqdXN0IGxpa2UgImFsbCBub2RlcyBvZiB0eXBlIHggZG8gdGhpczsgdHlwZSB5IGRv ZXMNCj4gdGhhdCIsIHNvIGl0IGlzIGEgaGlnaCBsZXZlbCBwb2xpY3kuDQo+ID4+DQo+ID4+ICAg ICAgICAgIEl0IHJlYWxseSBoZWxwcyBhIGxvdC4gVGhhbmtzLg0KPiA+Pg0KPiA+PiAgICAgICAg ICBJIGdlbmVyYWxseSBhZ3JlZSB3aXRoIHlvdXIgc3RhdGVtZW50LCB3aGlsZSBzb21lIG90aGVy IHBlcnNvbmFsDQo+IHVuZGVyc3RhbmRpbmdzIGV4aXN0IChzaG93biBiZWxvdykuIEJ1dCBwZXJo YXBzIHRoZXkgY2FuIGJlIGRpc2N1c3NlZA0KPiBsYXRlci4NCj4gPj4NCj4gPj4gQmVzdCBSZWdh cmRzDQo+ID4+IFpvbmdwZW5nIER1DQo+ID4+DQo+ID4+IEZyb206IE1pY2hhZWwgQmVocmluZ2Vy IChtYmVocmluZykgW21haWx0bzptYmVocmluZ0BjaXNjby5jb21dDQo+ID4+IFNlbnQ6IFdlZG5l c2RheSwgTWFyY2ggMTYsIDIwMTYgNDoxNyBQTQ0KPiA+PiBUbzogRHV6b25ncGVuZzsgYW5pbWFA aWV0Zi5vcmcNCj4gPj4gQ2M6IExhdXJlbnQgQ2lhdmFnbGlhOyBKw6lmZXJzb24gQ2FtcG9zIE5v YnJlOyBTaGVuZyBKaWFuZw0KPiA+PiBTdWJqZWN0OiBSRTogW0FuaW1hXSBEb2VzIEF1dG9ub21p YyBuZXR3b3JrIG5lZWQgc29tZSBpbnRlcnZlbnRpb24NCj4gPj4gYmV5b25kIGludGVudA0KPiA+ Pg0KPiA+PiBIaSBab25ncGVuZywNCj4gPj4NCj4gPj4gWWVzLCB0aGlzIGlzIGEgcmVjdXJyaW5n IHF1ZXN0aW9uIGFuZCB0b3BpYy4gTGV0IG1lIGdpdmUgbXkgdmlldyBvbiB0d28NCj4gbGV2ZWxz Og0KPiA+Pg0KPiA+PiBGaXJzdCwgQU5JTUEgaXNuJ3QgdGhlIG9ubHkgdGhpbmcgImluZmx1ZW5j aW5nIiBvciAic3RlZXJpbmciIGEgbmV0d29yaywgYXQNCj4gbGVhc3Qgbm90IGluaXRpYWxseS4g VGhlcmUgaXMgc3RpbGwgdHJhZGl0aW9uYWwgY29uZmlnLCBwbHVzIGFsbCB0aGUgZXhpc3Rpbmcg YW5kDQo+IGVtZXJnaW5nIFNETiBtb2RlbHMuIE15IHZpZXcgaGFzIGFsd2F5cyBiZWVuIHRoYXQg QU5JTUEgc2V0cyBhICJkZWZhdWx0Ig0KPiBwb2xpY3kgaW4gYSBuZXR3b3JrLCBhbmQgYWxsIG90 aGVyIG1lYW5zIGFyZSBtb3JlIHNwZWNpZmljIGFuZCBjYW4gb3ZlcnJpZGUNCj4gaW50ZW50LiBU aGlzIGlzIGRlc2NyaWJlZCBpbiBSRkM3NTc1LiBNYWluIHBvaW50IGhlcmU6IEFOSU1BIEludGVu dCBpc24ndCB0aGUNCj4gb25seSB0b29sIGluIGEgbmV0d29yay4NCj4gPj4NCj4gPj4gW2R1em9u Z3BlbmddIEkgYWdyZWUgdGhhdCBpbnRlbnQgY2FuIGNvZXhpc3Qgd2l0aCB0cmFkaXRpb25hbCBj b25maWcuDQo+ID4+IEkgZ2VuZXJhbGx5IGFncmVlIHdpdGggdGhlIHBvbGljeSBwcmlvcml0aWVz LCBidXQgSSBoYXZlIHNvbWUgb3RoZXINCj4gY29uY2VybnMuDQo+ID4+IFNvbWUgZGVmYXVsdCBw b2xpY2llcy9wYXJhbWV0ZXJzIHNldCB0aHJvdWdoICJpbnRlbnQiIHBlcmhhcHMgYXJlDQo+IGZ1 bmRhbWVudGFsIGZvciB0aGUgdXBsYXllciBzZXJ2aWNlcy4NCj4gPj4gSXQgc2hvdWxkIGJlIHBy ZXZlbnRlZCB0aGF0IGh1bWFuIG9wZXJhdG9yIHVzZXMgc29tZSBDTElzIHRvIGNoYW5nZQ0KPiB0 aGVtIGJ5IG1pc3Rha2UsIGFuZCB0aGUgdXBsYXllciBzZXJ2aWNlcyBhcmUgaW50ZXJydXB0ZWQu DQo+ID4+IEF0IGxlYXN0LCBzb21lIHdhcm5pbmcgc2hvdWxkIGJlIHJlcG9ydGVkIHRvIHRoZSBo dW1hbiBvcGVyYXRvci4NCj4gPj4gW2R1em9uZ3BlbmddDQo+ID4+DQo+ID4+IEludGVudCBjb3Vs ZCBwZXJmZWN0bHkgd2VsbCBjb3ZlciBhIGhpZ2ggbGV2ZWwgcG9saWN5IHN1Y2ggYXMgImFsbCBu b2RlcyBvZg0KPiB0eXBlIHggZG8gdGhpczsgdHlwZSB5IGRvZXMgdGhhdCIuIChXaGljaCBJIHRo aW5rIGlzIHdoYXQgeW91IG1lbnRpb24gYmVsb3cuKQ0KPiBXaGVuIHdlIHN0YXJ0ZWQgd2l0aCBJ bnRlbnQsIHRoZSB2aXNpb24gd2FzIHRvIGtlZXAgaXQgdmVyeSBoaWdoIGxldmVsLCBtb3JlDQo+ IGFsb25nIHRoZSBsaW5lICJkbyB0aGUgcmlnaHQgdGhpbmciIDopICBJdCBoYXMgYmVjb21lIGNs ZWFyIHRoYXQgaW4gdG9kYXkncw0KPiBuZXR3b3JraW5nLCBwZW9wbGUgd2lsbCB3YW50IHRvIHB1 c2ggbW9yZSAicGFyYW1ldGVycyIgdGhhbiAicG9saWN5IiwgYW5kIEkNCj4gdGhpbmsgd2UnbGwg anVzdCBoYXZlIHRvIGFja25vd2xlZGdlIHRoYXQuDQo+ID4+DQo+ID4+IFBlcnNvbmFsbHkgSSB0 aGluayBpdCdsbCBwYW4gb3V0IGxpa2UgdGhpczogSW50ZW50IHdpbGwgaGF2ZSBzZWN0aW9ucyBw ZXINCj4gYXV0b25vbWljIGZ1bmN0aW9uLiBBbiBJbnRlbnQgcGFyc2VyIHdpbGwgdmFsaWRhdGUg b3ZlcmFsbCBjb3JyZWN0bmVzcw0KPiAoc2lnbmF0dXJlLCBldGMpLCB0aGVuIHNlZ21lbnQgdGhl IEludGVudCBpbnRvIGl0cyBiaXRzOiBPbmUgcGFydCB3aWxsIGJlIHRoZQ0KPiBhdXRvbm9taWMg ZmFicmljIGl0c2VsZiwgdGhlbiB0aGVyZSB3aWxsIGJlIGJpdHMgZm9yIGVhY2ggYXV0b25vbWlj IGZ1bmN0aW9uLg0KPiA+Pg0KPiA+PiBTbyB0aGUgSW50ZW50IHBhcnNlciBvbiBhIG5vZGUgd2ls bCBncmFiIGludGVudCwgc2VlIHRoZXJlIGlzIHNvbWUgSW50ZW50DQo+IGZvciBhdXRvbm9taWMg ZnVuY3Rpb24gQSwgd2lsbCBzZWUgd2hldGhlciB0aGlzIGlzIHJ1bm5pbmcgbG9jYWxseSwgYW5k IGlmIHNvLCBpdCdsbA0KPiBqdXN0IHBhc3MgdGhhdCBwaWVjZSBvZiBJbnRlbnQgdG8gdGhhdCBm dW5jdGlvbi4NCj4gPj4NCj4gPj4gSW4gdGhpcyB3YXksIGEgZnVuY3Rpb24gb3duZXIgZGVmaW5l cyBib3RoIHRoZSBpbnRlbnQgaXRzZWxmLCBhbmQgaG93IGl0IGlzDQo+IGNvbnN1bWVkIGJ5IGhp cy9oZXIgZnVuY3Rpb24uDQo+ID4+DQo+ID4+IFtkdXpvbmdwZW5nXUFncmVlLltkdXpvbmdwZW5n XQ0KPiA+Pg0KPiA+PiBUaGUgZG93bnNpZGUgaXMgdGhhdCB0aGVyZSBpcyBubyAiY29udHJvbCIg YW4vL2QgcGVvcGxlIGNhbiBkbyB3aGF0IHRoZXkNCj4gd2FudCwgaW4gdGhlIHdvcnN0IGNhc2Ug ZXZlbiBwdXNoIGNvbmZpZywgd2hpY2ggaXMgY2xlYXJseSBOT1Qgd2hhdCBJbnRlbnQgaXMNCj4g bWVhbnQgdG8gZG8uIEknZCBob3BlIHRoYXQgcGVvcGxlIGRvbid0IHN0YXJ0IHdyaXRpbmcgSW50 ZW50IGxpa2UgIm5vZGUgMTc6DQo+IGhlcmUgaXMgeW91ciBDTEk7IG5vZGUgMjM6IGhlcmUgaXMg eW91ciBDTEkiOyBub2RlIDM3OiBoZXJlIGlzIHlvdXIgQ0xJIi4gQnV0DQo+IHdoYXQgSSBkZXNj cmliZSBhYm92ZSBDT1VMRCBiZSBhYnVzZWQgaW4gdGhpcyB3YXkuIFJGQzc1NzUgc2F5cyB0aGlz IGlzIE5PVA0KPiBJbnRlbnQuDQo+ID4+DQo+ID4+IFtkdXpvbmdwZW5nXUFncmVlLiBJdCBpcyBo YXJkIHRvIGdpdmUgYW4gZXhhY3QgYm91bmRhcnkuDQo+ID4+IEhvd2V2ZXIsIEkgdGhpbmsgd2Ug aGF2ZSB0d28gY2hvaWNlcyBoZXJlLg0KPiA+PiBUaGUgZmlyc3Qgb25lIGlzIHRvIGRlZmluZSBh IHBvbGljeSBmaXJzdCBpbiBldmVyeSBBU0EsIGJ1dCB3aXRoIHNvbWUgb3Blbg0KPiBwYXJhbWV0 ZXJzIHRoYXQgY2FuIGJlIGNvbmZpZ3VyZWQuIFdoYXQgd2UgbmVlZCB0byBkbyBpcyBqdXN0IHRv IGRpc3RyaWJ1dGUNCj4gdGhlIG5ldHdvcmstbGV2ZWwgcGFyYW1ldGVycyBmb3IgdGhhdCBmdW5j dGlvbi4gVGhpcyB3YXkgaXMgc3VpdGFibGUgdG8gc29tZQ0KPiByZWxhdGl2ZWx5IHN0YXRpYyBm dW5jdGlvbnMuDQo+ID4+IFRoZSBzZWNvbmQgaXMgbW9yZSBhZHZhbmNlZCwgaW4gdGhlIGlkZWFs IGNhc2UsIHRoZSBodW1hbiBvcGVyYXRvciBjYW4NCj4gY29tbXVuaWNhdGUgd2l0aCBuZXR3b3Jr IHVzaW5nIG5hdHVyYWwgbGFuZ3VhZ2Ugb3Igc29tZXRoaW5nIHNpbWlsYXIuIFRoaXMNCj4gd2F5 IGlzIG1vcmUgZmxleGlibGUsIGJ1dCBoYXJkZXIgdG8gcmVhbGl6ZS4NCj4gPj4gW2R1em9uZ3Bl bmddDQo+ID4+DQo+ID4+IEluIG90aGVyIHdvcmRzOiBMZXQncyBub3QgYmUgdG9vIHByZXNjcmlw dGl2ZSwgYW5kIGhvcGUgdGhhdCBmb2xrcyB3aWxsIGRvDQo+IG1vcmUgb3IgbGVzcyByZWFzb25h YmxlIHRoaW5ncy4NCj4gPj4NCj4gPj4gW2R1em9uZ3BlbmddQWdyZWUuW2R1em9uZ3BlbmddDQo+ ID4+DQo+ID4+IEkndmUgYmVlbiBtZWFuaW5nIHRvIGNvbW1lbnQgb24gdGhlIEludGVudCBkcmFm dHMgYW5kIHBvc3NpYmx5DQo+IGNvbnRyaWJ1dGUsIEknbSBqdXN0IHN0cnVnZ2xpbmcgd2l0aCBw cmlvcml0aWVzIGF0IHRoZSBtb21lbnQgKGFzIHlvdSBjYW4gc2VlDQo+IGJ5IG15IHJlbGF0aXZl IHNpbGVuY2Ugb3ZlciB0aGUgcGFzdCB3ZWVrcykuDQo+ID4+DQo+ID4+IE1pY2hhZWwNCj4gPj4N Cj4gPj4NCj4gPj4NCj4gPj4gRnJvbTogQW5pbWEgW21haWx0bzphbmltYS1ib3VuY2VzQGlldGYu b3JnXSBPbiBCZWhhbGYgT2YgRHV6b25ncGVuZw0KPiA+PiBTZW50OiAxNiBNYXJjaCAyMDE2IDAy OjM4DQo+ID4+IFRvOiBhbmltYUBpZXRmLm9yZzxtYWlsdG86YW5pbWFAaWV0Zi5vcmc+DQo+ID4+ IENjOiBMYXVyZW50IENpYXZhZ2xpYQ0KPiA+PiA8bGF1cmVudC5jaWF2YWdsaWFAbm9raWEuY29t PG1haWx0bzpsYXVyZW50LmNpYXZhZ2xpYUBub2tpYS5jb20+PjsNCj4gPj4gSsOpZmVyc29uIENh bXBvcyBOb2JyZQ0KPiA+PiA8amNub2JyZUBpbmYudWZyZ3MuYnI8bWFpbHRvOmpjbm9icmVAaW5m LnVmcmdzLmJyPj47IFNoZW5nIEppYW5nDQo+ID4+IDxqaWFuZ3NoZW5nQGh1YXdlaS5jb208bWFp bHRvOmppYW5nc2hlbmdAaHVhd2VpLmNvbT4+DQo+ID4+IFN1YmplY3Q6IFtBbmltYV0gRG9lcyBB dXRvbm9taWMgbmV0d29yayBuZWVkIHNvbWUgaW50ZXJ2ZW50aW9uDQo+IGJleW9uZA0KPiA+PiBp bnRlbnQNCj4gPj4NCj4gPj4gSGVsbG8sIGV2ZXJ5b25lDQo+ID4+DQo+ID4+ICAgICAgICAgIEkg dW5kZXJzdGFuZCB0aGF0IEF1dG9ub21pYyBuZXR3b3JrIHNob3VsZCBiZSBhYmxlIHRvIHdvcmsg d2l0aCBhcw0KPiBsZXNzIGNvbmZpZ3VyYXRpb24gYXMgcG9zc2libGUuIEluIHRoZSBpZGVhbCBj YXNlLCBhbiBBdXRvbm9taWMgbmV0d29yaw0KPiBzaG91bGQgYmUgYWJsZSB0byB3b3JrIHdlbGwg d2hpbGUgb25seSBpbnRlbnQgaXMgbmVlZGVkIGZyb20gaHVtYW4NCj4gb3BlcmF0b3JzLg0KPiA+ Pg0KPiA+PiAgICAgICAgICBIb3dldmVyLCBJIHdvbmRlciB3aGV0aGVyIEF1dG9ub21pYyBuZXR3 b3JrIG1heSBuZWVkIHNvbWUNCj4gaW50ZXJ2ZW50aW9uIGZyb20gaHVtYW4gb3BlcmF0b3JzIHRo cm91Z2ggc29tZSBsb3ctbGV2ZWwgcG9saWNpZXMgYmV5b25kDQo+IGludGVudHMuDQo+ID4+DQo+ ID4+ICAgICAgICAgIEZvciBleGFtcGxlLCBpbiB0aGUgaHR0cHM6Ly90b29scy5pZXRmLm9yZy9o dG1sL2RyYWZ0LWlldGYtYW5pbWEtDQo+IHByZWZpeC1tYW5hZ2VtZW50LTAwLCBpdCBpcyBzdWdn ZXN0ZWQgdGhhdCB0aGUgcHJlZml4IGxlbmd0aHMgZm9yIHRoZSBDU0csDQo+IEFTRywgUlNHIChk aWZmZXJlbnQgcm9sZXMgaW4gSVAgUkFOKSBjYW4gYmUgYXNzaWduZWQgYXMgYW4gImludGVudCIu IFRoZXNlDQo+IGNvbmZpZ3VyYXRpb24gcGFyYW1ldGVycyBuZWVkIHRvIGJlIGRpc3RyaWJ1dGVk IGluIHRoZSBhdXRvbm9taWMgZG9tYWluIHRvDQo+IGluZmx1ZW5jZSB0aGUgZGV0YWlsIGNvbmZp Z3VyYXRpb25zIG9uIGVhY2ggYXV0b25vbWljIG5vZGUuDQo+ID4+DQo+ID4+ICAgICAgICAgIEFz IGludGVudCBpcyBkZXNjcmliZWQgYXMgYW4gYWJzdHJhY3QsIGRlY2xhcmF0aXZlLCBoaWdoLWxl dmVsIHBvbGljeQ0KPiB1c2VkIHRvIG9wZXJhdGUgYW4gYXV0b25vbWljIGRvbWFpbiwgc3VjaCBh cyBhbiBlbnRlcnByaXNlIG5ldHdvcmsuDQo+IFBlcmhhcHMgd2UgY2FuIHJlbmFtZSB0aGUgImlu dGVudCIgaW4gdGhlICJwcmVmaXgtbWFuYWdlbWVudCBJRCIgdG8gc29tZQ0KPiBvdGhlciB3b3Jk cy4NCj4gPj4NCj4gPj4gSXMgIkFTQSBwYXJhbWV0ZXJzIiBvayBoZXJlPw0KPiA+Pg0KPiA+PiBJ IHRoaW5rIHNvbWUgY2hhcmFjdGVyaXN0aWNzIGZvciB0aGlzIGtpbmQgb2YgbG93LWxldmVsIHBv bGljaWVzIGFyZQ0KPiA+Pg0KPiA+PiAxLiAgICAgICBOZXR3b3JrIGxldmVsIHBhcmFtZXRlcnMs IG5lZWQgdG8gYmUgZGlzdHJpYnV0ZWQgYnkgR1JBU1AgYW5kDQo+IEFDUA0KPiA+Pg0KPiA+PiAy LiAgICAgICBSZWxhdGl2ZWx5IHN0YXRpYyBjb21wYXJlZCB0byBpbnRlbnQsIHBlcmhhcHMgb25s eSBuZWVkZWQgdG8gYmUNCj4gY29uZmlndXJlZCBvbmNlDQo+ID4+DQo+ID4+IDMuICAgICAgIFVu YXZvaWRhYmxlIGluIGN1cnJlbnQgbmV0d29yayBlbnZpcm9ubWVudCwgbW9zdGx5IGZvcg0KPiBl c3RhYmxpc2hpbmcgdGhlIG5ldHdvcmsgaW5mcmFzdHJ1Y3R1cmUNCj4gPj4NCj4gPj4gNC4gICAg ICAgUmFyZWx5IG5lZWQgY29vcmRpbmF0aW9ucyB3aXRoIG90aGVycyBwYXJhbWV0ZXJzLCB1c3Vh bGx5IGlzIGNsb3NlZA0KPiByZWxhdGVkIHRvIG9uZSBzcGVjaWZpYyBhdXRvbm9taWMgZnVuY3Rp b24NCj4gPj4NCj4gPj4gV2VsY29tZSBmb3IgY29tbWVudC4gWW91ciBzdWdnZXN0aW9ucyB3aWxs IGJlIGFwcHJlY2lhdGVkLg0KPiA+Pg0KPiA+Pg0KPiA+PiBCZXN0IFJlZ2FyZHMNCj4gPj4gWm9u Z3BlbmcgRHUNCj4gPj4NCj4gPj4NCj4gPj4NCj4gPj4gX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX18NCj4gPj4gQW5pbWEgbWFpbGluZyBsaXN0DQo+ID4+IEFu aW1hQGlldGYub3JnDQo+ID4+IGh0dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8v YW5pbWENCj4gPj4NCj4gPg0KDQo= From nobody Mon Mar 21 02:47:00 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E24CB12D705 for ; Mon, 21 Mar 2016 02:46:57 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.222 X-Spam-Level: X-Spam-Status: No, score=-4.222 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IOQBVstG0ktQ for ; Mon, 21 Mar 2016 02:46:54 -0700 (PDT) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 98A7512D6FE for ; Mon, 21 Mar 2016 02:46:53 -0700 (PDT) Received: from 172.18.7.190 (EHLO lhreml707-cah.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CLB06783; Mon, 21 Mar 2016 09:46:51 +0000 (GMT) Received: from NKGEML402-HUB.china.huawei.com (10.98.56.33) by lhreml707-cah.china.huawei.com (10.201.5.199) with Microsoft SMTP Server (TLS) id 14.3.235.1; Mon, 21 Mar 2016 09:46:50 +0000 Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml402-hub.china.huawei.com ([10.98.56.33]) with mapi id 14.03.0235.001; Mon, 21 Mar 2016 17:46:42 +0800 From: "Liubing (Leo)" To: Brian E Carpenter , Duzongpeng , "anima@ietf.org" Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAAE3aEAABGmDAABfNuwAAGs2jQAALi06gA== Date: Mon, 21 Mar 2016 09:46:41 +0000 Message-ID: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C997@nkgeml514-mbx.china.huawei.com> References: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> <56E9A8B2.8070504@gmail.com> <56EEF72A.5050309@gmail.com> In-Reply-To: <56EEF72A.5050309@gmail.com> Accept-Language: en-US, zh-CN Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.98.117] Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020202.56EFC30C.005B, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: 459087b6c3337b62f73452de91af5c9c Archived-At: Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "Michael Behringer \(mbehring\)" , Sheng Jiang Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 09:46:58 -0000 SGkgQnJpYW4gYW5kIGFsbCwNCg0KPiAtLS0tLU9yaWdpbmFsIE1lc3NhZ2UtLS0tLQ0KPiBGcm9t OiBBbmltYSBbbWFpbHRvOmFuaW1hLWJvdW5jZXNAaWV0Zi5vcmddIE9uIEJlaGFsZiBPZiBCcmlh biBFDQo+IENhcnBlbnRlcg0KPiBTZW50OiBNb25kYXksIE1hcmNoIDIxLCAyMDE2IDM6MTcgQU0N Cj4gVG86IER1em9uZ3Blbmc7IGFuaW1hQGlldGYub3JnDQo+IENjOiBMYXVyZW50IENpYXZhZ2xp YTsgSsOpZmVyc29uIENhbXBvcyBOb2JyZTsgTWljaGFlbCBCZWhyaW5nZXINCj4gKG1iZWhyaW5n KTsgU2hlbmcgSmlhbmcNCj4gU3ViamVjdDogUmU6IFtBbmltYV0gRG9lcyBBdXRvbm9taWMgbmV0 d29yayBuZWVkIHNvbWUgaW50ZXJ2ZW50aW9uDQo+IGJleW9uZCBpbnRlbnQNCj4gDQo+IEhpIFpv bmdwZW5nLA0KPiANCj4gPiBTaG91bGQgdGhlc2UgQVNBIHBhcmFtZXRlcnMgYmUgcGFydCBvZiBB TklNQSdzIHdvcmsgaXRlbXMganVzdCBhcyBpbnRlbnQ/DQo+IA0KPiBBcyBNaWNoYWVsJ3MgeWFu ZyBvdXRsaW5lIHNob3dlZCwgdGhlcmUgd2lsbCBiZSBuZXR3b3JrLXdpZGUgYXNwZWN0cyB0bw0K PiBpbnRlbnQgYW5kIGFsc28gQVNBLXNwZWNpZmljIGFzcGVjdHMuIEkgdGhpbmsgdGhhdCBpbiBw cmFjdGljZSB0aGVyZSB3aWxsIGJlDQo+IHNpbXBsZSBwYXJhbWV0ZXJzIGluIHRoZSBBU0Etc3Bl Y2lmaWMgcGFydHMsIHNvICh3aGV0aGVyIHdlIHVzZSB5YW5nIG9yDQo+IHNvbWUgb3RoZXIgc3lu dGF4KSBJIHNlZSBubyBhZHZhbnRhZ2UgaW4gaGF2aW5nIGEgc2VwYXJhdGUgbWVjaGFuaXNtIGZv cg0KPiB0aGUgcGFyYW1ldGVycyAtIGluIGEgc2Vuc2Ugd2UgbmV2ZXIgbmVlZCB0byBhbnN3ZXIg dGhlIHF1ZXN0aW9uLCB3ZSBzaW1wbHkNCj4gdHJlYXQgYWxsIHRoZSBpbmZvcm1hdGlvbiBpbiB0 aGUgc2FtZSB3YXksIGludGVudCBvciBwYXJhbWV0ZXIuDQo+IA0KPiBNYXliZSB3ZSBzaG91bGQg dGhpbmsgYWJvdXQgbGV2ZWxzIG9mIHNwZWNpYWxpemF0aW9uLiBUb3AgbGV2ZWwsIG5ldHdvcmsN Cj4gaW50ZW50LiBOZXh0IGxldmVsLCBpbnRlbnQgZm9yIGFsbCBBU0FzIHN1cHBvcnRpbmcgYSBn aXZlbiBvYmplY3RpdmUuDQo+IFRob3NlIHR3byBsZXZlbHMgd291bGQgYmUgZmxvb2RlZCBieSBH UkFTUC4gDQoNCltCaW5nXSBGb3IgIkFTQXMgc3VwcG9ydGluZyBhIGdpdmVuIG9iamVjdGl2ZSIs IGZsb29kaW5nIHN1cmVseSB3b3VsZCB3b3JrLiBCdXQgSSdtIHdvbmRlcmluZyB3aGV0aGVyIHdl IGFsc28gbmVlZCBhIHNlbGVjdGl2ZSBmbG9vZGluZyBtZWNoYW5pc20gdG8gY3V0IHNvbWUgdW5u ZWNlc3Nhcnkgc2lnbmFsaW5nIGFtcGxpZmljYXRpb24uDQpXaGVuIG9uZSBub2RlIHJlY2VpdmUg dGhlIGludGVudCwgaXQgb25seSByZXBsaWNhdGVzIHRoZSBpbnRlbnQgdG8gdGhlIG5laWdoYm9y cyB0aGF0IGZpdCBmb3Igc29tZSBjb25kaXRpb25zLiBUaGlzIG5vZGUgYmVoYXZpb3IgaW1wbGll cyBzb21lIGFzc3VtcHRpb246DQoxKSBUaGUgaW50ZW50IGNvbnRlbnQgaW5jbHVkZXMgdGhlICJj b25kaXRpb25zIiAoZS5nLiwgZmxvb2QgdG8gYSBzcGVjaWZpYyB0dW5uZWw7IG9yIGZsb29kIHRv IGEgY2VydGFpbiBzZXQgb2YgcnVsZXMgZXRjLikNCjIpIFRoZSBub2RlcyBrbm93IGFib3V0IHRo ZWlyIG5laWdoYm9ycycgaW5mbyB0byBpZGVudGlmeSB3aGV0aGVyIHRoZSBuZWlnaGJvcnMgZml0 IHRoZSBjb25kaXRpb25zLg0KDQpEbyB5b3UgY29uc2lkZXIgdGhpcyBhcyBhIGNvbXBsZW1lbnRh cnkgbWVjaGFuaXNtIHRvIG5vcm1hbCBmbG9vZGluZz8NCg0KQmVzdCByZWdhcmRzLA0KQmluZw0K DQo+IE5leHQgbGV2ZWwsIGludGVudCBmb3IgYSBzaW5nbGUNCj4gQVNBLiBUaGF0IHdvdWxkIGJl IHRyYW5zbWl0dGVkIGJ5IHVuaWNhc3QgR1JBU1Agc3luY2hyb25pemF0aW9uLg0KPiANCj4gUmVn YXJkcw0KPiAgICBCcmlhbg0KPiANCj4gT24gMTgvMDMvMjAxNiAyMTozMywgRHV6b25ncGVuZyB3 cm90ZToNCj4gPiBIaSwgQnJpYW4NCj4gPg0KPiA+IAlUaGFua3MgZm9yIHRoZSByZXNwb25zZS4g SSB0aGluayB3ZSBoYXZlIGFncmVlZCB0aGF0DQo+ICJwcmVmaXgtbWFuYWdlbWVudCBJRCIgY2Fu IGJlIGFuIEF1dG9ub21pYyBpbnRlbnQuDQo+ID4NCj4gPiAJSG93ZXZlciwgc29tZSBBU0EgcGFy YW1ldGVycyBtYXkgYXBwZWFyIHRvIGJlIG1vcmUgc3RhdGljLCBhbmQNCj4gcGVyaGFwcyBjYW4g ZnVsZmlsbCB0aGUgc3RhdGVtZW50IGJlbG93Og0KPiA+IAkxLiBOZXR3b3JrIGxldmVsIHBhcmFt ZXRlcnMsIG5lZWQgdG8gYmUgZGlzdHJpYnV0ZWQgYnkgR1JBU1AgYW5kIEFDUA0KPiA+IAkyLiBS ZWxhdGl2ZWx5IHN0YXRpYyBjb21wYXJlZCB0byBpbnRlbnQsIHBlcmhhcHMgb25seSBuZWVkZWQg dG8gYmUNCj4gY29uZmlndXJlZCBvbmNlDQo+ID4gCTMuIFVuYXZvaWRhYmxlIGluIGN1cnJlbnQg bmV0d29yayBlbnZpcm9ubWVudCwgbW9zdGx5IGZvciBlc3RhYmxpc2hpbmcNCj4gdGhlIG5ldHdv cmsgaW5mcmFzdHJ1Y3R1cmUNCj4gPiAJNC4gUmFyZWx5IG5lZWQgY29vcmRpbmF0aW9ucyB3aXRo IG90aGVycyBwYXJhbWV0ZXJzLCB1c3VhbGx5IGlzDQo+ID4gY2xvc2VkIHJlbGF0ZWQgdG8gb25l IHNwZWNpZmljIGF1dG9ub21pYyBmdW5jdGlvbg0KPiA+DQo+ID4gCVNvbWUgZXhhbXBsZXMgY2Fu IGJlDQo+ID4gCVdoYXQga2luZCBvZiBJR1AgdGhlIEF1dG9ub21pYyBuZXR3b3JrIGRvbWFpbiBz aG91bGQgdXNlIChzdWNoIGFzDQo+IElTLUlTLCBPU1BGKTsNCj4gPiAJV2hhdCBraW5kIG9mIHRy YW5zcG9ydCBsYXllciBuZXR3b3JrIHRlY2hub2xvZ3kgdGhlIEF1dG9ub21pYw0KPiBuZXR3b3Jr DQo+ID4gZG9tYWluIHNob3VsZCB1c2UgKHN1Y2ggYXMgTVBMUywgSVApDQo+ID4NCj4gPiAJU2hv dWxkIHRoZXNlIEFTQSBwYXJhbWV0ZXJzIGJlIHBhcnQgb2YgQU5JTUEncyB3b3JrIGl0ZW1zIGp1 c3QgYXMNCj4gaW50ZW50Pw0KPiA+DQo+ID4gQmVzdCBSZWdhcmRzDQo+ID4gWm9uZ3BlbmcgRHUN Cj4gPg0KPiA+IC0tLS0tT3JpZ2luYWwgTWVzc2FnZS0tLS0tDQo+ID4gRnJvbTogQnJpYW4gRSBD YXJwZW50ZXIgW21haWx0bzpicmlhbi5lLmNhcnBlbnRlckBnbWFpbC5jb21dDQo+ID4gU2VudDog VGh1cnNkYXksIE1hcmNoIDE3LCAyMDE2IDI6NDEgQU0NCj4gPiBUbzogRHV6b25ncGVuZzsgTWlj aGFlbCBCZWhyaW5nZXIgKG1iZWhyaW5nKTsgYW5pbWFAaWV0Zi5vcmcNCj4gPiBDYzogTGF1cmVu dCBDaWF2YWdsaWE7IErDqWZlcnNvbiBDYW1wb3MgTm9icmU7IFNoZW5nIEppYW5nDQo+ID4gU3Vi amVjdDogUmU6IFtBbmltYV0gRG9lcyBBdXRvbm9taWMgbmV0d29yayBuZWVkIHNvbWUgaW50ZXJ2 ZW50aW9uDQo+ID4gYmV5b25kIGludGVudA0KPiA+DQo+ID4gWWVzLCBJIHRoaW5rIHRoYXQgd2hh dGV2ZXIgdGhlICJwaGlsb3NvcGh5IiBvZiBpbnRlbnQgaXMsIHdlIHNob3VsZCBtYWtlDQo+IHN1 cmUgdGhhdCB0aGUgaW1wbGVtZW50YXRpb24gaXMgcHJhZ21hdGljIGFuZCB0aGF0IHdlIGNhbiBh ZGFwdCB0bw0KPiBvcGVyYXRvcidzIHJlcXVpcmVtZW50cy4NCj4gPg0KPiA+IFJlZ2FyZHMNCj4g PiAgICBCcmlhbg0KPiA+DQo+ID4gT24gMTcvMDMvMjAxNiAwMTo0MSwgRHV6b25ncGVuZyB3cm90 ZToNCj4gPj4gSGksIE1pY2hhZWwsDQo+ID4+DQo+ID4+ICAgICAgICAgIFRoYW5rcyBmb3IgdGhl IHJlcGx5LiBJIGNhbiB1bmRlcnN0YW5kIHRoYXQgaW50ZW50IGlzIG5vdCBpbiB0aGUNCj4gY3Vy cmVudCBtaWxlc3RvbmVzLCBhbmQgeW91IGFyZSB3b3JraW5nIG9uIHNldmVyYWwgbWlsZXN0b25l cyBkcmFmdHMgbm93Lg0KPiA+Pg0KPiA+PiAgICAgICAgICBZb3UgYXJlIHdlbGNvbWUgdG8gY29u dHJpYnV0ZSB0byB0aGUgaW50ZW50IGRyYWZ0IGF0IGFueSB0aW1lLA0KPiBhbmQgdGhhbmtzIGlu IGFkdmFuY2UuDQo+ID4+DQo+ID4+ICAgICAgICAgIEFjY29yZGluZyB0byB5b3VyIHN0YXRlbWVu dHMsIGluIG15IHVuZGVyc3RhbmRpbmcsIHdlIGNhbg0KPiBrZWVwIHJlZ2FyZGluZyB0aGUgImlu dGVudCIgZXhhbXBsZSBpbiAicHJlZml4LW1hbmFnZW1lbnQgSUQiIGFzIGFuDQo+IEF1dG9ub21p YyBpbnRlbnQuDQo+ID4+ICAgICAgICAgIEl0J3Mgc2VtYW50aWNzIGFyZSBqdXN0IGxpa2UgImFs bCBub2RlcyBvZiB0eXBlIHggZG8gdGhpczsgdHlwZSB5DQo+IGRvZXMgdGhhdCIsIHNvIGl0IGlz IGEgaGlnaCBsZXZlbCBwb2xpY3kuDQo+ID4+DQo+ID4+ICAgICAgICAgIEl0IHJlYWxseSBoZWxw cyBhIGxvdC4gVGhhbmtzLg0KPiA+Pg0KPiA+PiAgICAgICAgICBJIGdlbmVyYWxseSBhZ3JlZSB3 aXRoIHlvdXIgc3RhdGVtZW50LCB3aGlsZSBzb21lIG90aGVyDQo+IHBlcnNvbmFsIHVuZGVyc3Rh bmRpbmdzIGV4aXN0IChzaG93biBiZWxvdykuIEJ1dCBwZXJoYXBzIHRoZXkgY2FuIGJlDQo+IGRp c2N1c3NlZCBsYXRlci4NCj4gPj4NCj4gPj4gQmVzdCBSZWdhcmRzDQo+ID4+IFpvbmdwZW5nIER1 DQo+ID4+DQo+ID4+IEZyb206IE1pY2hhZWwgQmVocmluZ2VyIChtYmVocmluZykgW21haWx0bzpt YmVocmluZ0BjaXNjby5jb21dDQo+ID4+IFNlbnQ6IFdlZG5lc2RheSwgTWFyY2ggMTYsIDIwMTYg NDoxNyBQTQ0KPiA+PiBUbzogRHV6b25ncGVuZzsgYW5pbWFAaWV0Zi5vcmcNCj4gPj4gQ2M6IExh dXJlbnQgQ2lhdmFnbGlhOyBKw6lmZXJzb24gQ2FtcG9zIE5vYnJlOyBTaGVuZyBKaWFuZw0KPiA+ PiBTdWJqZWN0OiBSRTogW0FuaW1hXSBEb2VzIEF1dG9ub21pYyBuZXR3b3JrIG5lZWQgc29tZSBp bnRlcnZlbnRpb24NCj4gPj4gYmV5b25kIGludGVudA0KPiA+Pg0KPiA+PiBIaSBab25ncGVuZywN Cj4gPj4NCj4gPj4gWWVzLCB0aGlzIGlzIGEgcmVjdXJyaW5nIHF1ZXN0aW9uIGFuZCB0b3BpYy4g TGV0IG1lIGdpdmUgbXkgdmlldyBvbiB0d28NCj4gbGV2ZWxzOg0KPiA+Pg0KPiA+PiBGaXJzdCwg QU5JTUEgaXNuJ3QgdGhlIG9ubHkgdGhpbmcgImluZmx1ZW5jaW5nIiBvciAic3RlZXJpbmciIGEg bmV0d29yaywgYXQNCj4gbGVhc3Qgbm90IGluaXRpYWxseS4gVGhlcmUgaXMgc3RpbGwgdHJhZGl0 aW9uYWwgY29uZmlnLCBwbHVzIGFsbCB0aGUgZXhpc3RpbmcgYW5kDQo+IGVtZXJnaW5nIFNETiBt b2RlbHMuIE15IHZpZXcgaGFzIGFsd2F5cyBiZWVuIHRoYXQgQU5JTUEgc2V0cyBhICJkZWZhdWx0 Ig0KPiBwb2xpY3kgaW4gYSBuZXR3b3JrLCBhbmQgYWxsIG90aGVyIG1lYW5zIGFyZSBtb3JlIHNw ZWNpZmljIGFuZCBjYW4gb3ZlcnJpZGUNCj4gaW50ZW50LiBUaGlzIGlzIGRlc2NyaWJlZCBpbiBS RkM3NTc1LiBNYWluIHBvaW50IGhlcmU6IEFOSU1BIEludGVudCBpc24ndCB0aGUNCj4gb25seSB0 b29sIGluIGEgbmV0d29yay4NCj4gPj4NCj4gPj4gW2R1em9uZ3BlbmddIEkgYWdyZWUgdGhhdCBp bnRlbnQgY2FuIGNvZXhpc3Qgd2l0aCB0cmFkaXRpb25hbCBjb25maWcuDQo+ID4+IEkgZ2VuZXJh bGx5IGFncmVlIHdpdGggdGhlIHBvbGljeSBwcmlvcml0aWVzLCBidXQgSSBoYXZlIHNvbWUgb3Ro ZXINCj4gY29uY2VybnMuDQo+ID4+IFNvbWUgZGVmYXVsdCBwb2xpY2llcy9wYXJhbWV0ZXJzIHNl dCB0aHJvdWdoICJpbnRlbnQiIHBlcmhhcHMgYXJlDQo+IGZ1bmRhbWVudGFsIGZvciB0aGUgdXBs YXllciBzZXJ2aWNlcy4NCj4gPj4gSXQgc2hvdWxkIGJlIHByZXZlbnRlZCB0aGF0IGh1bWFuIG9w ZXJhdG9yIHVzZXMgc29tZSBDTElzIHRvIGNoYW5nZQ0KPiB0aGVtIGJ5IG1pc3Rha2UsIGFuZCB0 aGUgdXBsYXllciBzZXJ2aWNlcyBhcmUgaW50ZXJydXB0ZWQuDQo+ID4+IEF0IGxlYXN0LCBzb21l IHdhcm5pbmcgc2hvdWxkIGJlIHJlcG9ydGVkIHRvIHRoZSBodW1hbiBvcGVyYXRvci4NCj4gPj4g W2R1em9uZ3BlbmddDQo+ID4+DQo+ID4+IEludGVudCBjb3VsZCBwZXJmZWN0bHkgd2VsbCBjb3Zl ciBhIGhpZ2ggbGV2ZWwgcG9saWN5IHN1Y2ggYXMgImFsbCBub2RlcyBvZg0KPiB0eXBlIHggZG8g dGhpczsgdHlwZSB5IGRvZXMgdGhhdCIuIChXaGljaCBJIHRoaW5rIGlzIHdoYXQgeW91IG1lbnRp b24gYmVsb3cuKQ0KPiBXaGVuIHdlIHN0YXJ0ZWQgd2l0aCBJbnRlbnQsIHRoZSB2aXNpb24gd2Fz IHRvIGtlZXAgaXQgdmVyeSBoaWdoIGxldmVsLCBtb3JlDQo+IGFsb25nIHRoZSBsaW5lICJkbyB0 aGUgcmlnaHQgdGhpbmciIDopICBJdCBoYXMgYmVjb21lIGNsZWFyIHRoYXQgaW4gdG9kYXkncw0K PiBuZXR3b3JraW5nLCBwZW9wbGUgd2lsbCB3YW50IHRvIHB1c2ggbW9yZSAicGFyYW1ldGVycyIg dGhhbiAicG9saWN5IiwgYW5kIEkNCj4gdGhpbmsgd2UnbGwganVzdCBoYXZlIHRvIGFja25vd2xl ZGdlIHRoYXQuDQo+ID4+DQo+ID4+IFBlcnNvbmFsbHkgSSB0aGluayBpdCdsbCBwYW4gb3V0IGxp a2UgdGhpczogSW50ZW50IHdpbGwgaGF2ZSBzZWN0aW9ucyBwZXINCj4gYXV0b25vbWljIGZ1bmN0 aW9uLiBBbiBJbnRlbnQgcGFyc2VyIHdpbGwgdmFsaWRhdGUgb3ZlcmFsbCBjb3JyZWN0bmVzcw0K PiAoc2lnbmF0dXJlLCBldGMpLCB0aGVuIHNlZ21lbnQgdGhlIEludGVudCBpbnRvIGl0cyBiaXRz OiBPbmUgcGFydCB3aWxsIGJlIHRoZQ0KPiBhdXRvbm9taWMgZmFicmljIGl0c2VsZiwgdGhlbiB0 aGVyZSB3aWxsIGJlIGJpdHMgZm9yIGVhY2ggYXV0b25vbWljIGZ1bmN0aW9uLg0KPiA+Pg0KPiA+ PiBTbyB0aGUgSW50ZW50IHBhcnNlciBvbiBhIG5vZGUgd2lsbCBncmFiIGludGVudCwgc2VlIHRo ZXJlIGlzIHNvbWUgSW50ZW50DQo+IGZvciBhdXRvbm9taWMgZnVuY3Rpb24gQSwgd2lsbCBzZWUg d2hldGhlciB0aGlzIGlzIHJ1bm5pbmcgbG9jYWxseSwgYW5kIGlmIHNvLA0KPiBpdCdsbCBqdXN0 IHBhc3MgdGhhdCBwaWVjZSBvZiBJbnRlbnQgdG8gdGhhdCBmdW5jdGlvbi4NCj4gPj4NCj4gPj4g SW4gdGhpcyB3YXksIGEgZnVuY3Rpb24gb3duZXIgZGVmaW5lcyBib3RoIHRoZSBpbnRlbnQgaXRz ZWxmLCBhbmQgaG93IGl0IGlzDQo+IGNvbnN1bWVkIGJ5IGhpcy9oZXIgZnVuY3Rpb24uDQo+ID4+ DQo+ID4+IFtkdXpvbmdwZW5nXUFncmVlLltkdXpvbmdwZW5nXQ0KPiA+Pg0KPiA+PiBUaGUgZG93 bnNpZGUgaXMgdGhhdCB0aGVyZSBpcyBubyAiY29udHJvbCIgYW4vL2QgcGVvcGxlIGNhbiBkbyB3 aGF0IHRoZXkNCj4gd2FudCwgaW4gdGhlIHdvcnN0IGNhc2UgZXZlbiBwdXNoIGNvbmZpZywgd2hp Y2ggaXMgY2xlYXJseSBOT1Qgd2hhdCBJbnRlbnQgaXMNCj4gbWVhbnQgdG8gZG8uIEknZCBob3Bl IHRoYXQgcGVvcGxlIGRvbid0IHN0YXJ0IHdyaXRpbmcgSW50ZW50IGxpa2UgIm5vZGUgMTc6DQo+ IGhlcmUgaXMgeW91ciBDTEk7IG5vZGUgMjM6IGhlcmUgaXMgeW91ciBDTEkiOyBub2RlIDM3OiBo ZXJlIGlzIHlvdXIgQ0xJIi4gQnV0DQo+IHdoYXQgSSBkZXNjcmliZSBhYm92ZSBDT1VMRCBiZSBh YnVzZWQgaW4gdGhpcyB3YXkuIFJGQzc1NzUgc2F5cyB0aGlzIGlzDQo+IE5PVCBJbnRlbnQuDQo+ ID4+DQo+ID4+IFtkdXpvbmdwZW5nXUFncmVlLiBJdCBpcyBoYXJkIHRvIGdpdmUgYW4gZXhhY3Qg Ym91bmRhcnkuDQo+ID4+IEhvd2V2ZXIsIEkgdGhpbmsgd2UgaGF2ZSB0d28gY2hvaWNlcyBoZXJl Lg0KPiA+PiBUaGUgZmlyc3Qgb25lIGlzIHRvIGRlZmluZSBhIHBvbGljeSBmaXJzdCBpbiBldmVy eSBBU0EsIGJ1dCB3aXRoIHNvbWUgb3Blbg0KPiBwYXJhbWV0ZXJzIHRoYXQgY2FuIGJlIGNvbmZp Z3VyZWQuIFdoYXQgd2UgbmVlZCB0byBkbyBpcyBqdXN0IHRvIGRpc3RyaWJ1dGUNCj4gdGhlIG5l dHdvcmstbGV2ZWwgcGFyYW1ldGVycyBmb3IgdGhhdCBmdW5jdGlvbi4gVGhpcyB3YXkgaXMgc3Vp dGFibGUgdG8gc29tZQ0KPiByZWxhdGl2ZWx5IHN0YXRpYyBmdW5jdGlvbnMuDQo+ID4+IFRoZSBz ZWNvbmQgaXMgbW9yZSBhZHZhbmNlZCwgaW4gdGhlIGlkZWFsIGNhc2UsIHRoZSBodW1hbiBvcGVy YXRvciBjYW4NCj4gY29tbXVuaWNhdGUgd2l0aCBuZXR3b3JrIHVzaW5nIG5hdHVyYWwgbGFuZ3Vh Z2Ugb3Igc29tZXRoaW5nIHNpbWlsYXIuIFRoaXMNCj4gd2F5IGlzIG1vcmUgZmxleGlibGUsIGJ1 dCBoYXJkZXIgdG8gcmVhbGl6ZS4NCj4gPj4gW2R1em9uZ3BlbmddDQo+ID4+DQo+ID4+IEluIG90 aGVyIHdvcmRzOiBMZXQncyBub3QgYmUgdG9vIHByZXNjcmlwdGl2ZSwgYW5kIGhvcGUgdGhhdCBm b2xrcyB3aWxsIGRvDQo+IG1vcmUgb3IgbGVzcyByZWFzb25hYmxlIHRoaW5ncy4NCj4gPj4NCj4g Pj4gW2R1em9uZ3BlbmddQWdyZWUuW2R1em9uZ3BlbmddDQo+ID4+DQo+ID4+IEkndmUgYmVlbiBt ZWFuaW5nIHRvIGNvbW1lbnQgb24gdGhlIEludGVudCBkcmFmdHMgYW5kIHBvc3NpYmx5DQo+IGNv bnRyaWJ1dGUsIEknbSBqdXN0IHN0cnVnZ2xpbmcgd2l0aCBwcmlvcml0aWVzIGF0IHRoZSBtb21l bnQgKGFzIHlvdSBjYW4gc2VlDQo+IGJ5IG15IHJlbGF0aXZlIHNpbGVuY2Ugb3ZlciB0aGUgcGFz dCB3ZWVrcykuDQo+ID4+DQo+ID4+IE1pY2hhZWwNCj4gPj4NCj4gPj4NCj4gPj4NCj4gPj4gRnJv bTogQW5pbWEgW21haWx0bzphbmltYS1ib3VuY2VzQGlldGYub3JnXSBPbiBCZWhhbGYgT2YgRHV6 b25ncGVuZw0KPiA+PiBTZW50OiAxNiBNYXJjaCAyMDE2IDAyOjM4DQo+ID4+IFRvOiBhbmltYUBp ZXRmLm9yZzxtYWlsdG86YW5pbWFAaWV0Zi5vcmc+DQo+ID4+IENjOiBMYXVyZW50IENpYXZhZ2xp YQ0KPiA+PiA8bGF1cmVudC5jaWF2YWdsaWFAbm9raWEuY29tPG1haWx0bzpsYXVyZW50LmNpYXZh Z2xpYUBub2tpYS5jb20+PjsNCj4gPj4gSsOpZmVyc29uIENhbXBvcyBOb2JyZQ0KPiA+PiA8amNu b2JyZUBpbmYudWZyZ3MuYnI8bWFpbHRvOmpjbm9icmVAaW5mLnVmcmdzLmJyPj47IFNoZW5nIEpp YW5nDQo+ID4+IDxqaWFuZ3NoZW5nQGh1YXdlaS5jb208bWFpbHRvOmppYW5nc2hlbmdAaHVhd2Vp LmNvbT4+DQo+ID4+IFN1YmplY3Q6IFtBbmltYV0gRG9lcyBBdXRvbm9taWMgbmV0d29yayBuZWVk IHNvbWUgaW50ZXJ2ZW50aW9uDQo+IGJleW9uZA0KPiA+PiBpbnRlbnQNCj4gPj4NCj4gPj4gSGVs bG8sIGV2ZXJ5b25lDQo+ID4+DQo+ID4+ICAgICAgICAgIEkgdW5kZXJzdGFuZCB0aGF0IEF1dG9u b21pYyBuZXR3b3JrIHNob3VsZCBiZSBhYmxlIHRvIHdvcmsNCj4gd2l0aCBhcyBsZXNzIGNvbmZp Z3VyYXRpb24gYXMgcG9zc2libGUuIEluIHRoZSBpZGVhbCBjYXNlLCBhbiBBdXRvbm9taWMNCj4g bmV0d29yayBzaG91bGQgYmUgYWJsZSB0byB3b3JrIHdlbGwgd2hpbGUgb25seSBpbnRlbnQgaXMg bmVlZGVkIGZyb20gaHVtYW4NCj4gb3BlcmF0b3JzLg0KPiA+Pg0KPiA+PiAgICAgICAgICBIb3dl dmVyLCBJIHdvbmRlciB3aGV0aGVyIEF1dG9ub21pYyBuZXR3b3JrIG1heSBuZWVkDQo+IHNvbWUg aW50ZXJ2ZW50aW9uIGZyb20gaHVtYW4gb3BlcmF0b3JzIHRocm91Z2ggc29tZSBsb3ctbGV2ZWwg cG9saWNpZXMNCj4gYmV5b25kIGludGVudHMuDQo+ID4+DQo+ID4+ICAgICAgICAgIEZvciBleGFt cGxlLCBpbiB0aGUNCj4gaHR0cHM6Ly90b29scy5pZXRmLm9yZy9odG1sL2RyYWZ0LWlldGYtYW5p bWEtcHJlZml4LW1hbmFnZW1lbnQtMDAsIGl0IGlzDQo+IHN1Z2dlc3RlZCB0aGF0IHRoZSBwcmVm aXggbGVuZ3RocyBmb3IgdGhlIENTRywgQVNHLCBSU0cgKGRpZmZlcmVudCByb2xlcyBpbiBJUA0K PiBSQU4pIGNhbiBiZSBhc3NpZ25lZCBhcyBhbiAiaW50ZW50Ii4gVGhlc2UgY29uZmlndXJhdGlv biBwYXJhbWV0ZXJzIG5lZWQNCj4gdG8gYmUgZGlzdHJpYnV0ZWQgaW4gdGhlIGF1dG9ub21pYyBk b21haW4gdG8gaW5mbHVlbmNlIHRoZSBkZXRhaWwNCj4gY29uZmlndXJhdGlvbnMgb24gZWFjaCBh dXRvbm9taWMgbm9kZS4NCj4gPj4NCj4gPj4gICAgICAgICAgQXMgaW50ZW50IGlzIGRlc2NyaWJl ZCBhcyBhbiBhYnN0cmFjdCwgZGVjbGFyYXRpdmUsIGhpZ2gtbGV2ZWwNCj4gcG9saWN5IHVzZWQg dG8gb3BlcmF0ZSBhbiBhdXRvbm9taWMgZG9tYWluLCBzdWNoIGFzIGFuIGVudGVycHJpc2UgbmV0 d29yay4NCj4gUGVyaGFwcyB3ZSBjYW4gcmVuYW1lIHRoZSAiaW50ZW50IiBpbiB0aGUgInByZWZp eC1tYW5hZ2VtZW50IElEIiB0byBzb21lDQo+IG90aGVyIHdvcmRzLg0KPiA+Pg0KPiA+PiBJcyAi QVNBIHBhcmFtZXRlcnMiIG9rIGhlcmU/DQo+ID4+DQo+ID4+IEkgdGhpbmsgc29tZSBjaGFyYWN0 ZXJpc3RpY3MgZm9yIHRoaXMga2luZCBvZiBsb3ctbGV2ZWwgcG9saWNpZXMgYXJlDQo+ID4+DQo+ ID4+IDEuICAgICAgIE5ldHdvcmsgbGV2ZWwgcGFyYW1ldGVycywgbmVlZCB0byBiZSBkaXN0cmli dXRlZCBieSBHUkFTUA0KPiBhbmQgQUNQDQo+ID4+DQo+ID4+IDIuICAgICAgIFJlbGF0aXZlbHkg c3RhdGljIGNvbXBhcmVkIHRvIGludGVudCwgcGVyaGFwcyBvbmx5IG5lZWRlZCB0byBiZQ0KPiBj b25maWd1cmVkIG9uY2UNCj4gPj4NCj4gPj4gMy4gICAgICAgVW5hdm9pZGFibGUgaW4gY3VycmVu dCBuZXR3b3JrIGVudmlyb25tZW50LCBtb3N0bHkgZm9yDQo+IGVzdGFibGlzaGluZyB0aGUgbmV0 d29yayBpbmZyYXN0cnVjdHVyZQ0KPiA+Pg0KPiA+PiA0LiAgICAgICBSYXJlbHkgbmVlZCBjb29y ZGluYXRpb25zIHdpdGggb3RoZXJzIHBhcmFtZXRlcnMsIHVzdWFsbHkgaXMNCj4gY2xvc2VkIHJl bGF0ZWQgdG8gb25lIHNwZWNpZmljIGF1dG9ub21pYyBmdW5jdGlvbg0KPiA+Pg0KPiA+PiBXZWxj b21lIGZvciBjb21tZW50LiBZb3VyIHN1Z2dlc3Rpb25zIHdpbGwgYmUgYXBwcmVjaWF0ZWQuDQo+ ID4+DQo+ID4+DQo+ID4+IEJlc3QgUmVnYXJkcw0KPiA+PiBab25ncGVuZyBEdQ0KPiA+Pg0KPiA+ Pg0KPiA+Pg0KPiA+PiBfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fXw0KPiA+PiBBbmltYSBtYWlsaW5nIGxpc3QNCj4gPj4gQW5pbWFAaWV0Zi5vcmcNCj4gPj4g aHR0cHM6Ly93d3cuaWV0Zi5vcmcvbWFpbG1hbi9saXN0aW5mby9hbmltYQ0KPiA+Pg0KPiA+DQo+ IA0KPiBfX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXw0KPiBB bmltYSBtYWlsaW5nIGxpc3QNCj4gQW5pbWFAaWV0Zi5vcmcNCj4gaHR0cHM6Ly93d3cuaWV0Zi5v cmcvbWFpbG1hbi9saXN0aW5mby9hbmltYQ0K From nobody Mon Mar 21 03:02:41 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1AC3C12D6A2 for ; Mon, 21 Mar 2016 03:02:40 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.222 X-Spam-Level: X-Spam-Status: No, score=-4.222 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ODwceW8tMpUv for ; Mon, 21 Mar 2016 03:02:37 -0700 (PDT) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0190012D66D for ; Mon, 21 Mar 2016 03:02:36 -0700 (PDT) Received: from 172.18.7.190 (EHLO lhreml707-cah.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CLB09245; Mon, 21 Mar 2016 10:02:34 +0000 (GMT) Received: from NKGEML402-HUB.china.huawei.com (10.98.56.33) by lhreml707-cah.china.huawei.com (10.201.5.199) with Microsoft SMTP Server (TLS) id 14.3.235.1; Mon, 21 Mar 2016 10:02:33 +0000 Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml402-hub.china.huawei.com ([10.98.56.33]) with mapi id 14.03.0235.001; Mon, 21 Mar 2016 18:02:29 +0800 From: "Liubing (Leo)" To: "Michael Behringer (mbehring)" , "anima@ietf.org" Thread-Topic: CBOR for Intent?-//RE: Intent yang model - some thoughts Thread-Index: AdGA84Aiy7EwHtVtTUWSCghLdxIQCwCYxtng Date: Mon, 21 Mar 2016 10:02:28 +0000 Message-ID: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> In-Reply-To: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> Accept-Language: en-US, zh-CN Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.98.117] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-CFilter-Loop: Reflected X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020201.56EFC6BB.009E, ss=1, re=0.000, recu=0.000, reip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32 X-Mirapoint-Loop-Id: cdb128c886e9b40c796fdf2aa0f8003e Archived-At: Subject: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 10:02:40 -0000 Hi Michael and all, One quick comment aside from the Intent content discussion: since GRASP use= s CBOR, isn't it good for Intent also encoded by CBOR? YANG model can surely describe an Intent, but it includes lots of features = for data manipulating (for Netconf), which might not necessary for Intent d= escribing?.=20 Best regards, Bing > -----Original Message----- > From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael > Behringer (mbehring) > Sent: Friday, March 18, 2016 5:52 PM > To: anima@ietf.org > Subject: [Anima] Intent yang model - some thoughts >=20 > WRT the current discussion on what should be in Intent and what shouldn't= , > as mentioned before, I think we should leave this to the owner of an > autonomic function to decide. This risks "abuse" in a way that wasn't > foreseen, but it gives maximum agility, which is what we need to get ANIM= A > off the ground, IMO. If we are in the nice position to find that it gets > adopted a lot, and if there are things that need fixing then, we can work= on > version 2. >=20 > Several months ago I started with a yang model for Intent, but never fini= shed. > Maybe this helps with this discussion. Note I'm not at all a yang expert,= so > please look more at the general structure for now. Not sure if it helps t= he > discussion... >=20 > +--rw intent > +--rw meta-data > | +--rw intent-version uint32 // the version of the intent > specification > | +--rw timestamp uint32 // latest time stamp wins > | +--rw administrator string // who created that intent > | +--rw device? string // on which node was it created > | +--rw execution-time uint32 // plus other parameters on when to > execute > | +--rw expiration uint32 // when does this intent cease to be > valid > | +--rw feedback // how to let the operator know > about success/failure > | +--rw signature ?? // intent must be signed > | > +--rw scope > +--rw domains > +--rw domain* [name] > +--rw name string // is there a fqdn type defined? > +--rw meta-data // general about the domain > | +--rw include-subdomains boolean // should > sub-domains follow this policy? > | +--rw fail-open boolean // fail-open or fail-closed > policy > | +--rw maintenance-window // when can critical > operations take place > | +--rw ... // other general policies on > domain level > | ... > | > +--rw ani-intent > +-- (contains all policies which the ANI should follow) > | ex: ACP policies, coordination between AN functions, > security policies, etc > | for ani-intent, scope is always "all" (ie all nodes must > process) > | ... > ... > All of the above MUST be standardised, because it refers to the AN > infrastructure itself. > Of the below we need to standardise the structure, but not content for ea= ch > function Conceptually, we define Intent for a domain. > ... > | > +--rw intent-for-autonomic-functions > +--rw function_1 [name] > | +--rw scope // which devices > in the domain need to listen. > | | +--rw all boolean // all devices > | | +--rw role string // as defined in > certificate > | | +--rw dynamic_role string // as determined > dynamically > | | +--rw tag string // can be used for > location > | +-- function_intent // to be defined by > function owner > | > +--rw function_2 [name] > | +--rw scope // which devices > in the domain need to listen. > | | +--rw all boolean // all devices > | | +--rw role string // as defined in > certificate > | | +--rw dynamic_role string // as determined > dynamically > | | +--rw tag string // can be used for > location > | +-- function_intent // to be defined by > function owner > | > ... >=20 > If a node receives Intent, it checks: > 1) meta data: Mostly for forwarding and processing of Intent > 2) domain info: domain specific policy > 3) ANI Intent: Intent for the ANI itself; every node must be able to proc= ess > that > 4) For each of the "function" blocks: > - should I run this function? > - do I run this function? > - if so, forward the rest of that block to the corresponding handler. >=20 > I think we want a clean architectural separation between the ANI and the > functions. This way every function can do what it wants. >=20 > Thoughts? > Michael >=20 > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima From nobody Mon Mar 21 03:45:09 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C0BFE12D66D for ; Mon, 21 Mar 2016 03:45:08 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.92 X-Spam-Level: X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id m17hEhTD5Amh for ; Mon, 21 Mar 2016 03:45:05 -0700 (PDT) Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 37D4512D640 for ; Mon, 21 Mar 2016 03:45:05 -0700 (PDT) Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id A306B9B8EB9DC; Mon, 21 Mar 2016 10:45:00 +0000 (GMT) Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2LAj2oF027557 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 21 Mar 2016 10:45:02 GMT Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2LAio4V011826 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 21 Mar 2016 11:45:01 +0100 Received: from [172.27.205.106] (135.239.27.41) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Mon, 21 Mar 2016 11:44:33 +0100 To: "EXT Liubing (Leo)" , "Michael Behringer (mbehring)" , "anima@ietf.org" References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> From: Laurent Ciavaglia Organization: Bell Labs Message-ID: <56EFD090.1070501@alcatel-lucent.com> Date: Mon, 21 Mar 2016 11:44:32 +0100 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0 MIME-Version: 1.0 In-Reply-To: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> Content-Type: multipart/alternative; boundary="------------060205080101030307030700" X-Originating-IP: [135.239.27.41] Archived-At: Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 10:45:09 -0000 --------------060205080101030307030700 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Hello, I think that before considering encodings, we should first agree on the definition(s) of what we will manipulate. Let's start with a (formal) definiton of the "ANIMA" intent policy, and the hierarchy envisaged and coarse "format". If we can work out an example with we think capture well enough and large enough how and what the "ANIMA" is used for, then I think we will already have made great progress. Best regards, Laurent. On 21/03/2016 11:02, EXT Liubing (Leo) wrote: > Hi Michael and all, > > One quick comment aside from the Intent content discussion: since GRASP uses CBOR, isn't it good for Intent also encoded by CBOR? > > YANG model can surely describe an Intent, but it includes lots of features for data manipulating (for Netconf), which might not necessary for Intent describing?. > > Best regards, > Bing > >> -----Original Message----- >> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael >> Behringer (mbehring) >> Sent: Friday, March 18, 2016 5:52 PM >> To: anima@ietf.org >> Subject: [Anima] Intent yang model - some thoughts >> >> WRT the current discussion on what should be in Intent and what shouldn't, >> as mentioned before, I think we should leave this to the owner of an >> autonomic function to decide. This risks "abuse" in a way that wasn't >> foreseen, but it gives maximum agility, which is what we need to get ANIMA >> off the ground, IMO. If we are in the nice position to find that it gets >> adopted a lot, and if there are things that need fixing then, we can work on >> version 2. >> >> Several months ago I started with a yang model for Intent, but never finished. >> Maybe this helps with this discussion. Note I'm not at all a yang expert, so >> please look more at the general structure for now. Not sure if it helps the >> discussion... >> >> +--rw intent >> +--rw meta-data >> | +--rw intent-version uint32 // the version of the intent >> specification >> | +--rw timestamp uint32 // latest time stamp wins >> | +--rw administrator string // who created that intent >> | +--rw device? string // on which node was it created >> | +--rw execution-time uint32 // plus other parameters on when to >> execute >> | +--rw expiration uint32 // when does this intent cease to be >> valid >> | +--rw feedback // how to let the operator know >> about success/failure >> | +--rw signature ?? // intent must be signed >> | >> +--rw scope >> +--rw domains >> +--rw domain* [name] >> +--rw name string // is there a fqdn type defined? >> +--rw meta-data // general about the domain >> | +--rw include-subdomains boolean // should >> sub-domains follow this policy? >> | +--rw fail-open boolean // fail-open or fail-closed >> policy >> | +--rw maintenance-window // when can critical >> operations take place >> | +--rw ... // other general policies on >> domain level >> | ... >> | >> +--rw ani-intent >> +-- (contains all policies which the ANI should follow) >> | ex: ACP policies, coordination between AN functions, >> security policies, etc >> | for ani-intent, scope is always "all" (ie all nodes must >> process) >> | ... >> ... >> All of the above MUST be standardised, because it refers to the AN >> infrastructure itself. >> Of the below we need to standardise the structure, but not content for each >> function Conceptually, we define Intent for a domain. >> ... >> | >> +--rw intent-for-autonomic-functions >> +--rw function_1 [name] >> | +--rw scope // which devices >> in the domain need to listen. >> | | +--rw all boolean // all devices >> | | +--rw role string // as defined in >> certificate >> | | +--rw dynamic_role string // as determined >> dynamically >> | | +--rw tag string // can be used for >> location >> | +-- function_intent // to be defined by >> function owner >> | >> +--rw function_2 [name] >> | +--rw scope // which devices >> in the domain need to listen. >> | | +--rw all boolean // all devices >> | | +--rw role string // as defined in >> certificate >> | | +--rw dynamic_role string // as determined >> dynamically >> | | +--rw tag string // can be used for >> location >> | +-- function_intent // to be defined by >> function owner >> | >> ... >> >> If a node receives Intent, it checks: >> 1) meta data: Mostly for forwarding and processing of Intent >> 2) domain info: domain specific policy >> 3) ANI Intent: Intent for the ANI itself; every node must be able to process >> that >> 4) For each of the "function" blocks: >> - should I run this function? >> - do I run this function? >> - if so, forward the rest of that block to the corresponding handler. >> >> I think we want a clean architectural separation between the ANI and the >> functions. This way every function can do what it wants. >> >> Thoughts? >> Michael >> >> _______________________________________________ >> Anima mailing list >> Anima@ietf.org >> https://www.ietf.org/mailman/listinfo/anima > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima > -- Laurent Ciavaglia Senior Research Manager Bell Labs, Nokia +33 160 402 636 Route de Villejust | 91620 Nozay | France LinkedIn: laurentciavaglia --------------060205080101030307030700 Content-Type: text/html; charset="windows-1252" Content-Transfer-Encoding: 8bit Hello,

I think that before considering encodings, we should first agree on the definition(s) of what we will manipulate.
Let's start with a (formal) definiton of the "ANIMA" intent policy, and the hierarchy envisaged and coarse "format".
If we can work out an example with we think capture well enough and large enough how and what the "ANIMA" is used for, then I think we will already have made great progress.

Best regards, Laurent.

On 21/03/2016 11:02, EXT Liubing (Leo) wrote:
Hi Michael and all,

One quick comment aside from the Intent content discussion: since GRASP uses CBOR, isn't it good for Intent also encoded by CBOR?

YANG model can surely describe an Intent, but it includes lots of features for data manipulating (for Netconf), which might not necessary for Intent describing?. 

Best regards,
Bing


-----Original Message-----
From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael
Behringer (mbehring)
Sent: Friday, March 18, 2016 5:52 PM
To: anima@ietf.org
Subject: [Anima] Intent yang model - some thoughts

WRT the current discussion on what should be in Intent and what shouldn't,
as mentioned before, I think we should leave this to the owner of an
autonomic function to decide. This risks "abuse" in a way that wasn't
foreseen, but it gives maximum agility, which is what we need to get ANIMA
off the ground, IMO. If we are in the nice position to find that it gets
adopted a lot, and if there are things that need fixing then, we can work on
version 2.

Several months ago I started with a yang model for Intent, but never finished.
Maybe this helps with this discussion. Note I'm not at all a yang expert, so
please look more at the general structure for now. Not sure if it helps the
discussion...

+--rw intent
   +--rw meta-data
   |  +--rw intent-version  uint32   // the version of the intent
specification
   |  +--rw timestamp 		uint32   // latest time stamp wins
   |  +--rw administrator 	string   // who created that intent
   |  +--rw device?   		string   // on which node was it created
   |  +--rw execution-time	uint32   // plus other parameters on when to
execute
   |  +--rw expiration    	uint32   // when does this intent cease to be
valid
   |  +--rw feedback                 // how to let the operator know
about success/failure
   |  +--rw signature		??       // intent must be signed
   |
   +--rw scope
      +--rw domains
         +--rw domain* [name]
            +--rw name		string    // is there a fqdn type defined?
            +--rw meta-data           // general about the domain
            |  +--rw include-subdomains	boolean  // should
sub-domains follow this policy?
            |  +--rw fail-open  boolean  // fail-open or fail-closed
policy
            |  +--rw maintenance-window  // when can critical
operations take place
            |  +--rw ...                 // other general policies on
domain level
|        ...
            |
            +--rw ani-intent
            +-- (contains all policies which the ANI should follow)
            |  ex: ACP policies, coordination between AN functions,
security policies, etc
            |  for ani-intent, scope is always "all" (ie all nodes must
process)
            |  ...
...
All of the above MUST be standardised, because it refers to the AN
infrastructure itself.
Of the below we need to standardise the structure, but not content for each
function Conceptually, we define Intent for a domain.
...
            |
            +--rw intent-for-autonomic-functions
               +--rw function_1 [name]
               |  +--rw scope                    // which devices
in the domain need to listen.
               |  |  +--rw all    boolean        // all devices
               |  |  +--rw role   string         // as defined in
certificate
               |  |  +--rw dynamic_role  string  // as determined
dynamically
               |  |  +--rw tag    string         // can be used for
location
               |  +-- function_intent            // to be defined by
function owner
               |
               +--rw function_2 [name]
               |  +--rw scope                    // which devices
in the domain need to listen.
               |  |  +--rw all    boolean        // all devices
               |  |  +--rw role   string         // as defined in
certificate
               |  |  +--rw dynamic_role  string  // as determined
dynamically
               |  |  +--rw tag    string         // can be used for
location
               |  +-- function_intent            // to be defined by
function owner
               |
...

If a node receives Intent, it checks:
1) meta data: Mostly for forwarding and processing of Intent
2) domain info: domain specific policy
3) ANI Intent: Intent for the ANI itself; every node must be able to process
that
4) For each of the "function" blocks:
   - should I run this function?
   - do I run this function?
   - if so, forward the rest of that block to the corresponding handler.

I think we want a clean architectural separation between the ANI and the
functions. This way every function can do what it wants.

Thoughts?
Michael

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima

--

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

 

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia

 

--------------060205080101030307030700-- From nobody Mon Mar 21 03:45:23 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7390912D71B for ; Mon, 21 Mar 2016 03:45:12 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.92 X-Spam-Level: X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZDB0f7pWlJ8A for ; Mon, 21 Mar 2016 03:45:10 -0700 (PDT) Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 82FA512D640 for ; Mon, 21 Mar 2016 03:45:09 -0700 (PDT) Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id 9029212EA8D89; Mon, 21 Mar 2016 10:45:05 +0000 (GMT) Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2LAj7Ck027590 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 21 Mar 2016 10:45:07 GMT Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2LAio4Y011826 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 21 Mar 2016 11:45:06 +0100 Received: from [172.27.205.106] (135.239.27.38) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Mon, 21 Mar 2016 11:44:44 +0100 To: "EXT Liubing (Leo)" , "Michael Behringer (mbehring)" , "anima@ietf.org" References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> From: Laurent Ciavaglia Organization: Bell Labs Message-ID: <56EFD096.1000106@alcatel-lucent.com> Date: Mon, 21 Mar 2016 11:44:38 +0100 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0 MIME-Version: 1.0 In-Reply-To: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> Content-Type: multipart/alternative; boundary="------------000303070701070508030004" X-Originating-IP: [135.239.27.38] Archived-At: Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 10:45:12 -0000 --------------000303070701070508030004 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Hello, I think that before considering encodings, we should first agree on the definition(s) of what we will manipulate. Let's start with a (formal) definition of the "ANIMA" intent policy, and the hierarchy envisaged and coarse "format". If we can work out an example with we think capture well enough and large enough how and what the "ANIMA" is used for, then I think we will already have made great progress. Best regards, Laurent. On 21/03/2016 11:02, EXT Liubing (Leo) wrote: > Hi Michael and all, > > One quick comment aside from the Intent content discussion: since GRASP uses CBOR, isn't it good for Intent also encoded by CBOR? > > YANG model can surely describe an Intent, but it includes lots of features for data manipulating (for Netconf), which might not necessary for Intent describing?. > > Best regards, > Bing > >> -----Original Message----- >> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael >> Behringer (mbehring) >> Sent: Friday, March 18, 2016 5:52 PM >> To: anima@ietf.org >> Subject: [Anima] Intent yang model - some thoughts >> >> WRT the current discussion on what should be in Intent and what shouldn't, >> as mentioned before, I think we should leave this to the owner of an >> autonomic function to decide. This risks "abuse" in a way that wasn't >> foreseen, but it gives maximum agility, which is what we need to get ANIMA >> off the ground, IMO. If we are in the nice position to find that it gets >> adopted a lot, and if there are things that need fixing then, we can work on >> version 2. >> >> Several months ago I started with a yang model for Intent, but never finished. >> Maybe this helps with this discussion. Note I'm not at all a yang expert, so >> please look more at the general structure for now. Not sure if it helps the >> discussion... >> >> +--rw intent >> +--rw meta-data >> | +--rw intent-version uint32 // the version of the intent >> specification >> | +--rw timestamp uint32 // latest time stamp wins >> | +--rw administrator string // who created that intent >> | +--rw device? string // on which node was it created >> | +--rw execution-time uint32 // plus other parameters on when to >> execute >> | +--rw expiration uint32 // when does this intent cease to be >> valid >> | +--rw feedback // how to let the operator know >> about success/failure >> | +--rw signature ?? // intent must be signed >> | >> +--rw scope >> +--rw domains >> +--rw domain* [name] >> +--rw name string // is there a fqdn type defined? >> +--rw meta-data // general about the domain >> | +--rw include-subdomains boolean // should >> sub-domains follow this policy? >> | +--rw fail-open boolean // fail-open or fail-closed >> policy >> | +--rw maintenance-window // when can critical >> operations take place >> | +--rw ... // other general policies on >> domain level >> | ... >> | >> +--rw ani-intent >> +-- (contains all policies which the ANI should follow) >> | ex: ACP policies, coordination between AN functions, >> security policies, etc >> | for ani-intent, scope is always "all" (ie all nodes must >> process) >> | ... >> ... >> All of the above MUST be standardised, because it refers to the AN >> infrastructure itself. >> Of the below we need to standardise the structure, but not content for each >> function Conceptually, we define Intent for a domain. >> ... >> | >> +--rw intent-for-autonomic-functions >> +--rw function_1 [name] >> | +--rw scope // which devices >> in the domain need to listen. >> | | +--rw all boolean // all devices >> | | +--rw role string // as defined in >> certificate >> | | +--rw dynamic_role string // as determined >> dynamically >> | | +--rw tag string // can be used for >> location >> | +-- function_intent // to be defined by >> function owner >> | >> +--rw function_2 [name] >> | +--rw scope // which devices >> in the domain need to listen. >> | | +--rw all boolean // all devices >> | | +--rw role string // as defined in >> certificate >> | | +--rw dynamic_role string // as determined >> dynamically >> | | +--rw tag string // can be used for >> location >> | +-- function_intent // to be defined by >> function owner >> | >> ... >> >> If a node receives Intent, it checks: >> 1) meta data: Mostly for forwarding and processing of Intent >> 2) domain info: domain specific policy >> 3) ANI Intent: Intent for the ANI itself; every node must be able to process >> that >> 4) For each of the "function" blocks: >> - should I run this function? >> - do I run this function? >> - if so, forward the rest of that block to the corresponding handler. >> >> I think we want a clean architectural separation between the ANI and the >> functions. This way every function can do what it wants. >> >> Thoughts? >> Michael >> >> _______________________________________________ >> Anima mailing list >> Anima@ietf.org >> https://www.ietf.org/mailman/listinfo/anima > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima > -- Laurent Ciavaglia Senior Research Manager Bell Labs, Nokia +33 160 402 636 Route de Villejust | 91620 Nozay | France LinkedIn: laurentciavaglia --------------000303070701070508030004 Content-Type: text/html; charset="windows-1252" Content-Transfer-Encoding: 8bit Hello,

I think that before considering encodings, we should first agree on the definition(s) of what we will manipulate.
Let's start with a (formal) definition of the "ANIMA" intent policy, and the hierarchy envisaged and coarse "format".
If we can work out an example with we think capture well enough and large enough how and what the "ANIMA" is used for, then I think we will already have made great progress.

Best regards, Laurent.

On 21/03/2016 11:02, EXT Liubing (Leo) wrote:
Hi Michael and all,

One quick comment aside from the Intent content discussion: since GRASP uses CBOR, isn't it good for Intent also encoded by CBOR?

YANG model can surely describe an Intent, but it includes lots of features for data manipulating (for Netconf), which might not necessary for Intent describing?. 

Best regards,
Bing


-----Original Message-----
From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael
Behringer (mbehring)
Sent: Friday, March 18, 2016 5:52 PM
To: anima@ietf.org
Subject: [Anima] Intent yang model - some thoughts

WRT the current discussion on what should be in Intent and what shouldn't,
as mentioned before, I think we should leave this to the owner of an
autonomic function to decide. This risks "abuse" in a way that wasn't
foreseen, but it gives maximum agility, which is what we need to get ANIMA
off the ground, IMO. If we are in the nice position to find that it gets
adopted a lot, and if there are things that need fixing then, we can work on
version 2.

Several months ago I started with a yang model for Intent, but never finished.
Maybe this helps with this discussion. Note I'm not at all a yang expert, so
please look more at the general structure for now. Not sure if it helps the
discussion...

+--rw intent
   +--rw meta-data
   |  +--rw intent-version  uint32   // the version of the intent
specification
   |  +--rw timestamp 		uint32   // latest time stamp wins
   |  +--rw administrator 	string   // who created that intent
   |  +--rw device?   		string   // on which node was it created
   |  +--rw execution-time	uint32   // plus other parameters on when to
execute
   |  +--rw expiration    	uint32   // when does this intent cease to be
valid
   |  +--rw feedback                 // how to let the operator know
about success/failure
   |  +--rw signature		??       // intent must be signed
   |
   +--rw scope
      +--rw domains
         +--rw domain* [name]
            +--rw name		string    // is there a fqdn type defined?
            +--rw meta-data           // general about the domain
            |  +--rw include-subdomains	boolean  // should
sub-domains follow this policy?
            |  +--rw fail-open  boolean  // fail-open or fail-closed
policy
            |  +--rw maintenance-window  // when can critical
operations take place
            |  +--rw ...                 // other general policies on
domain level
|        ...
            |
            +--rw ani-intent
            +-- (contains all policies which the ANI should follow)
            |  ex: ACP policies, coordination between AN functions,
security policies, etc
            |  for ani-intent, scope is always "all" (ie all nodes must
process)
            |  ...
...
All of the above MUST be standardised, because it refers to the AN
infrastructure itself.
Of the below we need to standardise the structure, but not content for each
function Conceptually, we define Intent for a domain.
...
            |
            +--rw intent-for-autonomic-functions
               +--rw function_1 [name]
               |  +--rw scope                    // which devices
in the domain need to listen.
               |  |  +--rw all    boolean        // all devices
               |  |  +--rw role   string         // as defined in
certificate
               |  |  +--rw dynamic_role  string  // as determined
dynamically
               |  |  +--rw tag    string         // can be used for
location
               |  +-- function_intent            // to be defined by
function owner
               |
               +--rw function_2 [name]
               |  +--rw scope                    // which devices
in the domain need to listen.
               |  |  +--rw all    boolean        // all devices
               |  |  +--rw role   string         // as defined in
certificate
               |  |  +--rw dynamic_role  string  // as determined
dynamically
               |  |  +--rw tag    string         // can be used for
location
               |  +-- function_intent            // to be defined by
function owner
               |
...

If a node receives Intent, it checks:
1) meta data: Mostly for forwarding and processing of Intent
2) domain info: domain specific policy
3) ANI Intent: Intent for the ANI itself; every node must be able to process
that
4) For each of the "function" blocks:
   - should I run this function?
   - do I run this function?
   - if so, forward the rest of that block to the corresponding handler.

I think we want a clean architectural separation between the ANI and the
functions. This way every function can do what it wants.

Thoughts?
Michael

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima

--

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

 

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia

 

--------------000303070701070508030004-- From nobody Mon Mar 21 03:54:36 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3F42512D648 for ; Mon, 21 Mar 2016 03:54:34 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.501 X-Spam-Level: X-Spam-Status: No, score=-14.501 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, T_FILL_THIS_FORM_SHORT=0.01, T_KAM_HTML_FONT_INVALID=0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Qjk6Lbz5VElq for ; Mon, 21 Mar 2016 03:54:30 -0700 (PDT) Received: from alln-iport-7.cisco.com (alln-iport-7.cisco.com [173.37.142.94]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 492D412D557 for ; Mon, 21 Mar 2016 03:54:30 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=43716; q=dns/txt; s=iport; t=1458557670; x=1459767270; h=from:to:subject:date:message-id:references:in-reply-to: mime-version; bh=byey5QjKhAg+wWTu6rKArn3jlClAkcpem+N7JVIZ1WI=; b=M9UAQ+MNM5Wx8vyLMutq6AHjy1nYkEUlEAMR8LjyMqrWxKtJSR6qotJY wwjABnTrpwgRrA1AuW95oHmzqrznuPm2FV7IkLn3c133t/93CRb/Fb6Tg aM8+zuUvditNWw6svjIE4Q26VYuufwWzfojtXQj7o0XPCzL1mAXfeWzwP U=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AaBQAI0u9W/51dJa1egmdMU3IGuiyBb?= =?us-ascii?q?AMXAQmFbAKBJDkTAQEBAQEBAWQnhEEBAQEEAQEBKj4DFwQCAQgRBAEBIQEGByc?= =?us-ascii?q?LFAkIAQEEARIIiB8OvhYBAQEBAQEBAQEBAQEBAQEBAQEBAQEVhh6DRX+EaQmFI?= =?us-ascii?q?AWNO4VJhFMBhXCIDIFshEqIWI8FAQ8SAUCBfoFnaokXfgEBAQ?= X-IronPort-AV: E=Sophos;i="5.24,371,1454976000"; d="scan'208,217";a="251702989" Received: from rcdn-core-6.cisco.com ([173.37.93.157]) by alln-iport-7.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 21 Mar 2016 10:54:28 +0000 Received: from XCH-RCD-006.cisco.com (xch-rcd-006.cisco.com [173.37.102.16]) by rcdn-core-6.cisco.com (8.14.5/8.14.5) with ESMTP id u2LAsSfk014564 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Mon, 21 Mar 2016 10:54:28 GMT Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-006.cisco.com (173.37.102.16) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Mon, 21 Mar 2016 05:54:27 -0500 Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Mon, 21 Mar 2016 05:54:27 -0500 From: "Michael Behringer (mbehring)" To: Laurent Ciavaglia , "EXT Liubing (Leo)" , "anima@ietf.org" Thread-Topic: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts Thread-Index: AdGA84Aiy7EwHtVtTUWSCghLdxIQCwCYxtngAAx9CgAACi75sA== Date: Mon, 21 Mar 2016 10:54:27 +0000 Message-ID: <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> In-Reply-To: <56EFD090.1070501@alcatel-lucent.com> Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.55.238.139] Content-Type: multipart/alternative; boundary="_000_7471201223ba4962a66dd5058190ecedXCHRCD006ciscocom_" MIME-Version: 1.0 Archived-At: Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 10:54:34 -0000 --_000_7471201223ba4962a66dd5058190ecedXCHRCD006ciscocom_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Agree; the encoding format question is secondary. My YANG example was just "playing around", at this point I'm trying to find= the right structure. So don't look at the details, just the structure. Michael From: Laurent Ciavaglia [mailto:laurent.ciavaglia@nokia.com] Sent: 21 March 2016 11:45 To: EXT Liubing (Leo) ; Michael Behringer (mbehring= ) ; anima@ietf.org Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some though= ts Hello, I think that before considering encodings, we should first agree on the def= inition(s) of what we will manipulate. Let's start with a (formal) definiton of the "ANIMA" intent policy, and the= hierarchy envisaged and coarse "format". If we can work out an example with we think capture well enough and large e= nough how and what the "ANIMA" is used for, then I think we will already ha= ve made great progress. Best regards, Laurent. On 21/03/2016 11:02, EXT Liubing (Leo) wrote: Hi Michael and all, One quick comment aside from the Intent content discussion: since GRASP use= s CBOR, isn't it good for Intent also encoded by CBOR? YANG model can surely describe an Intent, but it includes lots of features = for data manipulating (for Netconf), which might not necessary for Intent d= escribing?. Best regards, Bing -----Original Message----- From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael Behringer (mbehring) Sent: Friday, March 18, 2016 5:52 PM To: anima@ietf.org Subject: [Anima] Intent yang model - some thoughts WRT the current discussion on what should be in Intent and what shouldn't, as mentioned before, I think we should leave this to the owner of an autonomic function to decide. This risks "abuse" in a way that wasn't foreseen, but it gives maximum agility, which is what we need to get ANIMA off the ground, IMO. If we are in the nice position to find that it gets adopted a lot, and if there are things that need fixing then, we can work o= n version 2. Several months ago I started with a yang model for Intent, but never finish= ed. Maybe this helps with this discussion. Note I'm not at all a yang expert, s= o please look more at the general structure for now. Not sure if it helps the discussion... +--rw intent +--rw meta-data | +--rw intent-version uint32 // the version of the intent specification | +--rw timestamp uint32 // latest time stamp wins | +--rw administrator string // who created that intent | +--rw device? string // on which node was it created | +--rw execution-time uint32 // plus other parameters on when = to execute | +--rw expiration uint32 // when does this intent cease to= be valid | +--rw feedback // how to let the operator know about success/failure | +--rw signature ?? // intent must be signed | +--rw scope +--rw domains +--rw domain* [name] +--rw name string // is there a fqdn type defined? +--rw meta-data // general about the domain | +--rw include-subdomains boolean // should sub-domains follow this policy? | +--rw fail-open boolean // fail-open or fail-closed policy | +--rw maintenance-window // when can critical operations take place | +--rw ... // other general policies on domain level | ... | +--rw ani-intent +-- (contains all policies which the ANI should follow) | ex: ACP policies, coordination between AN functions, security policies, etc | for ani-intent, scope is always "all" (ie all nodes must process) | ... ... All of the above MUST be standardised, because it refers to the AN infrastructure itself. Of the below we need to standardise the structure, but not content for each function Conceptually, we define Intent for a domain. ... | +--rw intent-for-autonomic-functions +--rw function_1 [name] | +--rw scope // which devices in the domain need to listen. | | +--rw all boolean // all devices | | +--rw role string // as defined in certificate | | +--rw dynamic_role string // as determined dynamically | | +--rw tag string // can be used for location | +-- function_intent // to be defined by function owner | +--rw function_2 [name] | +--rw scope // which devices in the domain need to listen. | | +--rw all boolean // all devices | | +--rw role string // as defined in certificate | | +--rw dynamic_role string // as determined dynamically | | +--rw tag string // can be used for location | +-- function_intent // to be defined by function owner | ... If a node receives Intent, it checks: 1) meta data: Mostly for forwarding and processing of Intent 2) domain info: domain specific policy 3) ANI Intent: Intent for the ANI itself; every node must be able to proces= s that 4) For each of the "function" blocks: - should I run this function? - do I run this function? - if so, forward the rest of that block to the corresponding handler. I think we want a clean architectural separation between the ANI and the functions. This way every function can do what it wants. Thoughts? Michael _______________________________________________ Anima mailing list Anima@ietf.org https://www.ietf.org/mailman/listinfo/anima _______________________________________________ Anima mailing list Anima@ietf.org https://www.ietf.org/mailman/listinfo/anima -- Laurent Ciavaglia Senior Research Manager Bell Labs, Nokia +33 160 402 636 Route de Villejust | 91620 Nozay | France LinkedIn: laurentciavaglia --_000_7471201223ba4962a66dd5058190ecedXCHRCD006ciscocom_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

= Agree; the = encoding format question is secondary.

= My YANG exa= mple was just “playing around”, at this point I’m trying = to find the right structure. So don’t look at the details, just the structure.

= Michael

=  =

From: Laurent Ciavaglia [mailto:laurent.ciavaglia@nokia.com]
Sent: 21 March 2016 11:45 To: EXT Liubing (Leo) <le= o.liubing@huawei.com>; Michael Behringer (mbehring) <mbehring@cisco.c= om>; anima@ietf.org
Subject: Re: [Anima] CBOR fo= r Intent?-//RE: Intent yang model - some thoughts<= /p>

=  <= /font>

Hello,

I think that before considering encodings, we should first agree on the def= inition(s) of what we will manipulate.
Let's start with a (formal) definiton of the "ANIMA" intent polic= y, and the hierarchy envisaged and coarse "format".
If we can work out an example with we think capture well enough and large e= nough how and what the "ANIMA" is used for, then I think we will = already have made great progress.

Best regards, Laurent.

On 21/03/2016 11:02, EXT Liubing (Leo) wrote:= 

Hi Michael and all,
 
One quick comment aside from the Intent=
 content discussion: since GRASP uses CBOR, isn't it good for Intent also e=
ncoded by CBOR?
 
YANG model can surely describe an Inten=
t, but it includes lots of features for data manipulating (for Netconf), wh=
ich might not necessary for Intent describing?. 

 


Best regards,<=
/pre>

Bing


 




-----Original Message-----


From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael


Behringer (mbehring)<=
/font>


Sent: Friday, March 18, 2016 5:52 PM


To: a=
nima@ietf.org


Subject: [Anima] Intent yang model - so=
me thoughts


 


WRT the current discussion on what shou=
ld be in Intent and what shouldn't,


as mentioned before, I think we should =
leave this to the owner of an


autonomic function to decide. This risk=
s "abuse" in a way that wasn't


foreseen, but it gives maximum agility,=
 which is what we need to get ANIMA


off the ground, IMO. If we are in the n=
ice position to find that it gets


adopted a lot, and if there are things =
that need fixing then, we can work on


version 2.

 


Several months ago I started with a yan=
g model for Intent, but never finished.


Maybe this helps with this discussion. =
Note I'm not at all a yang expert, so


please look more at the general structu=
re for now. Not sure if it helps the


discussion...<=
/pre>

 


+--rw intent


   +--rw meta-data


   |  +--rw intent-v=
ersion  uint32   // the version of the intent


specification<=
/pre>

   |  +--rw timestam=
p            uint32 =
  // latest time stamp wins


   |  +--rw administ=
rator        string   // who create=
d that intent


   |  +--rw device?&=
nbsp;             st=
ring   // on which node was it created

   |  +--rw executio=
n-time       uint32   // plus other=
 parameters on when to


execute


   |  +--rw expirati=
on           uint32  =
; // when does this intent cease to be


valid


   |  +--rw feedback=
            &nb=
sp;    // how to let the operator know


about success/failure=



   |  +--rw signatur=
e            ?? =
;      // intent must be signed<=
/font>


   |=



   +--rw scope=



      +--r=
w domains


      &nb=
sp;  +--rw domain* [name]


      &nb=
sp;     +--rw name     &nb=
sp;     string    // is there a fqdn typ=
e defined?


      &nb=
sp;     +--rw meta-data    &nbs=
p;      // general about the domain


      &nb=
sp;     |  +--rw include-subdomains  bool=
ean  // should


sub-domains follow this policy?


      &nb=
sp;     |  +--rw fail-open  boolean =
 // fail-open or fail-closed


policy


      &nb=
sp;     |  +--rw maintenance-window  // w=
hen can critical


operations take place=



      &nb=
sp;     |  +--rw ...    &n=
bsp;            // o=
ther general policies on


domain level

|      &n=
bsp; ...


      &nb=
sp;     |


      &nb=
sp;     +--rw ani-intent

      &nb=
sp;     +-- (contains all policies which the ANI sh=
ould follow)


      &nb=
sp;     |  ex: ACP policies, coordination between =
AN functions,


security policies, etc


      &nb=
sp;     |  for ani-intent, scope is always "a=
ll" (ie all nodes must


process)


      &nb=
sp;     |  ...


...


All of the above MUST be standardised, =
because it refers to the AN


infrastructure itself.


Of the below we need to standardise the=
 structure, but not content for each


function Conceptually, we define Intent=
 for a domain.


...


      &nb=
sp;     |


      &nb=
sp;     +--rw intent-for-autonomic-functions


      &nb=
sp;        +--rw function_1 [name]


      &nb=
sp;        |  +--rw scope =
            &nb=
sp;      // which devices=



in the domain need to listen.


      &nb=
sp;        |  |  +--rw all=
    boolean        // all=
 devices


      &nb=
sp;        |  |  +--rw rol=
e   string         // as =
defined in


certificate

      &nb=
sp;        |  |  +--rw dyn=
amic_role  string  // as determined

dynamically

      &nb=
sp;        |  |  +--rw tag=
    string         /=
/ can be used for


location


      &nb=
sp;        |  +-- function_inte=
nt            // to =
be defined by


function owner=



      &nb=
sp;        |

      &nb=
sp;        +--rw function_2 [name]


      &nb=
sp;        |  +--rw scope =
             &n=
bsp;     // which devices=



in the domain need to listen.


      &nb=
sp;        |  |  +--rw all=
    boolean        // all=
 devices


      &nb=
sp;        |  |  +--rw rol=
e   string         // as =
defined in


certificate

      &nb=
sp;        |  |  +--rw dyn=
amic_role  string  // as determined

dynamically

      &nb=
sp;        |  |  +--rw tag=
    string         /=
/ can be used for


location


      &nb=
sp;        |  +-- function_inte=
nt            // to =
be defined by


function owner=



      &nb=
sp;        |

...


 


If a node receives Intent, it checks:


1) meta data: Mostly for forwarding and=
 processing of Intent


2) domain info: domain specific policy<=
o:p>


3) ANI Intent: Intent for the ANI itsel=
f; every node must be able to process


that


4) For each of the "function"=
 blocks:


   - should I run this functi=
on?


   - do I run this function?<=
o:p>


   - if so, forward the rest =
of that block to the corresponding handler.


 


I think we want a clean architectural s=
eparation between the ANI and the


functions. This way every function can =
do what it wants.


 


Thoughts?

Michael


 


_______________________________________=
________


Anima mailing list


Anima=
@ietf.org


https://www.ietf.org/mailman/listinfo/anima=





 


_______________________________________=
________


Anima mailing list


Anima=
@ietf.org


https://www.ietf.org/mailman/listinfo/anima=



 

=  <= /font>

--

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

 

+33 160 402 636

Route de Villejust | 91620 Nozay | France=

LinkedIn: laurentciavaglia=

 

--_000_7471201223ba4962a66dd5058190ecedXCHRCD006ciscocom_-- From nobody Mon Mar 21 04:00:19 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7AACF12D6FE for ; Mon, 21 Mar 2016 04:00:18 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.522 X-Spam-Level: X-Spam-Status: No, score=-14.522 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xn9OTvMQbgvL for ; Mon, 21 Mar 2016 04:00:16 -0700 (PDT) Received: from alln-iport-7.cisco.com (alln-iport-7.cisco.com [173.37.142.94]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 79A6112D640 for ; Mon, 21 Mar 2016 04:00:16 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=8178; q=dns/txt; s=iport; t=1458558016; x=1459767616; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=mFQh/oDd6ZlBKjGLgmSrnTJCHJR/j62sHtXsYKhLkD4=; b=BwIJ9RsAVlrqkzLKERZvEd6opU5ZAzS926jiQ5dzS5FrGR5Qhp7VBbxv 5aOQX+RUSgkBVL+VYk9iJ4XcYyj/XglZAove22b5LfwTpA6cVqQhG2q+i AFg8zw7SxxMfvQWn2oIESNJtBvTW9JJg2VmtaoYtJienMbhYUebHU/Bo2 s=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D+AQBW0+9W/5NdJa1egzNTcgauUYtNA?= =?us-ascii?q?Q2BbxcKhWwCHIEIOBQBAQEBAQEBZCeEQQEBAQMBAQEBIBE3AxAHBAIBCBEEAQE?= =?us-ascii?q?BAgIjAwICAh8GCxQBCAgBAQQBEgiICgMKCA6ufIodDYRiAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEBAQEBAQEBEQR8hSKERII+ghSCaoJWBYdihVmJazEBjA6BboFshEqIWIcxh1Q?= =?us-ascii?q?BHgEBQoF+gWdqAYkWfgEBAQ?= X-IronPort-AV: E=Sophos;i="5.24,371,1454976000"; d="scan'208";a="251705310" Received: from rcdn-core-11.cisco.com ([173.37.93.147]) by alln-iport-7.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 21 Mar 2016 11:00:15 +0000 Received: from XCH-RCD-010.cisco.com (xch-rcd-010.cisco.com [173.37.102.20]) by rcdn-core-11.cisco.com (8.14.5/8.14.5) with ESMTP id u2LB0FkE028042 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Mon, 21 Mar 2016 11:00:15 GMT Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-010.cisco.com (173.37.102.20) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Mon, 21 Mar 2016 06:00:14 -0500 Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Mon, 21 Mar 2016 06:00:14 -0500 From: "Michael Behringer (mbehring)" To: Brian E Carpenter , "anima@ietf.org" Thread-Topic: [Anima] Intent yang model - some thoughts Thread-Index: AdGA84Aiy7EwHtVtTUWSCghLdxIQCwCFPbeAABYDdBA= Date: Mon, 21 Mar 2016 11:00:14 +0000 Message-ID: <144b00dacf2442c89761389e85a68f8b@XCH-RCD-006.cisco.com> References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <56EEF9A7.90709@gmail.com> In-Reply-To: <56EEF9A7.90709@gmail.com> Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.55.238.139] Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 MIME-Version: 1.0 Archived-At: Subject: Re: [Anima] Intent yang model - some thoughts X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 11:00:18 -0000 PiAtLS0tLU9yaWdpbmFsIE1lc3NhZ2UtLS0tLQ0KPiBGcm9tOiBCcmlhbiBFIENhcnBlbnRlciBb bWFpbHRvOmJyaWFuLmUuY2FycGVudGVyQGdtYWlsLmNvbV0NCj4gU2VudDogMjAgTWFyY2ggMjAx NiAyMDoyOA0KPiBUbzogTWljaGFlbCBCZWhyaW5nZXIgKG1iZWhyaW5nKSA8bWJlaHJpbmdAY2lz Y28uY29tPjsgYW5pbWFAaWV0Zi5vcmcNCj4gU3ViamVjdDogUmU6IFtBbmltYV0gSW50ZW50IHlh bmcgbW9kZWwgLSBzb21lIHRob3VnaHRzDQo+IA0KPiBIaSBNaWNoYWVsLA0KPiANCj4gQXQgYSBn bGFuY2UsIGZyb20gc29tZW9uZSB3aG8gaXMgYSB5YW5nIGJlZ2lubmVyLCB0aGlzIGxvb2tzIHBs YXVzaWJsZSwNCj4gYWx0aG91Z2ggSSB0aGluayB3ZSBuZWVkIHNvbWUgYW5hbHlzaXMgb2YgdGhl IHB1cnBvc2Ugb2YgbWFueSBvZiB0aGUgaXRlbXMuDQo+IEp1c3QgdG8gcGljayBhIHJhbmRvbSBl eGFtcGxlLCBjb25zaWRlciAibWFpbnRlbmFuY2Utd2luZG93Ii4NCj4gIC0gaXMgdGhpcyBmaXhl ZCBvciBkb2VzIGl0IGNoYW5nZT8NCj4gIC0gaWYgaXQgY2hhbmdlcywgaG93IGxvbmcgaW4gYWR2 YW5jZSBkb2VzIHRoZSBuZXcgaW50ZW50IGdldCBzZW50Pw0KPiAgLSB3aGF0IGFyZSB0aGUgc2Vt YW50aWNzIChtdXN0IGFsbCBBU0FzIHN0b3Agb3BlcmF0aW9uLCBldGMuKT8NCj4gIC0gd2hhdCBo YXBwZW5zIGlmIG1haW50ZW5hbmNlIGVuZHMgZWFybHk/DQo+ICAtIHdoYXQgaGFwcGVucyBpZiBt YWludGVuYW5jZSBlbmRzIGxhdGU/DQo+IEluIG90aGVyIHdvcmRzIHRoZXNlIGl0ZW1zIGFyZSBu b3Qgc2VsZi1kZWZpbmluZzsgZWFjaCBvbmUgbmVlZHMgd29yay4NCg0KQ29uc2lkZXIgbXkgcHJv cG9zYWwgbW9yZSBvZiBhIGJyYWluc3Rvcm1pbmcgc2Vzc2lvbiB0aGFuIGEgcmVhZHktdG8tdXNl IG1vZGVsLiBJJ20gZm9jdXNpbmcganVzdCBvbiB0aGUgc3RydWN0dXJlLCBhbmQgdGhlIGZpZWxk cyBhcmUgZXhhbXBsZXMgb2Ygd2hhdCBjb3VsZCBiZSB0aGVyZS4gDQogDQo+IEEgYmVnaW5uZXIn cyBxdWVzdGlvbjogaG93IGlzIHlhbmcgc2VyaWFsaXNlZCBvbiB0aGUgd2lyZT8NCj4gDQo+IEFu ZCBhbm90aGVyOiBjYW4gd2Ugc2VuZCBwYXJ0aWFsIHlhbmcsIGkuZS4gb25seSBzZW5kIHRoZSBi aXRzIHRoYXQgY2hhbmdlZD8NCg0KTm90IGEgWUFORyBleHBlcnQgZWl0aGVyLCBhbmQgYXMgbWVu dGlvbmVkIGJ5IExhdXJlbnQsIHRoZSBlbmNvZGluZyBpcyBhIHNlY29uZGFyeSBxdWVzdGlvbi4g DQoNCk1pY2hhZWwNCg0KPiBSZWdhcmRzDQo+ICAgIEJyaWFuDQo+IA0KPiANCj4gT24gMTgvMDMv MjAxNiAyMjo1MiwgTWljaGFlbCBCZWhyaW5nZXIgKG1iZWhyaW5nKSB3cm90ZToNCj4gPiBXUlQg dGhlIGN1cnJlbnQgZGlzY3Vzc2lvbiBvbiB3aGF0IHNob3VsZCBiZSBpbiBJbnRlbnQgYW5kIHdo YXQNCj4gc2hvdWxkbid0LCBhcyBtZW50aW9uZWQgYmVmb3JlLCBJIHRoaW5rIHdlIHNob3VsZCBs ZWF2ZSB0aGlzIHRvIHRoZSBvd25lciBvZg0KPiBhbiBhdXRvbm9taWMgZnVuY3Rpb24gdG8gZGVj aWRlLiBUaGlzIHJpc2tzICJhYnVzZSIgaW4gYSB3YXkgdGhhdCB3YXNuJ3QNCj4gZm9yZXNlZW4s IGJ1dCBpdCBnaXZlcyBtYXhpbXVtIGFnaWxpdHksIHdoaWNoIGlzIHdoYXQgd2UgbmVlZCB0byBn ZXQgQU5JTUENCj4gb2ZmIHRoZSBncm91bmQsIElNTy4gSWYgd2UgYXJlIGluIHRoZSBuaWNlIHBv c2l0aW9uIHRvIGZpbmQgdGhhdCBpdCBnZXRzIGFkb3B0ZWQNCj4gYSBsb3QsIGFuZCBpZiB0aGVy ZSBhcmUgdGhpbmdzIHRoYXQgbmVlZCBmaXhpbmcgdGhlbiwgd2UgY2FuIHdvcmsgb24gdmVyc2lv biAyLg0KPiA+DQo+ID4gU2V2ZXJhbCBtb250aHMgYWdvIEkgc3RhcnRlZCB3aXRoIGEgeWFuZyBt b2RlbCBmb3IgSW50ZW50LCBidXQgbmV2ZXINCj4gZmluaXNoZWQuIE1heWJlIHRoaXMgaGVscHMg d2l0aCB0aGlzIGRpc2N1c3Npb24uIE5vdGUgSSdtIG5vdCBhdCBhbGwgYSB5YW5nDQo+IGV4cGVy dCwgc28gcGxlYXNlIGxvb2sgbW9yZSBhdCB0aGUgZ2VuZXJhbCBzdHJ1Y3R1cmUgZm9yIG5vdy4g Tm90IHN1cmUgaWYgaXQNCj4gaGVscHMgdGhlIGRpc2N1c3Npb24uLi4NCj4gPg0KPiA+ICstLXJ3 IGludGVudA0KPiA+ICAgICstLXJ3IG1ldGEtZGF0YQ0KPiA+ICAgIHwgICstLXJ3IGludGVudC12 ZXJzaW9uICB1aW50MzIgICAvLyB0aGUgdmVyc2lvbiBvZiB0aGUgaW50ZW50IHNwZWNpZmljYXRp b24NCj4gPiAgICB8ICArLS1ydyB0aW1lc3RhbXAgCQl1aW50MzIgICAvLyBsYXRlc3QgdGltZSBz dGFtcCB3aW5zDQo+ID4gICAgfCAgKy0tcncgYWRtaW5pc3RyYXRvciAJc3RyaW5nICAgLy8gd2hv IGNyZWF0ZWQgdGhhdCBpbnRlbnQNCj4gPiAgICB8ICArLS1ydyBkZXZpY2U/ICAgCQlzdHJpbmcg ICAvLyBvbiB3aGljaCBub2RlIHdhcyBpdCBjcmVhdGVkDQo+ID4gICAgfCAgKy0tcncgZXhlY3V0 aW9uLXRpbWUJdWludDMyICAgLy8gcGx1cyBvdGhlciBwYXJhbWV0ZXJzIG9uIHdoZW4gdG8NCj4g ZXhlY3V0ZQ0KPiA+ICAgIHwgICstLXJ3IGV4cGlyYXRpb24gICAgCXVpbnQzMiAgIC8vIHdoZW4g ZG9lcyB0aGlzIGludGVudCBjZWFzZSB0byBiZQ0KPiB2YWxpZA0KPiA+ICAgIHwgICstLXJ3IGZl ZWRiYWNrICAgICAgICAgICAgICAgICAvLyBob3cgdG8gbGV0IHRoZSBvcGVyYXRvciBrbm93IGFi b3V0DQo+IHN1Y2Nlc3MvZmFpbHVyZQ0KPiA+ICAgIHwgICstLXJ3IHNpZ25hdHVyZQkJPz8gICAg ICAgLy8gaW50ZW50IG11c3QgYmUgc2lnbmVkDQo+ID4gICAgfA0KPiA+ICAgICstLXJ3IHNjb3Bl DQo+ID4gICAgICAgKy0tcncgZG9tYWlucw0KPiA+ICAgICAgICAgICstLXJ3IGRvbWFpbiogW25h bWVdDQo+ID4gICAgICAgICAgICAgKy0tcncgbmFtZQkJc3RyaW5nICAgIC8vIGlzIHRoZXJlIGEg ZnFkbiB0eXBlIGRlZmluZWQ/DQo+ID4gICAgICAgICAgICAgKy0tcncgbWV0YS1kYXRhICAgICAg ICAgICAvLyBnZW5lcmFsIGFib3V0IHRoZSBkb21haW4NCj4gPiAgICAgICAgICAgICB8ICArLS1y dyBpbmNsdWRlLXN1YmRvbWFpbnMJYm9vbGVhbiAgLy8gc2hvdWxkIHN1Yi1kb21haW5zDQo+IGZv bGxvdyB0aGlzIHBvbGljeT8NCj4gPiAgICAgICAgICAgICB8ICArLS1ydyBmYWlsLW9wZW4gIGJv b2xlYW4gIC8vIGZhaWwtb3BlbiBvciBmYWlsLWNsb3NlZCBwb2xpY3kNCj4gPiAgICAgICAgICAg ICB8ICArLS1ydyBtYWludGVuYW5jZS13aW5kb3cgIC8vIHdoZW4gY2FuIGNyaXRpY2FsIG9wZXJh dGlvbnMgdGFrZQ0KPiBwbGFjZQ0KPiA+ICAgICAgICAgICAgIHwgICstLXJ3IC4uLiAgICAgICAg ICAgICAgICAgLy8gb3RoZXIgZ2VuZXJhbCBwb2xpY2llcyBvbiBkb21haW4gbGV2ZWwNCj4gPiB8 ICAgICAgICAuLi4NCj4gPiAgICAgICAgICAgICB8DQo+ID4gICAgICAgICAgICAgKy0tcncgYW5p LWludGVudA0KPiA+ICAgICAgICAgICAgICstLSAoY29udGFpbnMgYWxsIHBvbGljaWVzIHdoaWNo IHRoZSBBTkkgc2hvdWxkIGZvbGxvdykNCj4gPiAgICAgICAgICAgICB8ICBleDogQUNQIHBvbGlj aWVzLCBjb29yZGluYXRpb24gYmV0d2VlbiBBTiBmdW5jdGlvbnMsIHNlY3VyaXR5DQo+IHBvbGlj aWVzLCBldGMNCj4gPiAgICAgICAgICAgICB8ICBmb3IgYW5pLWludGVudCwgc2NvcGUgaXMgYWx3 YXlzICJhbGwiIChpZSBhbGwgbm9kZXMgbXVzdCBwcm9jZXNzKQ0KPiA+ICAgICAgICAgICAgIHwg IC4uLg0KPiA+IC4uLg0KPiA+IEFsbCBvZiB0aGUgYWJvdmUgTVVTVCBiZSBzdGFuZGFyZGlzZWQs IGJlY2F1c2UgaXQgcmVmZXJzIHRvIHRoZSBBTg0KPiBpbmZyYXN0cnVjdHVyZSBpdHNlbGYuDQo+ ID4gT2YgdGhlIGJlbG93IHdlIG5lZWQgdG8gc3RhbmRhcmRpc2UgdGhlIHN0cnVjdHVyZSwgYnV0 IG5vdCBjb250ZW50IGZvcg0KPiA+IGVhY2ggZnVuY3Rpb24gQ29uY2VwdHVhbGx5LCB3ZSBkZWZp bmUgSW50ZW50IGZvciBhIGRvbWFpbi4NCj4gPiAuLi4NCj4gPiAgICAgICAgICAgICB8DQo+ID4g ICAgICAgICAgICAgKy0tcncgaW50ZW50LWZvci1hdXRvbm9taWMtZnVuY3Rpb25zDQo+ID4gICAg ICAgICAgICAgICAgKy0tcncgZnVuY3Rpb25fMSBbbmFtZV0NCj4gPiAgICAgICAgICAgICAgICB8 ICArLS1ydyBzY29wZSAgICAgICAgICAgICAgICAgICAgLy8gd2hpY2ggZGV2aWNlcyBpbiB0aGUg ZG9tYWluIG5lZWQgdG8NCj4gbGlzdGVuLg0KPiA+ICAgICAgICAgICAgICAgIHwgIHwgICstLXJ3 IGFsbCAgICBib29sZWFuICAgICAgICAvLyBhbGwgZGV2aWNlcw0KPiA+ICAgICAgICAgICAgICAg IHwgIHwgICstLXJ3IHJvbGUgICBzdHJpbmcgICAgICAgICAvLyBhcyBkZWZpbmVkIGluIGNlcnRp ZmljYXRlDQo+ID4gICAgICAgICAgICAgICAgfCAgfCAgKy0tcncgZHluYW1pY19yb2xlICBzdHJp bmcgIC8vIGFzIGRldGVybWluZWQgZHluYW1pY2FsbHkNCj4gPiAgICAgICAgICAgICAgICB8ICB8 ICArLS1ydyB0YWcgICAgc3RyaW5nICAgICAgICAgLy8gY2FuIGJlIHVzZWQgZm9yIGxvY2F0aW9u DQo+ID4gICAgICAgICAgICAgICAgfCAgKy0tIGZ1bmN0aW9uX2ludGVudCAgICAgICAgICAgIC8v IHRvIGJlIGRlZmluZWQgYnkgZnVuY3Rpb24gb3duZXINCj4gPiAgICAgICAgICAgICAgICB8DQo+ ID4gICAgICAgICAgICAgICAgKy0tcncgZnVuY3Rpb25fMiBbbmFtZV0NCj4gPiAgICAgICAgICAg ICAgICB8ICArLS1ydyBzY29wZSAgICAgICAgICAgICAgICAgICAgLy8gd2hpY2ggZGV2aWNlcyBp biB0aGUgZG9tYWluIG5lZWQgdG8NCj4gbGlzdGVuLg0KPiA+ICAgICAgICAgICAgICAgIHwgIHwg ICstLXJ3IGFsbCAgICBib29sZWFuICAgICAgICAvLyBhbGwgZGV2aWNlcw0KPiA+ICAgICAgICAg ICAgICAgIHwgIHwgICstLXJ3IHJvbGUgICBzdHJpbmcgICAgICAgICAvLyBhcyBkZWZpbmVkIGlu IGNlcnRpZmljYXRlDQo+ID4gICAgICAgICAgICAgICAgfCAgfCAgKy0tcncgZHluYW1pY19yb2xl ICBzdHJpbmcgIC8vIGFzIGRldGVybWluZWQgZHluYW1pY2FsbHkNCj4gPiAgICAgICAgICAgICAg ICB8ICB8ICArLS1ydyB0YWcgICAgc3RyaW5nICAgICAgICAgLy8gY2FuIGJlIHVzZWQgZm9yIGxv Y2F0aW9uDQo+ID4gICAgICAgICAgICAgICAgfCAgKy0tIGZ1bmN0aW9uX2ludGVudCAgICAgICAg ICAgIC8vIHRvIGJlIGRlZmluZWQgYnkgZnVuY3Rpb24gb3duZXINCj4gPiAgICAgICAgICAgICAg ICB8DQo+ID4gLi4uDQo+ID4NCj4gPiBJZiBhIG5vZGUgcmVjZWl2ZXMgSW50ZW50LCBpdCBjaGVj a3M6DQo+ID4gMSkgbWV0YSBkYXRhOiBNb3N0bHkgZm9yIGZvcndhcmRpbmcgYW5kIHByb2Nlc3Np bmcgb2YgSW50ZW50DQo+ID4gMikgZG9tYWluIGluZm86IGRvbWFpbiBzcGVjaWZpYyBwb2xpY3kN Cj4gPiAzKSBBTkkgSW50ZW50OiBJbnRlbnQgZm9yIHRoZSBBTkkgaXRzZWxmOyBldmVyeSBub2Rl IG11c3QgYmUgYWJsZSB0bw0KPiA+IHByb2Nlc3MgdGhhdA0KPiA+IDQpIEZvciBlYWNoIG9mIHRo ZSAiZnVuY3Rpb24iIGJsb2NrczoNCj4gPiAgICAtIHNob3VsZCBJIHJ1biB0aGlzIGZ1bmN0aW9u Pw0KPiA+ICAgIC0gZG8gSSBydW4gdGhpcyBmdW5jdGlvbj8NCj4gPiAgICAtIGlmIHNvLCBmb3J3 YXJkIHRoZSByZXN0IG9mIHRoYXQgYmxvY2sgdG8gdGhlIGNvcnJlc3BvbmRpbmcgaGFuZGxlci4N Cj4gPg0KPiA+IEkgdGhpbmsgd2Ugd2FudCBhIGNsZWFuIGFyY2hpdGVjdHVyYWwgc2VwYXJhdGlv biBiZXR3ZWVuIHRoZSBBTkkgYW5kIHRoZQ0KPiBmdW5jdGlvbnMuIFRoaXMgd2F5IGV2ZXJ5IGZ1 bmN0aW9uIGNhbiBkbyB3aGF0IGl0IHdhbnRzLg0KPiA+DQo+ID4gVGhvdWdodHM/DQo+ID4gTWlj aGFlbA0KPiA+DQo+ID4gX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX18NCj4gPiBBbmltYSBtYWlsaW5nIGxpc3QNCj4gPiBBbmltYUBpZXRmLm9yZw0KPiA+IGh0 dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vYW5pbWENCj4gPg0K From nobody Mon Mar 21 04:34:01 2016 Return-Path: X-Original-To: anima@ietfa.amsl.com Delivered-To: anima@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EC97312D677 for ; Mon, 21 Mar 2016 04:33:59 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.92 X-Spam-Level: X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3HzwvixZwy_5 for ; Mon, 21 Mar 2016 04:33:56 -0700 (PDT) Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D184412D73E for ; Mon, 21 Mar 2016 04:33:41 -0700 (PDT) Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id D1210661C6E71; Mon, 21 Mar 2016 11:33:36 +0000 (GMT) Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2LBXcqH002463 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 21 Mar 2016 11:33:39 GMT Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2LBX133026689 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 21 Mar 2016 12:33:35 +0100 Received: from [172.27.205.106] (135.239.27.38) by FR711WXCHHUB02.zeu.alcatel-lucent.com (135.239.2.112) with Microsoft SMTP Server (TLS) id 14.3.195.1; Mon, 21 Mar 2016 12:33:14 +0100 To: "EXT Michael Behringer (mbehring)" , Brian E Carpenter , Duzongpeng , "anima@ietf.org" References: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> <56E9A8B2.8070504@gmail.com> <56EEF72A.5050309@gmail.com> <782ad410ef844630a942c27f6279a49f@XCH-RCD-006.cisco.com> From: Laurent Ciavaglia Organization: Bell Labs Message-ID: <56EFDBF9.9020206@alcatel-lucent.com> Date: Mon, 21 Mar 2016 12:33:13 +0100 User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0 MIME-Version: 1.0 In-Reply-To: <782ad410ef844630a942c27f6279a49f@XCH-RCD-006.cisco.com> Content-Type: multipart/alternative; boundary="------------090100060506080504000207" X-Originating-IP: [135.239.27.38] Archived-At: Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , Sheng Jiang Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent X-BeenThere: anima@ietf.org X-Mailman-Version: 2.1.17 Precedence: list List-Id: Autonomic Networking Integrated Model and Approach List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 21 Mar 2016 11:34:00 -0000 --------------090100060506080504000207 Content-Type: text/plain; charset="utf-8"; format=flowed Content-Transfer-Encoding: 8bit Hello, Please see below for my comments. On 21/03/2016 10:31, EXT Michael Behringer (mbehring) wrote: >> -----Original Message----- >> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com] >> Sent: 20 March 2016 20:17 >> To: Duzongpeng ; anima@ietf.org >> Cc: Laurent Ciavaglia ; Jéferson Campos Nobre >> ; Sheng Jiang ; Michael >> Behringer (mbehring) >> Subject: Re: [Anima] Does Autonomic network need some intervention >> beyond intent >> >> Hi Zongpeng, >> >>> Should these ASA parameters be part of ANIMA's work items just as >> intent? >> >> As Michael's yang outline showed, there will be network-wide aspects to >> intent and also ASA-specific aspects. I think that in practice there will be >> simple parameters in the ASA-specific parts, so (whether we use yang or >> some other syntax) I see no advantage in having a separate mechanism for >> the parameters - in a sense we never need to answer the question, we >> simply treat all the information in the same way, intent or parameter. > Agree so far. If we refer to so-called policy-continuum, a policy is defined and refined from a high-level/abstract point, usually with network-wide considerations, down to low-level device parameters. This is thus completely part of the policy based management approach. What we need in ANIMA is to define (re-use), and agree on the levels of the hierarchy and how we call (again re-use terms) the levels and elements in each level. > >> Maybe we should think about levels of specialization. Top level, network >> intent. Next level, intent for all ASAs supporting a given objective. >> Those two levels would be flooded by GRASP. Next level, intent for a single >> ASA. That would be transmitted by unicast GRASP synchronization. Comment to Brian's point: usually we write (multiple) policies based on the needs/objectives of the network operations. It is the resolution/processing of the policies that determine the set of target entities. > Here I don't have the same view. The first two levels we agree on. Anything that's for a specific node doesn't qualify as "Intent" in my view. (Ignore that the flooding mechanism may contain unicast transmissions as part of the overall scheme). > > If Intent is a high-level policy (which I think we all agree on?) then I can think of two more "things" that are NOT Intent, to add to my list of "not-Intent": > > - anything that's for a specific node Agree, but we should be careful in the interpretation: -By nature, a policy targets a group of entities/role/function. -It is the evaluation/resolution of the policies that determines the set of target entities. This set may contain only a single "node"/entity. > - anything that's negotiated dynamically. (Intent is always originating from something non-technical - it's a policy). I don't really understand what you mean by "negotiating" and "dynamically"... Which are the entities negotiating? What is dynamic important? (how do you negotiate something statically?) With "non-technical" do you have in mind something like business-level policy? How would you qualify a policy written as "secure all communications"? or "maximize link usage"...? --> my guess is that we need to define more than the "ANIMA" intent-policy, and include at least the "levels" in the hierarchy to understand better where the intent-policy starts and stops (and then is translated into something else) > - anything that needs to change frequently (high-level policy is normally pretty stable) I would tend to agree, but would slightly change the angle, because with your definition the issue is how to set the cursor on frequency. A policy is a statement or goal is not time dependent. As long as the policy is "active", it should be evaluated (here time/context might come into consideration). Whether the policy is activated/deactivated, changed, erased... is not dependent on the system, but on the user. > - anything that's not a "policy", ie doesn't come from a non-technical reasoning. (that's not a good way to define it, but can't think of a better one right now) Let's try to dig in... If I guess right, one example would be ASAs converging on the settings of common parameters...? Again, if we define a hierarchy this will clear out what is a policy and where it starts/ends, and the other things we manipulate. However, one note in the context of ANIMA / autonomic networking: we consider an approach where ASAs exchange "policy" among each others in a peer-to-peer manner... Maybe we have to clarify if this is indeed policy or something else. > > I think you're suggesting an optimisation of Intent *distribution*, Brian, right? Where, if I know only a few nodes run a certain function, I just send that particular Intent to those nodes only? (is that what you're thinking of?) > > This would certainly be tempting, but I think it's a slippery slope, and would quickly get into conflict with current policy models. I'd prefer if we keep "our" Intent as "must be flooded everywhere". If it is really high level guidance, the optimization shouldn't be needed. Said differently: If we start pushing huge data models, maybe we're on the wrong track in the first place? I have mixed feeling over this one. I can buy the "flood everywhere" for the simplicity (but how important is it compared to other criteria...?). I see at least two issues: 1. flooding a "big" intent file (cf your current tentative example) may be bandwidth hungry, if one considers how frequently such an intent file could be distributed (ASA stop/start, device join/leave, policy change/update...) - compared to distributing only what is required / where it is required. 2. security: flooding the entire file reveal information to target (possibly undesirable/rogue) entities that do not need it (and thus may mis-use it). I think the first step would be again to agree on our definitions, scope and hierarchy. Best regards, Laurent. > > Michael > >> Regards >> Brian >> >> On 18/03/2016 21:33, Duzongpeng wrote: >>> Hi, Brian >>> >>> Thanks for the response. I think we have agreed that "prefix- >> management ID" can be an Autonomic intent. >>> However, some ASA parameters may appear to be more static, and >> perhaps can fulfill the statement below: >>> 1. Network level parameters, need to be distributed by GRASP and >> ACP >>> 2. Relatively static compared to intent, perhaps only needed to be >> configured once >>> 3. Unavoidable in current network environment, mostly for >> establishing the network infrastructure >>> 4. Rarely need coordinations with others parameters, usually is >>> closed related to one specific autonomic function >>> >>> Some examples can be >>> What kind of IGP the Autonomic network domain should use (such as >> IS-IS, OSPF); >>> What kind of transport layer network technology the Autonomic >> network >>> domain should use (such as MPLS, IP) >>> >>> Should these ASA parameters be part of ANIMA's work items just as >> intent? >>> Best Regards >>> Zongpeng Du >>> >>> -----Original Message----- >>> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com] >>> Sent: Thursday, March 17, 2016 2:41 AM >>> To: Duzongpeng; Michael Behringer (mbehring); anima@ietf.org >>> Cc: Laurent Ciavaglia; Jéferson Campos Nobre; Sheng Jiang >>> Subject: Re: [Anima] Does Autonomic network need some intervention >>> beyond intent >>> >>> Yes, I think that whatever the "philosophy" of intent is, we should make >> sure that the implementation is pragmatic and that we can adapt to >> operator's requirements. >>> Regards >>> Brian >>> >>> On 17/03/2016 01:41, Duzongpeng wrote: >>>> Hi, Michael, >>>> >>>> Thanks for the reply. I can understand that intent is not in the current >> milestones, and you are working on several milestones drafts now. >>>> You are welcome to contribute to the intent draft at any time, and >> thanks in advance. >>>> According to your statements, in my understanding, we can keep >> regarding the "intent" example in "prefix-management ID" as an Autonomic >> intent. >>>> It's semantics are just like "all nodes of type x do this; type y does >> that", so it is a high level policy. >>>> It really helps a lot. Thanks. >>>> >>>> I generally agree with your statement, while some other personal >> understandings exist (shown below). But perhaps they can be discussed >> later. >>>> Best Regards >>>> Zongpeng Du >>>> >>>> From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com] >>>> Sent: Wednesday, March 16, 2016 4:17 PM >>>> To: Duzongpeng; anima@ietf.org >>>> Cc: Laurent Ciavaglia; Jéferson Campos Nobre; Sheng Jiang >>>> Subject: RE: [Anima] Does Autonomic network need some intervention >>>> beyond intent >>>> >>>> Hi Zongpeng, >>>> >>>> Yes, this is a recurring question and topic. Let me give my view on two >> levels: >>>> First, ANIMA isn't the only thing "influencing" or "steering" a network, at >> least not initially. There is still traditional config, plus all the existing and >> emerging SDN models. My view has always been that ANIMA sets a "default" >> policy in a network, and all other means are more specific and can override >> intent. This is described in RFC7575. Main point here: ANIMA Intent isn't the >> only tool in a network. >>>> [duzongpeng] I agree that intent can coexist with traditional config. >>>> I generally agree with the policy priorities, but I have some other >> concerns. >>>> Some default policies/parameters set through "intent" perhaps are >> fundamental for the uplayer services. >>>> It should be prevented that human operator uses some CLIs to change >> them by mistake, and the uplayer services are interrupted. >>>> At least, some warning should be reported to the human operator. >>>> [duzongpeng] >>>> >>>> Intent could perfectly well cover a high level policy such as "all nodes of >> type x do this; type y does that". (Which I think is what you mention below.) >> When we started with Intent, the vision was to keep it very high level, more >> along the line "do the right thing" :) It has become clear that in today's >> networking, people will want to push more "parameters" than "policy", and I >> think we'll just have to acknowledge that. >>>> Personally I think it'll pan out like this: Intent will have sections per >> autonomic function. An Intent parser will validate overall correctness >> (signature, etc), then segment the Intent into its bits: One part will be the >> autonomic fabric itself, then there will be bits for each autonomic function. >>>> So the Intent parser on a node will grab intent, see there is some Intent >> for autonomic function A, will see whether this is running locally, and if so, it'll >> just pass that piece of Intent to that function. >>>> In this way, a function owner defines both the intent itself, and how it is >> consumed by his/her function. >>>> [duzongpeng]Agree.[duzongpeng] >>>> >>>> The downside is that there is no "control" an//d people can do what they >> want, in the worst case even push config, which is clearly NOT what Intent is >> meant to do. I'd hope that people don't start writing Intent like "node 17: >> here is your CLI; node 23: here is your CLI"; node 37: here is your CLI". But >> what I describe above COULD be abused in this way. RFC7575 says this is NOT >> Intent. >>>> [duzongpeng]Agree. It is hard to give an exact boundary. >>>> However, I think we have two choices here. >>>> The first one is to define a policy first in every ASA, but with some open >> parameters that can be configured. What we need to do is just to distribute >> the network-level parameters for that function. This way is suitable to some >> relatively static functions. >>>> The second is more advanced, in the ideal case, the human operator can >> communicate with network using natural language or something similar. This >> way is more flexible, but harder to realize. >>>> [duzongpeng] >>>> >>>> In other words: Let's not be too prescriptive, and hope that folks will do >> more or less reasonable things. >>>> [duzongpeng]Agree.[duzongpeng] >>>> >>>> I've been meaning to comment on the Intent drafts and possibly >> contribute, I'm just struggling with priorities at the moment (as you can see >> by my relative silence over the past weeks). >>>> Michael >>>> >>>> >>>> >>>> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng >>>> Sent: 16 March 2016 02:38 >>>> To: anima@ietf.org >>>> Cc: Laurent Ciavaglia >>>> >; >>>> Jéferson Campos Nobre >>>> >; Sheng Jiang >>>> > >>>> Subject: [Anima] Does Autonomic network need some intervention >> beyond >>>> intent >>>> >>>> Hello, everyone >>>> >>>> I understand that Autonomic network should be able to work with as >> less configuration as possible. In the ideal case, an Autonomic network >> should be able to work well while only intent is needed from human >> operators. >>>> However, I wonder whether Autonomic network may need some >> intervention from human operators through some low-level policies beyond >> intents. >>>> For example, in the https://tools.ietf.org/html/draft-ietf-anima- >> prefix-management-00, it is suggested that the prefix lengths for the CSG, >> ASG, RSG (different roles in IP RAN) can be assigned as an "intent". These >> configuration parameters need to be distributed in the autonomic domain to >> influence the detail configurations on each autonomic node. >>>> As intent is described as an abstract, declarative, high-level policy >> used to operate an autonomic domain, such as an enterprise network. >> Perhaps we can rename the "intent" in the "prefix-management ID" to some >> other words. >>>> Is "ASA parameters" ok here? >>>> >>>> I think some characteristics for this kind of low-level policies are >>>> >>>> 1. Network level parameters, need to be distributed by GRASP and >> ACP >>>> 2. Relatively static compared to intent, perhaps only needed to be >> configured once >>>> 3. Unavoidable in current network environment, mostly for >> establishing the network infrastructure >>>> 4. Rarely need coordinations with others parameters, usually is closed >> related to one specific autonomic function >>>> Welcome for comment. Your suggestions will be appreciated. >>>> >>>> >>>> Best Regards >>>> Zongpeng Du >>>> >>>> >>>> >>>> _______________________________________________ >>>> Anima mailing list >>>> Anima@ietf.org >>>> https://www.ietf.org/mailman/listinfo/anima >>>> > _______________________________________________ > Anima mailing list > Anima@ietf.org > https://www.ietf.org/mailman/listinfo/anima -- Laurent Ciavaglia Senior Research Manager Bell Labs, Nokia +33 160 402 636 Route de Villejust | 91620 Nozay | France LinkedIn: laurentciavaglia --------------090100060506080504000207 Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: 8bit Hello,

Please see below for my comments.



On 21/03/2016 10:31, EXT Michael Behringer (mbehring) wrote:
-----Original Message-----
From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
Sent: 20 March 2016 20:17
To: Duzongpeng <duzongpeng@huawei.com>; anima@ietf.org
Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; Jéferson Campos Nobre
<jcnobre@inf.ufrgs.br>; Sheng Jiang <jiangsheng@huawei.com>; Michael
Behringer (mbehring) <mbehring@cisco.com>
Subject: Re: [Anima] Does Autonomic network need some intervention
beyond intent

Hi Zongpeng,


Should these ASA parameters be part of ANIMA's work items just as

intent?

As Michael's yang outline showed, there will be network-wide aspects to
intent and also ASA-specific aspects. I think that in practice there will be
simple parameters in the ASA-specific parts, so (whether we use yang or
some other syntax) I see no advantage in having a separate mechanism for
the parameters - in a sense we never need to answer the question, we
simply treat all the information in the same way, intent or parameter.

Agree so far.
If we refer to so-called policy-continuum, a policy is defined and refined from a high-level/abstract point, usually with network-wide considerations, down to low-level device parameters.
This is thus completely part of the policy based management approach.
What we need in ANIMA is to define (re-use), and agree on the levels of the hierarchy and how we call (again re-use terms) the levels and elements in each level.

 

Maybe we should think about levels of specialization. Top level, network
intent. Next level, intent for all ASAs supporting a given objective.
Those two levels would be flooded by GRASP. Next level, intent for a single
ASA. That would be transmitted by unicast GRASP synchronization.
Comment to Brian's point: usually we write (multiple) policies based on the needs/objectives of the network operations. It is the resolution/processing of the policies that determine the set of target entities.

Here I don't have the same view. The first two levels we agree on. Anything that's for a specific node doesn't qualify as "Intent" in my view. (Ignore that the flooding mechanism may contain unicast transmissions as part of the overall scheme). 

If Intent is a high-level policy (which I think we all agree on?) then I can think of two more "things" that are NOT Intent, to add to my list of "not-Intent": 

- anything that's for a specific node
Agree, but we should be careful in the interpretation:
    -By nature, a policy targets a group of entities/role/function.
    -It is the evaluation/resolution of the policies that determines the set of target entities. This set may contain only a single "node"/entity. 

- anything that's negotiated dynamically. (Intent is always originating from something non-technical - it's a policy).
I don't really understand what you mean by "negotiating" and "dynamically"...
Which are the entities negotiating?
What is dynamic important? (how do you negotiate something statically?)

With "non-technical" do you have in mind something like business-level policy?
How would you qualify a policy written as "secure all communications"? or "maximize link usage"...?

--> my guess is that we need to define more than the "ANIMA" intent-policy, and include at least the "levels" in the hierarchy to understand better where the intent-policy starts and stops (and then is translated into something else)

- anything that needs to change frequently (high-level policy is normally pretty stable)
I would tend to agree, but would slightly change the angle, because with your definition the issue is how to set the cursor on frequency.
A policy is a statement or goal is not time dependent. As long as the policy is "active", it should be evaluated (here time/context might come into consideration).
Whether the policy is activated/deactivated, changed, erased... is not dependent on the system, but on the user.

- anything that's not a "policy", ie doesn't come from a non-technical reasoning. (that's not a good way to define it, but can't think of a better one right now)
Let's try to dig in...
If I guess right, one example would be ASAs converging on the settings of common parameters...?
Again, if we define a hierarchy this will clear out what is a policy and where it starts/ends, and the other things we manipulate.
However, one note in the context of ANIMA / autonomic networking: we consider an approach where ASAs exchange "policy" among each others in a peer-to-peer manner... Maybe we have to clarify if this is indeed policy or something else.


I think you're suggesting an optimisation of Intent *distribution*, Brian, right? Where, if I know only a few nodes run a certain function, I just send that particular Intent to those nodes only? (is that what you're thinking of?) 

This would certainly be tempting, but I think it's a slippery slope, and would quickly get into conflict with current policy models. I'd prefer if we keep "our" Intent as "must be flooded everywhere". If it is really high level guidance, the optimization shouldn't be needed. Said differently: If we start pushing huge data models, maybe we're on the wrong track in the first place?
I have mixed feeling over this one.
I can buy the "flood everywhere" for the simplicity (but how important is it compared to other criteria...?).
I see at least two issues:
    1. flooding a "big" intent file (cf your current tentative example) may be bandwidth hungry, if one considers how frequently such an intent file could be distributed (ASA stop/start, device join/leave, policy change/update...)
     - compared to distributing only what is required / where it is required.
    2. security: flooding the entire file reveal information to target (possibly undesirable/rogue) entities that do not need it (and thus may mis-use it).

I think the first step would be again to agree on our definitions, scope and hierarchy.

Best regards, Laurent.


Michael


Regards
   Brian

On 18/03/2016 21:33, Duzongpeng wrote:

Hi, Brian

	Thanks for the response. I think we have agreed that "prefix-

management ID" can be an Autonomic intent.

	However, some ASA parameters may appear to be more static, and

perhaps can fulfill the statement below:

	1. Network level parameters, need to be distributed by GRASP and

ACP

	2. Relatively static compared to intent, perhaps only needed to be

configured once

	3. Unavoidable in current network environment, mostly for

establishing the network infrastructure

	4. Rarely need coordinations with others parameters, usually is
closed related to one specific autonomic function

	Some examples can be
	What kind of IGP the Autonomic network domain should use (such as

IS-IS, OSPF);

	What kind of transport layer network technology the Autonomic

network

domain should use (such as MPLS, IP)

	Should these ASA parameters be part of ANIMA's work items just as

intent?

Best Regards
Zongpeng Du

-----Original Message-----
From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
Sent: Thursday, March 17, 2016 2:41 AM
To: Duzongpeng; Michael Behringer (mbehring); anima@ietf.org
Cc: Laurent Ciavaglia; Jéferson Campos Nobre; Sheng Jiang
Subject: Re: [Anima] Does Autonomic network need some intervention
beyond intent

Yes, I think that whatever the "philosophy" of intent is, we should make

sure that the implementation is pragmatic and that we can adapt to
operator's requirements.

Regards
   Brian

On 17/03/2016 01:41, Duzongpeng wrote:

Hi, Michael,

         Thanks for the reply. I can understand that intent is not in the current

milestones, and you are working on several milestones drafts now.

         You are welcome to contribute to the intent draft at any time, and

thanks in advance.

         According to your statements, in my understanding, we can keep

regarding the "intent" example in "prefix-management ID" as an Autonomic
intent.

         It's semantics are just like "all nodes of type x do this; type y does

that", so it is a high level policy.

         It really helps a lot. Thanks.

         I generally agree with your statement, while some other personal

understandings exist (shown below). But perhaps they can be discussed
later.

Best Regards
Zongpeng Du

From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com]
Sent: Wednesday, March 16, 2016 4:17 PM
To: Duzongpeng; anima@ietf.org
Cc: Laurent Ciavaglia; Jéferson Campos Nobre; Sheng Jiang
Subject: RE: [Anima] Does Autonomic network need some intervention
beyond intent

Hi Zongpeng,

Yes, this is a recurring question and topic. Let me give my view on two

levels:

First, ANIMA isn't the only thing "influencing" or "steering" a network, at

least not initially. There is still traditional config, plus all the existing and
emerging SDN models. My view has always been that ANIMA sets a "default"
policy in a network, and all other means are more specific and can override
intent. This is described in RFC7575. Main point here: ANIMA Intent isn't the
only tool in a network.

[duzongpeng] I agree that intent can coexist with traditional config.
I generally agree with the policy priorities, but I have some other

concerns.

Some default policies/parameters set through "intent" perhaps are

fundamental for the uplayer services.

It should be prevented that human operator uses some CLIs to change

them by mistake, and the uplayer services are interrupted.

At least, some warning should be reported to the human operator.
[duzongpeng]

Intent could perfectly well cover a high level policy such as "all nodes of

type x do this; type y does that". (Which I think is what you mention below.)
When we started with Intent, the vision was to keep it very high level, more
along the line "do the right thing" :)  It has become clear that in today's
networking, people will want to push more "parameters" than "policy", and I
think we'll just have to acknowledge that.

Personally I think it'll pan out like this: Intent will have sections per

autonomic function. An Intent parser will validate overall correctness
(signature, etc), then segment the Intent into its bits: One part will be the
autonomic fabric itself, then there will be bits for each autonomic function.

So the Intent parser on a node will grab intent, see there is some Intent

for autonomic function A, will see whether this is running locally, and if so, it'll
just pass that piece of Intent to that function.

In this way, a function owner defines both the intent itself, and how it is

consumed by his/her function.

[duzongpeng]Agree.[duzongpeng]

The downside is that there is no "control" an//d people can do what they

want, in the worst case even push config, which is clearly NOT what Intent is
meant to do. I'd hope that people don't start writing Intent like "node 17:
here is your CLI; node 23: here is your CLI"; node 37: here is your CLI". But
what I describe above COULD be abused in this way. RFC7575 says this is NOT
Intent.

[duzongpeng]Agree. It is hard to give an exact boundary.
However, I think we have two choices here.
The first one is to define a policy first in every ASA, but with some open

parameters that can be configured. What we need to do is just to distribute
the network-level parameters for that function. This way is suitable to some
relatively static functions.

The second is more advanced, in the ideal case, the human operator can

communicate with network using natural language or something similar. This
way is more flexible, but harder to realize.

[duzongpeng]

In other words: Let's not be too prescriptive, and hope that folks will do

more or less reasonable things.

[duzongpeng]Agree.[duzongpeng]

I've been meaning to comment on the Intent drafts and possibly

contribute, I'm just struggling with priorities at the moment (as you can see
by my relative silence over the past weeks).

Michael



From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng
Sent: 16 March 2016 02:38
To: anima@ietf.org<mailto:anima@ietf.org>
Cc: Laurent Ciavaglia
<laurent.ciavaglia@nokia.com<mailto:laurent.ciavaglia@nokia.com>>;
Jéferson Campos Nobre
<jcnobre@inf.ufrgs.br<mailto:jcnobre@inf.ufrgs.br>>; Sheng Jiang
<jiangsheng@huawei.com<mailto:jiangsheng@huawei.com>>
Subject: [Anima] Does Autonomic network need some intervention

beyond

intent

Hello, everyone

         I understand that Autonomic network should be able to work with as

less configuration as possible. In the ideal case, an Autonomic network
should be able to work well while only intent is needed from human
operators.

         However, I wonder whether Autonomic network may need some

intervention from human operators through some low-level policies beyond
intents.

         For example, in the https://tools.ietf.org/html/draft-ietf-anima-

prefix-management-00, it is suggested that the prefix lengths for the CSG,
ASG, RSG (different roles in IP RAN) can be assigned as an "intent". These
configuration parameters need to be distributed in the autonomic domain to
influence the detail configurations on each autonomic node.

         As intent is described as an abstract, declarative, high-level policy

used to operate an autonomic domain, such as an enterprise network.
Perhaps we can rename the "intent" in the "prefix-management ID" to some
other words.

Is "ASA parameters" ok here?

I think some characteristics for this kind of low-level policies are

1.       Network level parameters, need to be distributed by GRASP and

ACP

2.       Relatively static compared to intent, perhaps only needed to be

configured once

3.       Unavoidable in current network environment, mostly for

establishing the network infrastructure

4.       Rarely need coordinations with others parameters, usually is closed

related to one specific autonomic function

Welcome for comment. Your suggestions will be appreciated.


Best Regards
Zongpeng Du



_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima



        

      

      
_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima


    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------090100060506080504000207--


From nobody Mon Mar 21 04:53:40 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EB96C12D747 for ; Mon, 21 Mar 2016 04:53:36 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.91
X-Spam-Level: 
X-Spam-Status: No, score=-6.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_KAM_HTML_FONT_INVALID=0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bBAjtai35i3M for ; Mon, 21 Mar 2016 04:53:28 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 077B912D5CD for ; Mon, 21 Mar 2016 04:53:28 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id 9B6DED6C93E92; Mon, 21 Mar 2016 11:53:23 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2LBrOQl031864 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 21 Mar 2016 11:53:26 GMT
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2LBrHEa003964 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 21 Mar 2016 12:53:23 +0100
Received: from [172.27.205.106] (135.239.27.38) by FR711WXCHHUB02.zeu.alcatel-lucent.com (135.239.2.112) with Microsoft SMTP Server (TLS) id 14.3.195.1; Mon, 21 Mar 2016 12:53:19 +0100
To: "EXT Michael Behringer (mbehring)" , Laurent Ciavaglia , "EXT Liubing (Leo)" , "anima@ietf.org" 
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56EFE0AE.5040007@alcatel-lucent.com>
Date: Mon, 21 Mar 2016 12:53:18 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com>
Content-Type: multipart/alternative; boundary="------------010608050202050503020003"
X-Originating-IP: [135.239.27.38]
Archived-At: 
Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 11:53:37 -0000

--------------010608050202050503020003
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 8bit

Michael, all,

Thanks for providing your example!

My first comment would be to consider "breaking" the model into parts.
Some clarifications:
     -I don't think the ANIMA intent should be one single (big) file. 
This will be cumbersome to manage/update, and contrary to the PBM 
approach principles.
     -Thus, an ANIMA network will have several, concurrent intents.

Therefore, I think we should define:
     -A format of the ANIMA intent that applies to the different target 
entities / roles.
     -e.g. the ANI, the ASAs.

Can/should the ANI intent and ASA intent use the same base model?
     -> I think this will depend if they share enough in common.

I will try to provide more details in the future (now busy with latest 
ID updates).

Thanks, best regards, Laurent.


On 21/03/2016 11:54, EXT Michael Behringer (mbehring) wrote:
>
> Agree; the encoding format question is secondary.
>
> My YANG example was just “playing around”, at this point I’m trying to 
> find the right structure. So don’t look at the details, just the 
> structure.
>
> Michael
>
> *From:*Laurent Ciavaglia [mailto:laurent.ciavaglia@nokia.com]
> *Sent:* 21 March 2016 11:45
> *To:* EXT Liubing (Leo) ; Michael Behringer 
> (mbehring) ; anima@ietf.org
> *Subject:* Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some 
> thoughts
>
> Hello,
>
> I think that before considering encodings, we should first agree on 
> the definition(s) of what we will manipulate.
> Let's start with a (formal) definiton of the "ANIMA" intent policy, 
> and the hierarchy envisaged and coarse "format".
> If we can work out an example with we think capture well enough and 
> large enough how and what the "ANIMA" is used for, then I think we 
> will already have made great progress.
>
> Best regards, Laurent.
>
> On 21/03/2016 11:02, EXT Liubing (Leo) wrote:
>
>     Hi Michael and all,
>
>     One quick comment aside from the Intent content discussion: since
>     GRASP uses CBOR, isn't it good for Intent also encoded by CBOR?
>
>     YANG model can surely describe an Intent, but it includes lots of
>     features for data manipulating (for Netconf), which might not
>     necessary for Intent describing?.
>
>     Best regards,
>
>     Bing
>
>         -----Original Message-----
>
>         From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael
>
>         Behringer (mbehring)
>
>         Sent: Friday, March 18, 2016 5:52 PM
>
>         To: anima@ietf.org 
>
>         Subject: [Anima] Intent yang model - some thoughts
>
>         WRT the current discussion on what should be in Intent and
>         what shouldn't,
>
>         as mentioned before, I think we should leave this to the owner
>         of an
>
>         autonomic function to decide. This risks "abuse" in a way that
>         wasn't
>
>         foreseen, but it gives maximum agility, which is what we need
>         to get ANIMA
>
>         off the ground, IMO. If we are in the nice position to find
>         that it gets
>
>         adopted a lot, and if there are things that need fixing then,
>         we can work on
>
>         version 2.
>
>         Several months ago I started with a yang model for Intent, but
>         never finished.
>
>         Maybe this helps with this discussion. Note I'm not at all a
>         yang expert, so
>
>         please look more at the general structure for now. Not sure if
>         it helps the
>
>         discussion...
>
>         +--rw intent
>
>            +--rw meta-data
>
>            |  +--rw intent-version  uint32   // the version of the intent
>
>         specification
>
>            |  +--rw timestamp            uint32   // latest time stamp
>         wins
>
>            |  +--rw administrator        string   // who created that
>         intent
>
>            |  +--rw device?              string   // on which node was
>         it created
>
>            |  +--rw execution-time       uint32   // plus other
>         parameters on when to
>
>         execute
>
>            |  +--rw expiration           uint32   // when does this
>         intent cease to be
>
>         valid
>
>            |  +--rw feedback                 // how to let the
>         operator know
>
>         about success/failure
>
>            |  +--rw signature            ??       // intent must be signed
>
>            |
>
>            +--rw scope
>
>               +--rw domains
>
>                  +--rw domain* [name]
>
>                     +--rw name           string    // is there a fqdn
>         type defined?
>
>                     +--rw meta-data           // general about the domain
>
>                     |  +--rw include-subdomains  boolean  // should
>
>         sub-domains follow this policy?
>
>                     |  +--rw fail-open  boolean  // fail-open or
>         fail-closed
>
>         policy
>
>                     |  +--rw maintenance-window  // when can critical
>
>         operations take place
>
>                     |  +--rw ...                 // other general
>         policies on
>
>         domain level
>
>         |        ...
>
>                     |
>
>                     +--rw ani-intent
>
>                     +-- (contains all policies which the ANI should
>         follow)
>
>                     |  ex: ACP policies, coordination between AN
>         functions,
>
>         security policies, etc
>
>                     |  for ani-intent, scope is always "all" (ie all
>         nodes must
>
>         process)
>
>                     |  ...
>
>         ...
>
>         All of the above MUST be standardised, because it refers to the AN
>
>         infrastructure itself.
>
>         Of the below we need to standardise the structure, but not
>         content for each
>
>         function Conceptually, we define Intent for a domain.
>
>         ...
>
>                     |
>
>                     +--rw intent-for-autonomic-functions
>
>                        +--rw function_1 [name]
>
>                        |  +--rw scope                    // which devices
>
>         in the domain need to listen.
>
>                        |  |  +--rw all    boolean        // all devices
>
>                        |  |  +--rw role   string         // as defined in
>
>         certificate
>
>                        |  |  +--rw dynamic_role  string  // as determined
>
>         dynamically
>
>                        |  |  +--rw tag    string         // can be
>         used for
>
>         location
>
>                        |  +-- function_intent            // to be
>         defined by
>
>         function owner
>
>                        |
>
>                        +--rw function_2 [name]
>
>                        |  +--rw scope                    // which devices
>
>         in the domain need to listen.
>
>                        |  |  +--rw all    boolean        // all devices
>
>                        |  |  +--rw role   string         // as defined in
>
>         certificate
>
>                        |  |  +--rw dynamic_role  string  // as determined
>
>         dynamically
>
>                        |  |  +--rw tag    string         // can be
>         used for
>
>         location
>
>                        |  +-- function_intent            // to be
>         defined by
>
>         function owner
>
>                        |
>
>         ...
>
>         If a node receives Intent, it checks:
>
>         1) meta data: Mostly for forwarding and processing of Intent
>
>         2) domain info: domain specific policy
>
>         3) ANI Intent: Intent for the ANI itself; every node must be
>         able to process
>
>         that
>
>         4) For each of the "function" blocks:
>
>            - should I run this function?
>
>            - do I run this function?
>
>            - if so, forward the rest of that block to the
>         corresponding handler.
>
>         I think we want a clean architectural separation between the
>         ANI and the
>
>         functions. This way every function can do what it wants.
>
>         Thoughts?
>
>         Michael
>
>         _______________________________________________
>
>         Anima mailing list
>
>         Anima@ietf.org 
>
>         https://www.ietf.org/mailman/listinfo/anima
>
>     _______________________________________________
>
>     Anima mailing list
>
>     Anima@ietf.org 
>
>     https://www.ietf.org/mailman/listinfo/anima
>
> -- 
>
> Laurent Ciavaglia
>
> Senior Research Manager
>
> Bell Labs, Nokia
>
> +33 160 402 636
>
> Route de Villejust | 91620 Nozay | France
>
> LinkedIn: laurentciavaglia 
>

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------010608050202050503020003
Content-Type: text/html; charset="windows-1252"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Michael, all,

      

      Thanks for providing your example!

      

      My first comment would be to consider "breaking" the model into
      parts.

      Some clarifications:

          -I don't think the ANIMA intent should be one single (big)
      file. This will be cumbersome to manage/update, and contrary to
      the PBM approach principles.

          -Thus, an ANIMA network will have several, concurrent intents.

      

      Therefore, I think we should define:

          -A format of the ANIMA intent that applies to the different
      target entities / roles.

          -e.g. the ANI, the ASAs.

      

      Can/should the ANI intent and ASA intent use the same base model?
      

          -> I think this will depend if they share enough in common.

      

      I will try to provide more details in the future (now busy with
      latest ID updates).

      

      Thanks, best regards, Laurent.

      

    

    
On 21/03/2016 11:54, EXT Michael
      Behringer (mbehring) wrote:

    

    

      
      
      
      

        
Agree;
              the encoding format question is secondary.
              

        
My
              YANG example was just “playing around”, at this point I’m
              trying to find the right structure. So don’t look at the
              details, just the structure. 

        
Michael

        
 

        

          

            

              

                From: Laurent Ciavaglia
                    [mailto:laurent.ciavaglia@nokia.com] 

                    Sent:
                    21 March 2016 11:45

                    To: EXT
                    Liubing (Leo) <leo.liubing@huawei.com>;
                    Michael Behringer (mbehring)
                    <mbehring@cisco.com>; anima@ietf.org

                    Subject:
                    Re: [Anima] CBOR for Intent?-//RE: Intent yang model
                    - some thoughts

            

          

          
 

          
Hello,

                

                I think that before considering encodings, we should
                first agree on the definition(s) of what we will
                manipulate.

                Let's start with a (formal) definiton of the "ANIMA"
                intent policy, and the hierarchy envisaged and coarse
                "format".

                If we can work out an example with we think capture well
                enough and large enough how and what the "ANIMA" is used
                for, then I think we will already have made great
                progress.
                

                

                Best regards, Laurent.

          

            
On
                  21/03/2016 11:02, EXT Liubing (Leo) wrote:

          

          

            
Hi Michael and all,

            
 

            
One quick comment aside from the Intent content discussion: since GRASP uses CBOR, isn't it good for Intent also encoded by CBOR?

            
 

            
YANG model can surely describe an Intent, but it includes lots of features for data manipulating (for Netconf), which might not necessary for Intent describing?. 

            
 

            
Best regards,

            
Bing

            
 

            

              
-----Original Message-----

              
From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael

              
Behringer (mbehring)

              
Sent: Friday, March 18, 2016 5:52 PM

              
To: anima@ietf.org

              
Subject: [Anima] Intent yang model - some thoughts

              
 

              
WRT the current discussion on what should be in Intent and what shouldn't,

              
as mentioned before, I think we should leave this to the owner of an

              
autonomic function to decide. This risks "abuse" in a way that wasn't

              
foreseen, but it gives maximum agility, which is what we need to get ANIMA

              
off the ground, IMO. If we are in the nice position to find that it gets

              
adopted a lot, and if there are things that need fixing then, we can work on

              
version 2.

              
 

              
Several months ago I started with a yang model for Intent, but never finished.

              
Maybe this helps with this discussion. Note I'm not at all a yang expert, so

              
please look more at the general structure for now. Not sure if it helps the

              
discussion...

              
 

              
+--rw intent

              
   +--rw meta-data

              
   |  +--rw intent-version  uint32   // the version of the intent

              
specification

              
   |  +--rw timestamp            uint32   // latest time stamp wins

              
   |  +--rw administrator        string   // who created that intent

              
   |  +--rw device?              string   // on which node was it created

              
   |  +--rw execution-time       uint32   // plus other parameters on when to

              
execute

              
   |  +--rw expiration           uint32   // when does this intent cease to be

              
valid

              
   |  +--rw feedback                 // how to let the operator know

              
about success/failure

              
   |  +--rw signature            ??       // intent must be signed

              
   |

              
   +--rw scope

              
      +--rw domains

              
         +--rw domain* [name]

              
            +--rw name           string    // is there a fqdn type defined?

              
            +--rw meta-data           // general about the domain

              
            |  +--rw include-subdomains  boolean  // should

              
sub-domains follow this policy?

              
            |  +--rw fail-open  boolean  // fail-open or fail-closed

              
policy

              
            |  +--rw maintenance-window  // when can critical

              
operations take place

              
            |  +--rw ...                 // other general policies on

              
domain level

              
|        ...

              
            |

              
            +--rw ani-intent

              
            +-- (contains all policies which the ANI should follow)

              
            |  ex: ACP policies, coordination between AN functions,

              
security policies, etc

              
            |  for ani-intent, scope is always "all" (ie all nodes must

              
process)

              
            |  ...

              
...

              
All of the above MUST be standardised, because it refers to the AN

              
infrastructure itself.

              
Of the below we need to standardise the structure, but not content for each

              
function Conceptually, we define Intent for a domain.

              
...

              
            |

              
            +--rw intent-for-autonomic-functions

              
               +--rw function_1 [name]

              
               |  +--rw scope                    // which devices

              
in the domain need to listen.

              
               |  |  +--rw all    boolean        // all devices

              
               |  |  +--rw role   string         // as defined in

              
certificate

              
               |  |  +--rw dynamic_role  string  // as determined

              
dynamically

              
               |  |  +--rw tag    string         // can be used for

              
location

              
               |  +-- function_intent            // to be defined by

              
function owner

              
               |

              
               +--rw function_2 [name]

              
               |  +--rw scope                    // which devices

              
in the domain need to listen.

              
               |  |  +--rw all    boolean        // all devices

              
               |  |  +--rw role   string         // as defined in

              
certificate

              
               |  |  +--rw dynamic_role  string  // as determined

              
dynamically

              
               |  |  +--rw tag    string         // can be used for

              
location

              
               |  +-- function_intent            // to be defined by

              
function owner

              
               |

              
...

              
 

              
If a node receives Intent, it checks:

              
1) meta data: Mostly for forwarding and processing of Intent

              
2) domain info: domain specific policy

              
3) ANI Intent: Intent for the ANI itself; every node must be able to process

              
that

              
4) For each of the "function" blocks:

              
   - should I run this function?

              
   - do I run this function?

              
   - if so, forward the rest of that block to the corresponding handler.

              
 

              
I think we want a clean architectural separation between the ANI and the

              
functions. This way every function can do what it wants.

              
 

              
Thoughts?

              
Michael

              
 

              
_______________________________________________

              
Anima mailing list

              
Anima@ietf.org

              
https://www.ietf.org/mailman/listinfo/anima

            

            
 

            
_______________________________________________

            
Anima mailing list

            
Anima@ietf.org

            
https://www.ietf.org/mailman/listinfo/anima

            
 

          

          
 

          

            

              -- 

                  

                

            

              Laurent Ciavaglia

            

              Senior Research Manager

            

              Bell Labs, Nokia

            

               

            

              +33 160 402 636
                

            

              Route de Villejust | 91620
                  Nozay | France

            

              LinkedIn:
                laurentciavaglia

            
 

          

        

      

    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------010608050202050503020003--


From nobody Mon Mar 21 04:53:56 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CF4FD12D5CD for ; Mon, 21 Mar 2016 04:53:47 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.201
X-Spam-Level: 
X-Spam-Status: No, score=-4.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, T_FILL_THIS_FORM_SHORT=0.01, T_KAM_HTML_FONT_INVALID=0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id h0xVokqMU_6r for ; Mon, 21 Mar 2016 04:53:39 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BE99512D745 for ; Mon, 21 Mar 2016 04:53:32 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml708-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGK04863; Mon, 21 Mar 2016 11:53:29 +0000 (GMT)
Received: from NKGEML412-HUB.china.huawei.com (10.98.56.73) by lhreml708-cah.china.huawei.com (10.201.5.202) with Microsoft SMTP Server (TLS) id 14.3.235.1; Mon, 21 Mar 2016 11:53:28 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml412-hub.china.huawei.com ([10.98.56.73]) with mapi id 14.03.0235.001; Mon, 21 Mar 2016 19:53:24 +0800
From: "Liubing (Leo)" 
To: "Michael Behringer (mbehring)" , Laurent Ciavaglia , "anima@ietf.org" 
Thread-Topic: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts
Thread-Index: AdGA84Aiy7EwHtVtTUWSCghLdxIQCwCYxtng//+J+gCAAALFgP//ao4g
Date: Mon, 21 Mar 2016 11:53:24 +0000
Message-ID: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5CB78@nkgeml514-mbx.china.huawei.com>
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com>
In-Reply-To: <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com>
Accept-Language: en-US, zh-CN
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.98.117]
Content-Type: multipart/alternative; boundary="_000_8AE0F17B87264D4CAC7DE0AA6C406F45C2D5CB78nkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020202.56EFE0BA.0199, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 7e907c2345d3a058fd35fabae67a2320
Archived-At: 
Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 11:53:48 -0000

--_000_8AE0F17B87264D4CAC7DE0AA6C406F45C2D5CB78nkgeml514mbxchi_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Hi Michael, Laurent, and all
Yes, encoding is a secondary issue. Just had a random comment:)

Best regards,
Bing

From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com]
Sent: Monday, March 21, 2016 6:54 PM
To: Laurent Ciavaglia; Liubing (Leo); anima@ietf.org
Subject: RE: [Anima] CBOR for Intent?-//RE: Intent yang model - some though=
ts

Agree; the encoding format question is secondary.
My YANG example was just "playing around", at this point I'm trying to find=
 the right structure. So don't look at the details, just the structure.
Michael

From: Laurent Ciavaglia [mailto:laurent.ciavaglia@nokia.com]
Sent: 21 March 2016 11:45
To: EXT Liubing (Leo) >; Michael Behringer (mbehring) >; anima@ietf.org
Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some though=
ts

Hello,

I think that before considering encodings, we should first agree on the def=
inition(s) of what we will manipulate.
Let's start with a (formal) definiton of the "ANIMA" intent policy, and the=
 hierarchy envisaged and coarse "format".
If we can work out an example with we think capture well enough and large e=
nough how and what the "ANIMA" is used for, then I think we will already ha=
ve made great progress.

Best regards, Laurent.
On 21/03/2016 11:02, EXT Liubing (Leo) wrote:

Hi Michael and all,



One quick comment aside from the Intent content discussion: since GRASP use=
s CBOR, isn't it good for Intent also encoded by CBOR?



YANG model can surely describe an Intent, but it includes lots of features =
for data manipulating (for Netconf), which might not necessary for Intent d=
escribing?.



Best regards,

Bing



-----Original Message-----

From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael

Behringer (mbehring)

Sent: Friday, March 18, 2016 5:52 PM

To: anima@ietf.org

Subject: [Anima] Intent yang model - some thoughts



WRT the current discussion on what should be in Intent and what shouldn't,

as mentioned before, I think we should leave this to the owner of an

autonomic function to decide. This risks "abuse" in a way that wasn't

foreseen, but it gives maximum agility, which is what we need to get ANIMA

off the ground, IMO. If we are in the nice position to find that it gets

adopted a lot, and if there are things that need fixing then, we can work o=
n

version 2.



Several months ago I started with a yang model for Intent, but never finish=
ed.

Maybe this helps with this discussion. Note I'm not at all a yang expert, s=
o

please look more at the general structure for now. Not sure if it helps the

discussion...



+--rw intent

   +--rw meta-data

   |  +--rw intent-version  uint32   // the version of the intent

specification

   |  +--rw timestamp            uint32   // latest time stamp wins

   |  +--rw administrator        string   // who created that intent

   |  +--rw device?              string   // on which node was it created

   |  +--rw execution-time       uint32   // plus other parameters on when =
to

execute

   |  +--rw expiration           uint32   // when does this intent cease to=
 be

valid

   |  +--rw feedback                 // how to let the operator know

about success/failure

   |  +--rw signature            ??       // intent must be signed

   |

   +--rw scope

      +--rw domains

         +--rw domain* [name]

            +--rw name           string    // is there a fqdn type defined?

            +--rw meta-data           // general about the domain

            |  +--rw include-subdomains  boolean  // should

sub-domains follow this policy?

            |  +--rw fail-open  boolean  // fail-open or fail-closed

policy

            |  +--rw maintenance-window  // when can critical

operations take place

            |  +--rw ...                 // other general policies on

domain level

|        ...

            |

            +--rw ani-intent

            +-- (contains all policies which the ANI should follow)

            |  ex: ACP policies, coordination between AN functions,

security policies, etc

            |  for ani-intent, scope is always "all" (ie all nodes must

process)

            |  ...

...

All of the above MUST be standardised, because it refers to the AN

infrastructure itself.

Of the below we need to standardise the structure, but not content for each

function Conceptually, we define Intent for a domain.

...

            |

            +--rw intent-for-autonomic-functions

               +--rw function_1 [name]

               |  +--rw scope                    // which devices

in the domain need to listen.

               |  |  +--rw all    boolean        // all devices

               |  |  +--rw role   string         // as defined in

certificate

               |  |  +--rw dynamic_role  string  // as determined

dynamically

               |  |  +--rw tag    string         // can be used for

location

               |  +-- function_intent            // to be defined by

function owner

               |

               +--rw function_2 [name]

               |  +--rw scope                    // which devices

in the domain need to listen.

               |  |  +--rw all    boolean        // all devices

               |  |  +--rw role   string         // as defined in

certificate

               |  |  +--rw dynamic_role  string  // as determined

dynamically

               |  |  +--rw tag    string         // can be used for

location

               |  +-- function_intent            // to be defined by

function owner

               |

...



If a node receives Intent, it checks:

1) meta data: Mostly for forwarding and processing of Intent

2) domain info: domain specific policy

3) ANI Intent: Intent for the ANI itself; every node must be able to proces=
s

that

4) For each of the "function" blocks:

   - should I run this function?

   - do I run this function?

   - if so, forward the rest of that block to the corresponding handler.



I think we want a clean architectural separation between the ANI and the

functions. This way every function can do what it wants.



Thoughts?

Michael



_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima



_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima



--
Laurent Ciavaglia
Senior Research Manager
Bell Labs, Nokia

+33 160 402 636
Route de Villejust | 91620 Nozay | France
LinkedIn: laurentciavaglia



--_000_8AE0F17B87264D4CAC7DE0AA6C406F45C2D5CB78nkgeml514mbxchi_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable











Hi Michael, Laurent, and all

Yes, encoding is a secondary issue. Just had a r=
andom commentJ


 


Best regards,


Bing


 









From: Michael Behringer (mbehring) [mailto:mbe=
hring@cisco.com]


Sent: Monday, March 21, 2016 6:54 PM

To: Laurent Ciavaglia; Liubing (Leo); anima@ietf.org

Subject: RE: [Anima] CBOR for Intent?-//RE: Intent yang model - some=
 thoughts






 


Agree; =
the encoding format question is secondary.



My YANG=
 example was just “playing around”, at this point I’m try=
ing to find the right structure. So don’t look at the details, just t=
he structure.



Michael=



&n=
bsp;









From: Laurent Ciavaglia [mailto:laurent.ciavaglia@nokia.com]


Sent: 21 March 2016 11:45

To: EXT Liubing (Leo) <=
leo.liubing@huawei.com>; Michael Behringer (mbehring) <mbehring@cisco.com>;
anima@ietf.org

Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some=
 thoughts






 


Hello,



I think that before considering encodings, we should first agree on the def=
inition(s) of what we will manipulate.

Let's start with a (formal) definiton of the "ANIMA" intent polic=
y, and the hierarchy envisaged and coarse "format".

If we can work out an example with we think capture well enough and large e=
nough how and what the "ANIMA" is used for, then I think we will =
already have made great progress.




Best regards, Laurent.




On 21/03/2016 11:02, EXT Liubing (Leo) wrote:=







Hi Michael and all,


 


One quick comment aside from the Intent content d=
iscussion: since GRASP uses CBOR, isn't it good for Intent also encoded by =
CBOR?


 


YANG model can surely describe an Intent, but it =
includes lots of features for data manipulating (for Netconf), which might =
not necessary for Intent describing?. 


 


Best regards,


Bing


 




-----Original Message-----

From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael


Behringer (mbehring)


Sent: Friday, March 18, 2016 5:52 PM


To: anima@ietf.=
org


Subject: [Anima] Intent yang model - some thought=
s


 


WRT the current discussion on what should be in I=
ntent and what shouldn't,


as mentioned before, I think we should leave this=
 to the owner of an


autonomic function to decide. This risks "ab=
use" in a way that wasn't


foreseen, but it gives maximum agility, which is =
what we need to get ANIMA


off the ground, IMO. If we are in the nice positi=
on to find that it gets


adopted a lot, and if there are things that need =
fixing then, we can work on


version 2.


 


Several months ago I started with a yang model fo=
r Intent, but never finished.


Maybe this helps with this discussion. Note I'm n=
ot at all a yang expert, so


please look more at the general structure for now=
. Not sure if it helps the


discussion...


 


+--rw intent


   +--rw meta-data


   |  +--rw intent-version&nbs=
p; uint32   // the version of the intent


specification


   |  +--rw timestamp  &n=
bsp;         uint32   // =
latest time stamp wins


   |  +--rw administrator &nbs=
p;      string   // who created that int=
ent


   |  +--rw device?  =
;            string =
  // on which node was it created


   |  +--rw execution-time&nbs=
p;      uint32   // plus other parameter=
s on when to


execute


   |  +--rw expiration &n=
bsp;         uint32   // when =
does this intent cease to be


valid


   |  +--rw feedback &nbs=
p;            &=
nbsp;  // how to let the operator know


about success/failure


   |  +--rw signature &nb=
sp;          ??  &nb=
sp;    // intent must be signed


   |


   +--rw scope

      +--rw domains<=
o:p>


         =
+--rw domain* [name]


        &=
nbsp;   +--rw name       &=
nbsp;   string    // is there a fqdn type defined?=



        &=
nbsp;   +--rw meta-data      &n=
bsp;    // general about the domain


        &=
nbsp;   |  +--rw include-subdomains  boolean  =
// should


sub-domains follow this policy?=



        &=
nbsp;   |  +--rw fail-open  boolean  // fail-o=
pen or fail-closed


policy


        &=
nbsp;   |  +--rw maintenance-window  // when can cr=
itical


operations take place


        &=
nbsp;   |  +--rw ...      =
           // other gener=
al policies on


domain level


|        ...


        &=
nbsp;   |


        &=
nbsp;   +--rw ani-intent


        &=
nbsp;   +-- (contains all policies which the ANI should follo=
w)


        &=
nbsp;   |  ex: ACP policies, coordination between AN functio=
ns,


security policies, etc


        &=
nbsp;   |  for ani-intent, scope is always "all" (=
ie all nodes must


process)


        &=
nbsp;   |  ...


...


All of the above MUST be standardised, because it=
 refers to the AN


infrastructure itself.


Of the below we need to standardise the structure=
, but not content for each


function Conceptually, we define Intent for a dom=
ain.


...


        &=
nbsp;   |


        &=
nbsp;   +--rw intent-for-autonomic-functions


        &=
nbsp;      +--rw function_1 [name]<=
/span>


        &=
nbsp;      |  +--rw scope  &nbs=
p;            &=
nbsp;    // which devices


in the domain need to listen.

        &=
nbsp;      |  |  +--rw all &nbs=
p;  boolean        // all devices


        &=
nbsp;      |  |  +--rw role &nb=
sp; string         // as defined in=



certificate


        &=
nbsp;      |  |  +--rw dynamic_role&=
nbsp; string  // as determined


dynamically


        &=
nbsp;      |  |  +--rw tag &nbs=
p;  string         // can be u=
sed for


location


        &=
nbsp;      |  +-- function_intent &n=
bsp;          // to be defined=
 by


function owner


        &=
nbsp;      |


        &=
nbsp;      +--rw function_2 [name]<=
/span>


        &=
nbsp;      |  +--rw scope  &nbs=
p;             =
    // which devices


in the domain need to listen.

        &=
nbsp;      |  |  +--rw all &nbs=
p;  boolean        // all devices


        &=
nbsp;      |  |  +--rw role &nb=
sp; string         // as defined in=



certificate


        &=
nbsp;      |  |  +--rw dynamic_role&=
nbsp; string  // as determined


dynamically


        &=
nbsp;      |  |  +--rw tag &nbs=
p;  string         // can be u=
sed for


location


        &=
nbsp;      |  +-- function_intent &n=
bsp;          // to be defined=
 by


function owner


        &=
nbsp;      |


...


 


If a node receives Intent, it checks:<=
/span>


1) meta data: Mostly for forwarding and processin=
g of Intent


2) domain info: domain specific policy=



3) ANI Intent: Intent for the ANI itself; every n=
ode must be able to process


that


4) For each of the "function" blocks:


   - should I run this function?


   - do I run this function?=



   - if so, forward the rest of that bl=
ock to the corresponding handler.


 


I think we want a clean architectural separation =
between the ANI and the


functions. This way every function can do what it=
 wants.


 


Thoughts?


Michael


 


_______________________________________________


Anima mailing list


Anima@ietf.org<=
/a>


https://www.ietf.org/mailman/listinfo/anima



 


_______________________________________________


Anima mailing list


Anima@ietf.org<=
/a>


https://www.ietf.org/mailman/listinfo/anima

 




 




--




Laurent Ciavaglia



Senior Research Manager



Bell Labs, Nokia



 



+33 160 402 63=
6 



Route de Villejust | 91620 N=
ozay | France



LinkedIn: lauren=
tciavaglia



 












--_000_8AE0F17B87264D4CAC7DE0AA6C406F45C2D5CB78nkgeml514mbxchi_--


From nobody Mon Mar 21 05:53:01 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AC5DB12D772 for ; Mon, 21 Mar 2016 05:53:00 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.902
X-Spam-Level: 
X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id J2QPdiv-_jpG for ; Mon, 21 Mar 2016 05:52:58 -0700 (PDT)
Received: from tuna.sandelman.ca (tuna.sandelman.ca [IPv6:2607:f0b0:f:3:216:3eff:fe7c:d1f3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9FC1712D748 for ; Mon, 21 Mar 2016 05:52:55 -0700 (PDT)
Received: from sandelman.ca (obiwan.sandelman.ca [209.87.249.21]) by tuna.sandelman.ca (Postfix) with ESMTP id 68F45200A5 for ; Mon, 21 Mar 2016 08:55:30 -0400 (EDT)
Received: from obiwan.sandelman.ca (localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id 14B306375A for ; Mon, 21 Mar 2016 08:52:54 -0400 (EDT)
From: Michael Richardson 
To: "anima\@ietf.org" 
In-Reply-To: <56EEF72A.5050309@gmail.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com>
X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2
X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m
Archived-At: 
Subject: [Anima] need for some ASA definitions definitions to help bootstrap
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 12:53:00 -0000

--=-=-=
Content-Type: text/plain


The ANIMA bootstrap work is coming along.  The question of how, prior to
secure bootstrap, the New Node (the pledge) finds the proxy has been resolved
with the use of dnssd/mDNS-service-discovery on the link local addresses.
This has been selected on the basis of code reuse for devices which need/want
to bootstrap, but which might never participate in the ACP itself.  Please
see our draft.

There remains two other things that need to discover which we do not cover,
which are more ANIMA general problems.  These are:
  1) the "discovery" ("revelation"?) of a node that it needs to become
     a proxy.

  2) the discovery of the proxy node of the registrar.

These are things that *could* be determined by "power-on-above", i.e. a
controller in the NOC could push (netconf-style or otherweise) configuration
into a node to tell it, "You will be a bootstrap proxy on interfaces A,C and
Q, and the registrar is at address BAR"

Or these things could be discovered via ASA.

number (2) -- discoverying the registrar by the proxy node should be a
       relatively straight forward application of GRASP over the secured
       ACP to discover an ASA.

number (1) -- whether to become a proxy on a particular interface ought also
       to be negotiated via the secured ACP, even though the proxy could
       observe the mDNS-service-discovery process, the node which replies
       could in fact be malicious.
       So this is a less straightforward application of GRASP, it needs to
       driven by an intent that says something like: "new nodes may join
       network BAR on interface FOO"  for suitable abstractions of "FOO"
       being device independant.

There is also the ASA which would cause a node to turn itself into a
registrar (or rather, enable a registrar container, process or VM) when it
realizes that there is no other registrar, but I think that this is a higher
level of independance of Autonomic than we are ready for yet.

Aside from socializing these requirements into the larger group, I think that
we also looking for someone who might step forward to take the lead on these
two documents.  Maybe someone new?

--
Michael Richardson , Sandelman Software Works
 -= IPv6 IoT consulting =-




--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBVu/uo4CLcPvd0N1lAQJthQgAtfr8DCL8iPbyo1xAVGCa+nIY3vSs6Hhv
fIIV6Dmib75xnbnXnB2qf3fL5sPJuFfnqZH8UwV1M63WMLpWGvk6tij1/QtPfAH1
VvLdjJNCeyr5brkMARQxr3VX1sZstUmFnTnL8bWcY1Hj0jsTnQgUaNpO0Sz/Mb+L
reOZzWr1hG5grIVatxTVt7TZnn3QggzuPYOks700ulpjbkqB1WVDAnUIg2MGxby2
vm9yTs6XljBvtPZxFQmHKgY4WUHe4GcnrbTvYspsoBG61b31MlDH8hRMO9UuG+QB
074PhDRSaT35SSeYsD1ayYtgtKtW5CGbmk3iZJn2ovCD15+qwWn8fw==
=jvHu
-----END PGP SIGNATURE-----
--=-=-=--


From nobody Mon Mar 21 10:08:02 2016
Return-Path: 
X-Original-To: anima@ietf.org
Delivered-To: anima@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 59E3012D91A; Mon, 21 Mar 2016 10:08:01 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: 
X-Test-IDTracker: no
X-IETF-IDTracker: 6.17.0
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20160321170801.31929.90988.idtracker@ietfa.amsl.com>
Date: Mon, 21 Mar 2016 10:08:01 -0700
Archived-At: 
Cc: anima@ietf.org
Subject: [Anima] I-D Action: draft-ietf-anima-reference-model-01.txt
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 17:08:01 -0000

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Autonomic Networking Integrated Model and Approach of the IETF.

        Title           : A Reference Model for Autonomic Networking
        Authors         : Michael H. Behringer
                          Brian Carpenter
                          Toerless Eckert
                          Laurent Ciavaglia
                          Bing Liu
                          Jeferson Campos Nobre
                          John Strassner
	Filename        : draft-ietf-anima-reference-model-01.txt
	Pages           : 23
	Date            : 2016-03-21

Abstract:
   This document describes a reference model for Autonomic Networking.
   The goal is to define how the various elements in an autonomic
   context work together, to describe their interfaces and relations.
   While the document is written as generally as possible, the initial
   solutions are limited to the chartered scope of the WG.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-anima-reference-model/

There's also a htmlized version available at:
https://tools.ietf.org/html/draft-ietf-anima-reference-model-01

A diff from the previous version is available at:
https://www.ietf.org/rfcdiff?url2=draft-ietf-anima-reference-model-01


Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

Internet-Drafts are also available by anonymous FTP at:
ftp://ftp.ietf.org/internet-drafts/


From nobody Mon Mar 21 11:14:31 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9701912D540 for ; Mon, 21 Mar 2016 11:14:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.522
X-Spam-Level: 
X-Spam-Status: No, score=-14.522 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hOakHf4Xmr29 for ; Mon, 21 Mar 2016 11:14:28 -0700 (PDT)
Received: from rcdn-iport-9.cisco.com (rcdn-iport-9.cisco.com [173.37.86.80]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C7EC512D84F for ; Mon, 21 Mar 2016 11:14:28 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=3498; q=dns/txt; s=iport; t=1458584068; x=1459793668; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=r/2F3WFDOW6XuP3JOOBNDp46B77ui/OcwtTWg4qOUkE=; b=VWnudZ9YJy4IoP0ODgYkeQjWoMKN1NC2USvc7sUBEliy3RfDx2UBbv3J +g+Lp8ULEAuf9TBkItionSfepIIgB2GFR2RGdMUCyu+Otx6vtZniLSaeW ZCbrWXcRibAIZrXG25/evRL1cFMS6ZuLMHWNVQZkllXlB9aUKeBbq7G7M 4=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AjAgALOfBW/4ENJK1egzRTcgauUotNA?= =?us-ascii?q?Q2BcCGFbAIcgRE4FAEBAQEBAQFkHAuEQQEBAQQjET4FDgQCAQgRBAEBAwIjAwI?= =?us-ascii?q?CAh8RFAEGAQEFAwIEEwgTh3cDEg6vX4pODYRUAQEBAQEBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEBFXyFIoREgj6BWCw0gkaCVgWNO4lrMQGFcIYegW6BbEuDf4hYhzGHVAEeAQF?= =?us-ascii?q?Cg2VqAYhZPX4BAQE?=
X-IronPort-AV: E=Sophos;i="5.24,372,1454976000"; d="scan'208";a="83005903"
Received: from alln-core-9.cisco.com ([173.36.13.129]) by rcdn-iport-9.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 21 Mar 2016 18:14:28 +0000
Received: from XCH-ALN-009.cisco.com (xch-aln-009.cisco.com [173.36.7.19]) by alln-core-9.cisco.com (8.14.5/8.14.5) with ESMTP id u2LIERVt019889 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL) for ; Mon, 21 Mar 2016 18:14:27 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-009.cisco.com (173.36.7.19) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Mon, 21 Mar 2016 13:14:27 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Mon, 21 Mar 2016 13:14:26 -0500
From: "Michael Behringer (mbehring)" 
To: "anima@ietf.org" 
Thread-Topic: New Version Notification for draft-ietf-anima-reference-model-01.txt
Thread-Index: AQHRg5Q8d2gB+tmCHk2o+SY3V9MVpp9kKPiA
Date: Mon, 21 Mar 2016 18:14:26 +0000
Message-ID: 
References: <20160321170801.31929.20778.idtracker@ietfa.amsl.com>
In-Reply-To: <20160321170801.31929.20778.idtracker@ietfa.amsl.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Subject: [Anima] FW: New Version Notification for draft-ietf-anima-reference-model-01.txt
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 18:14:30 -0000
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From nobody Mon Mar 21 12:13:34 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D302312D8A1 for ; Mon, 21 Mar 2016 12:13:33 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Mz_xrthDAWxt for ; Mon, 21 Mar 2016 12:13:32 -0700 (PDT)
Received: from mail-pf0-x231.google.com (mail-pf0-x231.google.com [IPv6:2607:f8b0:400e:c00::231]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E6E0A12D8AE for ; Mon, 21 Mar 2016 12:13:31 -0700 (PDT)
Received: by mail-pf0-x231.google.com with SMTP id u190so276730100pfb.3 for ; Mon, 21 Mar 2016 12:13:31 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=/ApQFqmcouCl5STIusqfuTJXZzGokXhI+Lz0vLMrzp0=; b=M8/P5xX/RyRrlk7KBQcXyq82++kqjaF0SjecRjsJEKw6ZsezrCEtr1O4iiwLSW7x50 BIsDCOwHeZCD4g9s/3B9kwF+NdZmA/dvqJxgRggC9J36VlxTLpNKD6hShABeEXgzFaQP Q1GzjIpoZ11v74i67GulegxhQoZWB/L3RrW5EvAp5dQ99cp4ps3r2ssmtbkIKmki60cL SOTu1DjfMJWStLuTUsmVyPbfM1a3cH339XVcv7yjWlqHLVkX5yovsTEjHepTofe5faYC enHdwtCNqTxr58oDh0huN2GwyhoRjjwS6CX+XEJ0nLFjgXsvZ+a92VwDtbCXYwflmsbG R7DQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=/ApQFqmcouCl5STIusqfuTJXZzGokXhI+Lz0vLMrzp0=; b=AQAATxXVLcL5xhuUvbo9pXRl422ZLwp0NkgAAVfjp2d1/JQE+VVl+WQXi2YlC5Kk2J scgCXSBl2rTgzJXqF/C4uxe6JZSpJmqAuRXi3fVx2yxG8+6BHALTb0dPq2Zduunp/xqK 42p9UHOwEsz8G0lTGKzD6MFE4UMaiZxFPoZD8hHSYQG0ULUdPslUsDvgyLtsHngEbJSk utHoYpbJBZf1+HPy6YQ0xJAKAZM49VJgZj+g/1vKXFOcZHdbvUQIuIQyxcUhVlyqSlIh FFsNUCVMIfpkpMExca0TCbpiIcpBfSRGSDGQjD4xLsJtTyrpZCgVD7nfTCr3jsWoT/NI GkWg==
X-Gm-Message-State: AD7BkJJeqBQuWpZDaBDlviO4odkv6TK/4EIsK+52uMu9JzPqLMAtW1FnCAOv/HMfh5zRKw==
X-Received: by 10.98.80.150 with SMTP id g22mr47819317pfj.132.1458587611543; Mon, 21 Mar 2016 12:13:31 -0700 (PDT)
Received: from ?IPv6:2406:e007:4373:1:28cc:dc4c:9703:6781? ([2406:e007:4373:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id i9sm42491465pfi.95.2016.03.21.12.13.27 (version=TLSv1/SSLv3 cipher=OTHER); Mon, 21 Mar 2016 12:13:29 -0700 (PDT)
To: "Michael Behringer (mbehring)" , Duzongpeng , "anima@ietf.org" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com> <782ad410ef844630a942c27f6279a49f@XCH-RCD-006.cisco.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F047D8.2040303@gmail.com>
Date: Tue, 22 Mar 2016 08:13:28 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <782ad410ef844630a942c27f6279a49f@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 19:13:34 -0000

On 21/03/2016 22:31, Michael Behringer (mbehring) wrote:

...
>> Maybe we should think about levels of specialization. Top level, network
>> intent. Next level, intent for all ASAs supporting a given objective.
>> Those two levels would be flooded by GRASP. Next level, intent for a single
>> ASA. That would be transmitted by unicast GRASP synchronization.
> 
> Here I don't have the same view. The first two levels we agree on. Anything that's for a specific node doesn't qualify as "Intent" in my view. (Ignore that the flooding mechanism may contain unicast transmissions as part of the overall scheme). 

Yes, OK, I like Laurent's formulation of a continuum - with intent at one end
and traditional configuration of a single parameter in a single ode at the other
end.

> 
> If Intent is a high-level policy (which I think we all agree on?) then I can think of two more "things" that are NOT Intent, to add to my list of "not-Intent": 
> 
> - anything that's for a specific node
> - anything that's negotiated dynamically. (Intent is always originating from something non-technical - it's a policy).
> - anything that needs to change frequently (high-level policy is normally pretty stable)
> - anything that's not a "policy", ie doesn't come from a non-technical reasoning. (that's not a good way to define it, but can't think of a better one right now)
> 
> I think you're suggesting an optimisation of Intent *distribution*, Brian, right? Where, if I know only a few nodes run a certain function, I just send that particular Intent to those nodes only? (is that what you're thinking of?) 

Yes, but only when we are a long way down the continuum, so maybe that would be a
more traditional "policy" rather than "intent".

> 
> This would certainly be tempting, but I think it's a slippery slope, and would quickly get into conflict with current policy models. I'd prefer if we keep "our" Intent as "must be flooded everywhere". If it is really high level guidance, the optimization shouldn't be needed. Said differently: If we start pushing huge data models, maybe we're on the wrong track in the first place? 

I agree with that.

On 21/03/2016 22:46, Liubing (Leo) wrote:

> [Bing] For "ASAs supporting a given objective", flooding surely would work. But I'm wondering whether we also need a selective flooding mechanism to cut some unnecessary signaling amplification.
> When one node receive the intent, it only replicates the intent to the neighbors that fit for some conditions. This node behavior implies some assumption:
> 1) The intent content includes the "conditions" (e.g., flood to a specific tunnel; or flood to a certain set of rules etc.)
> 2) The nodes know about their neighbors' info to identify whether the neighbors fit the conditions.
> 
> Do you consider this as a complementary mechanism to normal flooding?

I think it's quite complicated, because it is essentially a special-purpose
multicast mechanism, where group membership is defined at application level
(for example, all nodes interested in objective "X"). At the moment we have
a two level mechanism (all GRASP nodes = flooding, one GRASP node = synchronization).

As Michael suggested, if the total intent data model is not too big, we can
simply flood it everywhere and nodes will select the relevant items.

    Brian


From nobody Mon Mar 21 12:28:17 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5702812DA81 for ; Mon, 21 Mar 2016 12:28:16 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ghCg_DjHyo49 for ; Mon, 21 Mar 2016 12:28:14 -0700 (PDT)
Received: from mail-pf0-x22f.google.com (mail-pf0-x22f.google.com [IPv6:2607:f8b0:400e:c00::22f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 84E9E12DA95 for ; Mon, 21 Mar 2016 12:28:14 -0700 (PDT)
Received: by mail-pf0-x22f.google.com with SMTP id x3so276137892pfb.1 for ; Mon, 21 Mar 2016 12:28:14 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=HD2ipGn5RBzy7hF4yC2QN08XGhem/1vxjqqEUxH5mHg=; b=GV7paZs9dm7WYe9ribGv+mpi3TOidAn8y+mY/Ps5xmbYGeRval95dFtQT2QHIAH/Fr E73aaFP/Utmfip2b7dTwQue24nVWdf7nSFQsCfjHCau4IOoj5DghMQq3ezi1+TE5pjPZ omN4b4fUuZ14RqMS52b9921K0S0fHxE9M5icl4KHkKsoF9/DPIMGNtkhfvjtX0i+XAbO cHCL5NUlz+4diao3I7nfIhdCwQmbADsWdEq5QQJOl/k7mCWTrqQhnCMdmaDHXkIIu5x0 JuLNLjz/zmATCRueen3AePOW5z84ZqeDdOpAyFgV+Qvkr6xqRzGeMb8dKWJbhLTV9A7p RByg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=HD2ipGn5RBzy7hF4yC2QN08XGhem/1vxjqqEUxH5mHg=; b=fJhROe+Tbh6fYEeOLdpxSkUkwjObzFy5awwx0rD4YElxpFfhzFCDubra6EZTHAaLLI Y5aKskpLAy6awnMuUP2fz99kyv73iTN1C63uGLtTyTyww7OAFyu7D7DMuTp3fr6pgEso 28X3KO9ipv/z4NpJ8o07ZMRa1BxTsderkK9pvHD4JrdoFHmOpDxYvnTtHalEzhkvBXC2 A2mpRmPiBZ1Jf7baFLbxDnxEa15SbyfeGbbpBvDReCYXbciLOBkIs2A63OV+0RgFr99u qc5Um0exsZTHOZewcf3To3u4nFD1SeScX9sdec7X8ZI7flD/wPO76kgrD5c7wv7BqW38 JODQ==
X-Gm-Message-State: AD7BkJIk5PXwqoAXViQfXs8h+cU0KoZLm58c/vTSN/D/fq18SgonrfErd9pd0Zef2O/aNQ==
X-Received: by 10.66.124.167 with SMTP id mj7mr47915407pab.153.1458588494072;  Mon, 21 Mar 2016 12:28:14 -0700 (PDT)
Received: from ?IPv6:2406:e007:4373:1:28cc:dc4c:9703:6781? ([2406:e007:4373:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id g23sm42563682pfg.35.2016.03.21.12.28.10 (version=TLSv1/SSLv3 cipher=OTHER); Mon, 21 Mar 2016 12:28:12 -0700 (PDT)
To: Laurent Ciavaglia , Michael Behringer , "Liubing (Leo)" , "anima@ietf.org" 
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F04B4B.1060703@gmail.com>
Date: Tue, 22 Mar 2016 08:28:11 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56EFE0AE.5040007@alcatel-lucent.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 19:28:16 -0000

On 21/03/2016 23:44, Laurent Ciavaglia wrote:
> Hello,
>
> I think that before considering encodings, we should first agree on the=
 definition(s) of what we will manipulate.

That is good systems engineering, but I do want to understand the implica=
tions
on the wire as well. For example, WSDL was very popular for a while but i=
s a
horribly inefficient wire format. So if we are going to design in yang, I=
 want
to be sure that it will look reasonable on the wire.

[Bing - from my prototype, I've learned that pretty much anything can be =
encoded
in CBOR if you want. If you can represent a data structure as an object, =
the
CBOR 'dumps' and 'loads' functions convert to and from CBOR bytes. That w=
orks
in all the likely programming languages.]

On 22/03/2016 00:53, Laurent Ciavaglia wrote:
> Michael, all,
>=20
> Thanks for providing your example!
>=20
> My first comment would be to consider "breaking" the model into parts.

I fully agree. This will also make the mechanics of flooding easier and
also make incremental updates possible.

    Brian

> Some clarifications:
>     -I don't think the ANIMA intent should be one single (big) file. Th=
is will be cumbersome to manage/update, and contrary to
> the PBM approach principles.
>     -Thus, an ANIMA network will have several, concurrent intents.
>=20
> Therefore, I think we should define:
>     -A format of the ANIMA intent that applies to the different target =
entities / roles.
>     -e.g. the ANI, the ASAs.
>=20
> Can/should the ANI intent and ASA intent use the same base model?
>     -> I think this will depend if they share enough in common.
>=20
> I will try to provide more details in the future (now busy with latest =
ID updates).
>=20
> Thanks, best regards, Laurent.
>=20
>=20
> On 21/03/2016 11:54, EXT Michael Behringer (mbehring) wrote:
>>
>> Agree; the encoding format question is secondary.
>>
>> My YANG example was just =E2=80=9Cplaying around=E2=80=9D, at this poi=
nt I=E2=80=99m trying to find the right structure. So don=E2=80=99t look =
at the details,
>> just the structure.
>>
>> Michael
>>
>> *From:*Laurent Ciavaglia [mailto:laurent.ciavaglia@nokia.com]
>> *Sent:* 21 March 2016 11:45
>> *To:* EXT Liubing (Leo) ; Michael Behringer (m=
behring) ; anima@ietf.org
>> *Subject:* Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some=
 thoughts
>>
>> Hello,
>>
>> I think that before considering encodings, we should first agree on th=
e definition(s) of what we will manipulate.
>> Let's start with a (formal) definiton of the "ANIMA" intent policy, an=
d the hierarchy envisaged and coarse "format".
>> If we can work out an example with we think capture well enough and la=
rge enough how and what the "ANIMA" is used for, then I
>> think we will already have made great progress.
>>
>> Best regards, Laurent.
>>
>> On 21/03/2016 11:02, EXT Liubing (Leo) wrote:
>>
>>     Hi Michael and all,
>>
>>     One quick comment aside from the Intent content discussion: since
>>     GRASP uses CBOR, isn't it good for Intent also encoded by CBOR?
>>
>>     YANG model can surely describe an Intent, but it includes lots of
>>     features for data manipulating (for Netconf), which might not
>>     necessary for Intent describing?.
>>
>>     Best regards,
>>
>>     Bing
>>
>>         -----Original Message-----
>>
>>         From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Micha=
el
>>
>>         Behringer (mbehring)
>>
>>         Sent: Friday, March 18, 2016 5:52 PM
>>
>>         To: anima@ietf.org 
>>
>>         Subject: [Anima] Intent yang model - some thoughts
>>
>>         WRT the current discussion on what should be in Intent and
>>         what shouldn't,
>>
>>         as mentioned before, I think we should leave this to the owner=

>>         of an
>>
>>         autonomic function to decide. This risks "abuse" in a way that=

>>         wasn't
>>
>>         foreseen, but it gives maximum agility, which is what we need
>>         to get ANIMA
>>
>>         off the ground, IMO. If we are in the nice position to find
>>         that it gets
>>
>>         adopted a lot, and if there are things that need fixing then,
>>         we can work on
>>
>>         version 2.
>>
>>         Several months ago I started with a yang model for Intent, but=

>>         never finished.
>>
>>         Maybe this helps with this discussion. Note I'm not at all a
>>         yang expert, so
>>
>>         please look more at the general structure for now. Not sure if=

>>         it helps the
>>
>>         discussion...
>>
>>         +--rw intent
>>
>>            +--rw meta-data
>>
>>            |  +--rw intent-version  uint32   // the version of the int=
ent
>>
>>         specification
>>
>>            |  +--rw timestamp            uint32   // latest time stamp=

>>         wins
>>
>>            |  +--rw administrator        string   // who created that
>>         intent
>>
>>            |  +--rw device?              string   // on which node was=

>>         it created
>>
>>            |  +--rw execution-time       uint32   // plus other
>>         parameters on when to
>>
>>         execute
>>
>>            |  +--rw expiration           uint32   // when does this
>>         intent cease to be
>>
>>         valid
>>
>>            |  +--rw feedback                 // how to let the
>>         operator know
>>
>>         about success/failure
>>
>>            |  +--rw signature            ??       // intent must be si=
gned
>>
>>            |
>>
>>            +--rw scope
>>
>>               +--rw domains
>>
>>                  +--rw domain* [name]
>>
>>                     +--rw name           string    // is there a fqdn
>>         type defined?
>>
>>                     +--rw meta-data           // general about the dom=
ain
>>
>>                     |  +--rw include-subdomains  boolean  // should
>>
>>         sub-domains follow this policy?
>>
>>                     |  +--rw fail-open  boolean  // fail-open or
>>         fail-closed
>>
>>         policy
>>
>>                     |  +--rw maintenance-window  // when can critical
>>
>>         operations take place
>>
>>                     |  +--rw ...                 // other general
>>         policies on
>>
>>         domain level
>>
>>         |        ...
>>
>>                     |
>>
>>                     +--rw ani-intent
>>
>>                     +-- (contains all policies which the ANI should
>>         follow)
>>
>>                     |  ex: ACP policies, coordination between AN
>>         functions,
>>
>>         security policies, etc
>>
>>                     |  for ani-intent, scope is always "all" (ie all
>>         nodes must
>>
>>         process)
>>
>>                     |  ...
>>
>>         ...
>>
>>         All of the above MUST be standardised, because it refers to th=
e AN
>>
>>         infrastructure itself.
>>
>>         Of the below we need to standardise the structure, but not
>>         content for each
>>
>>         function Conceptually, we define Intent for a domain.
>>
>>         ...
>>
>>                     |
>>
>>                     +--rw intent-for-autonomic-functions
>>
>>                        +--rw function_1 [name]
>>
>>                        |  +--rw scope                    // which devi=
ces
>>
>>         in the domain need to listen.
>>
>>                        |  |  +--rw all    boolean        // all device=
s
>>
>>                        |  |  +--rw role   string         // as defined=
 in
>>
>>         certificate
>>
>>                        |  |  +--rw dynamic_role  string  // as determi=
ned
>>
>>         dynamically
>>
>>                        |  |  +--rw tag    string         // can be
>>         used for
>>
>>         location
>>
>>                        |  +-- function_intent            // to be
>>         defined by
>>
>>         function owner
>>
>>                        |
>>
>>                        +--rw function_2 [name]
>>
>>                        |  +--rw scope                    // which devi=
ces
>>
>>         in the domain need to listen.
>>
>>                        |  |  +--rw all    boolean        // all device=
s
>>
>>                        |  |  +--rw role   string         // as defined=
 in
>>
>>         certificate
>>
>>                        |  |  +--rw dynamic_role  string  // as determi=
ned
>>
>>         dynamically
>>
>>                        |  |  +--rw tag    string         // can be
>>         used for
>>
>>         location
>>
>>                        |  +-- function_intent            // to be
>>         defined by
>>
>>         function owner
>>
>>                        |
>>
>>         ...
>>
>>         If a node receives Intent, it checks:
>>
>>         1) meta data: Mostly for forwarding and processing of Intent
>>
>>         2) domain info: domain specific policy
>>
>>         3) ANI Intent: Intent for the ANI itself; every node must be
>>         able to process
>>
>>         that
>>
>>         4) For each of the "function" blocks:
>>
>>            - should I run this function?
>>
>>            - do I run this function?
>>
>>            - if so, forward the rest of that block to the
>>         corresponding handler.
>>
>>         I think we want a clean architectural separation between the
>>         ANI and the
>>
>>         functions. This way every function can do what it wants.
>>
>>         Thoughts?
>>
>>         Michael
>>
>>         _______________________________________________
>>
>>         Anima mailing list
>>
>>         Anima@ietf.org 
>>
>>         https://www.ietf.org/mailman/listinfo/anima
>>
>>     _______________________________________________
>>
>>     Anima mailing list
>>
>>     Anima@ietf.org 
>>
>>     https://www.ietf.org/mailman/listinfo/anima
>>
>> --=20
>>
>> Laurent Ciavaglia
>>
>> Senior Research Manager
>>
>> Bell Labs, Nokia
>>
>> +33 160 402 636
>>
>> Route de Villejust | 91620 Nozay | France
>>
>> LinkedIn: laurentciavaglia 
>>
>=20
>=20
>=20
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>=20


From nobody Mon Mar 21 14:39:34 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6A9CF12D09A for ; Mon, 21 Mar 2016 14:39:33 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fI-0ZkswCge5 for ; Mon, 21 Mar 2016 14:39:31 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D808712D0B8 for ; Mon, 21 Mar 2016 14:39:18 -0700 (PDT)
Received: from fr712umx3.dmz.alcatel-lucent.com (unknown [135.245.210.42]) by Websense Email Security Gateway with ESMTPS id 488046ADF2E40; Mon, 21 Mar 2016 21:39:12 +0000 (GMT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (fr711usmtp1.zeu.alcatel-lucent.com [135.239.2.122]) by fr712umx3.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2LLdFXN021891 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 21 Mar 2016 21:39:15 GMT
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id u2LLdFdX029518 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 21 Mar 2016 22:39:15 +0100
Received: from [135.224.218.45] (135.239.27.41) by FR711WXCHHUB02.zeu.alcatel-lucent.com (135.239.2.112) with Microsoft SMTP Server (TLS) id 14.3.195.1; Mon, 21 Mar 2016 22:39:15 +0100
To: EXT Brian E Carpenter , "Michael Behringer (mbehring)" , Duzongpeng , "anima@ietf.org" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com> <782ad410ef844630a942c27f6279a49f@XCH-RCD-006.cisco.com> <56F047D8.2040303@gmail.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56F069FF.2030201@alcatel-lucent.com>
Date: Mon, 21 Mar 2016 22:39:11 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <56F047D8.2040303@gmail.com>
Content-Type: multipart/alternative; boundary="------------080209050103030806030909"
X-Originating-IP: [135.239.27.41]
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 21:39:33 -0000

--------------080209050103030806030909
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit


On 21/03/2016 20:13, EXT Brian E Carpenter wrote:
>
>> Here I don't have the same view. The first two levels we agree on. Anything that's for a specific node doesn't qualify as "Intent" in my view. (Ignore that the flooding mechanism may contain unicast transmissions as part of the overall scheme).
> Yes, OK, I like Laurent's formulation of a continuum - with intent at one end
> and traditional configuration of a single parameter in a single ode at the other
> end.

The credits go to Bob Natale: 
https://mailarchive.ietf.org/arch/msg/anima/QL-FKVtU-XjCCdWg6C-qDJ8WWyw

Best regards, Laurent.



--------------080209050103030806030909
Content-Type: text/html; charset="windows-1252"
Content-Transfer-Encoding: 7bit


  
    
  
  
    

    
On 21/03/2016 20:13, EXT Brian E
      Carpenter wrote:

    

    


      

        
Here I don't have the same view. The first two levels we agree on. Anything that's for a specific node doesn't qualify as "Intent" in my view. (Ignore that the flooding mechanism may contain unicast transmissions as part of the overall scheme). 


      

      
Yes, OK, I like Laurent's formulation of a continuum - with intent at one end
and traditional configuration of a single parameter in a single ode at the other
end.

    

    

    The credits go to Bob Natale:
    https://mailarchive.ietf.org/arch/msg/anima/QL-FKVtU-XjCCdWg6C-qDJ8WWyw

    

    Best regards, Laurent.

    

    

  


--------------080209050103030806030909--


From nobody Mon Mar 21 15:50:51 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8826D12D0E0 for ; Mon, 21 Mar 2016 15:50:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QFMiIkekSYXU for ; Mon, 21 Mar 2016 15:50:39 -0700 (PDT)
Received: from mail-lb0-x22a.google.com (mail-lb0-x22a.google.com [IPv6:2a00:1450:4010:c04::22a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 24F2E12D0FF for ; Mon, 21 Mar 2016 15:50:39 -0700 (PDT)
Received: by mail-lb0-x22a.google.com with SMTP id oe12so138852874lbc.0 for ; Mon, 21 Mar 2016 15:50:39 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=NbV52CyW6i7RfNtzR+FKPWFE/ZY3xZx11/+gvNwi+CA=; b=bBGx/RG1t4AzNOG25ZSnXYreMVXyxZ7RbhJhmnII8Np1XaJXtpGC8ybHfr6JUwpuk3 KvJQFtYumprCu19AA1+wZnnHgI3Yn69BL16o5sKwVwMBMwPi13bnnovVd6OCYA2X0wuw b/ZPpsxO8S7yK6z6f8wdemdDEnVub24bB97RxcxFC4HfkUztOKHcreeGg2NC8CP+Kz6K D42xkr2KrJOlMQchXe/vr4ArajHuHZz2E5r5VrLE6+AQKYKSOX0Oqe6renntutjwJEmm vqc4dXGd4GVTV47n5LA1M8ghiiURh0Iu4vYncozIWIHuC9cLGxqiaiFaRmqMQiXEwjEU Z9PA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=NbV52CyW6i7RfNtzR+FKPWFE/ZY3xZx11/+gvNwi+CA=; b=AaAzXx+vZR2QV4HGc8ErEcyTv9YL7Z7buMfzMmhQcPqXfnqkps2MqCJzFXtXC5Pte5 1s872oCccfXvOfhyX0rpOY1avsGimhj0e8tBRKiT5oQGYgl7Jp0e0SNdYMVdIAsA7uyo SBAQs44dZG/I4qsQ9iatRJdgdQrISJrAYq7916knzQgWP7D6GWSkgw+GmUxIznh1VWUp 8J+KdApVvbIW89BKRmd8UMQ7DX2XowFco3AcHspe2BsHjFOpmDuJElXwSJPXrVuq48E4 7O2vh3HX9YEQ7qOOyNci83ZkhVjNAWlM+ykvHWJbKapHEESpV/yLCt990opZi0J0k5vo 3dVw==
X-Gm-Message-State: AD7BkJLH1LmpuQQ4K/O7tHy9PND15n39nmOkthP1W9w9CHi2nKaaEyPUPTbKrLC9AzsIOs6OmG1c1C8jXvb71Q==
MIME-Version: 1.0
X-Received: by 10.25.143.141 with SMTP id r135mr12993119lfd.100.1458600637129;  Mon, 21 Mar 2016 15:50:37 -0700 (PDT)
Received: by 10.25.23.219 with HTTP; Mon, 21 Mar 2016 15:50:37 -0700 (PDT)
In-Reply-To: 
References: 
Date: Mon, 21 Mar 2016 15:50:37 -0700
Message-ID: 
From: John Strassner 
To: Duzongpeng , John Strassner 
Content-Type: multipart/alternative; boundary=001a114016e4d8697e052e96ec5e
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 22:50:43 -0000

--001a114016e4d8697e052e96ec5e
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

 I understand that Autonomic network should be able to work with as
less configuration as possible.


Why is this? The only reason I could imagine for this is if the components
of the autonomic network were each significantly more intelligent than
current components, and hence, could reason about what needed to
change in their configuration. However, this doesn't necessarily mean
that the **internal** configuration is less complicated. Rather, it implies
that there should be less communication about configuration changes.
Right?


In the ideal case, an Autonomic network should be able to work well
while only intent is needed from human operators.


This depends on what you mean by intent. With all due respect, we
first need to tighten the definition of intent rather than speculate what
to do with something we don't all agree about.



         However, I wonder whether Autonomic network may need
some intervention from human operators through some low-level
policies beyond intents.


I think that this is implementation-specific. For example, in OpenStack
Congress, we use an extended version of datalog. There is no reason
in general to use anything else.

That being said, I want to agree, but would like to understand your
reasoning in more detail before I say "yes".



         For example, in the
 https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00
,
it is suggested that the prefix lengths for the CSG, ASG, RSG
(different roles in IP RAN) can be assigned as an "intent". These
configuration parameters need to be distributed in the autonomic
domain to influence the detail configurations on each autonomic
node.


I am missing what is intent here. This is just a simple mapping of
a role to something. We could do the exact same thing without
using intent.

In fact, I'm a bit worried that this is not a good application of
intent. How many users will know what a prefix is? :-) At least in
the ONF and the TMF, we have decided that intent really belongs
in the higher layers (at the user, business person, and app
developer layers), and not on the device.



         As intent is described as an abstract, declarative, high-level
policy used to operate an autonomic domain, such as an enterprise
network. Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the
=E2=80=9Cprefix-management ID=E2=80=9D to some other words.


I think that this would be a good idea...



Is =E2=80=9CASA parameters" ok here?



I think some characteristics for this kind of low-level policies are

1.      Network level parameters, need to be distributed by GRASP and ACP

2.      Relatively static compared to intent, perhaps only needed to be
configured once

3.      Unavoidable in current network environment, mostly for establishing
the network infrastructure
4.      Rarely need coordinations with others parameters, usually is closed
related to one specific autonomic function


If we want to talk about translating intent, we should
probably look at putting this in either the anima
reference model (or better, mention it there and then
point to a separate draft), or start a new draft.
Note that we have had discussions in SUPA, as well
as in the ONF and TMF, about such a function.

If there is sufficient interest, I could write up an I-D.


regards,
John



On Tue, Mar 15, 2016 at 6:37 PM, Duzongpeng  wrote:

> Hello, everyone
>
>
>
>          I understand that Autonomic network should be able to work with
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
>
>
>
>          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
>
>
>
>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
>
>
>
>          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
>
>
>
> Is =E2=80=9CASA parameters" ok here?
>
>
>
> I think some characteristics for this kind of low-level policies are
>
> 1.      Network level parameters, need to be distributed by GRASP and ACP
>
> 2.      Relatively static compared to intent, perhaps only needed to be
> configured once
>
> 3.      Unavoidable in current network environment, mostly for
> establishing the network infrastructure
>
> 4.      Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
>
>
>
> Welcome for comment. Your suggestions will be appreciated.
>
>
>
>
>
> Best Regards
>
> Zongpeng Du
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>


--=20
regards,
John

--001a114016e4d8697e052e96ec5e
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


=C2=A0I understand that Autonomic network should be a=
ble to work with as
less configuration as possible.
<jcs>
Why is this?=C2=A0The only reason I could=
 imagine for this is if the components
of the autonomic network w=
ere each significantly more intelligent than
current components, =
and hence, could reason about what needed to
change in their conf=
iguration. However, this doesn't necessarily mean
that the **=
internal** configuration is less complicated. Rather, it implies
=
that there should be less communication about configuration changes.
<=
div>Right?
</jcs>

In the ideal cas=
e, an Autonomic network should be able to work well
while only in=
tent is needed from human operators.

<jcs>
This depends on what you mean by intent. With all due respect, we<=
/div>
first need to tighten the definition of intent rather than specul=
ate what
to do with something we don't all agree about.
=

</jcs>
=
=C2=A0
=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 However, I wonder whether Aut=
onomic network may need
some intervention from human operators through some low-level=

policies beyond intents=
.

<jcs>
I think that=
 this is implementation-specific. For example, in OpenStack
Congress, we use an extended ver=
sion of datalog. There is no reason
in general to use anything else.

That being said, I want to agree, but would like to=
 understand your
=
reasoning in more detail before I say "yes".
</jcs>

=C2=A0
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Fo=
r example, in the
=C2=A0https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,<=
/div>
it is suggested that the=
 prefix lengths for the CSG, ASG, RSG
=
(different roles in IP RAN) can be assigned as an &quo=
t;intent". These
configuration parameters need to be distributed in the autonomic
domain to influence th=
e detail configurations on each autonomic
node.

<jcs>
I a=
m missing what is intent here. This is just a simple mapping of
a role to something. We coul=
d do the exact same thing without
using intent.

In fact, I'm a bit worried that this is not a good application of
intent. How many use=
rs will know what a prefix is? :-) At least in
the ONF and the TMF, we have decided that int=
ent really belongs
in the higher layers (at the user, business person, and app
<=
div class=3D"MsoNormal">developer layers), and not on =
the device.
</=
jcs>

=C2=A0
=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 As intent is described as an abstract, declarat=
ive, high-level
p=
olicy used to operate an autonomic domain, such as an enterprise
network. Perhaps we can ren=
ame the =E2=80=9Cintent=E2=80=9D in the 
=E2=80=9Cprefix-management ID=E2=80=9D to some othe=
r words.

<jcs>
I think that this would b=
e a good idea...
=
</jcs>
=C2=A0

Is =E2=
=80=9CASA parameters" ok here? 
=C2=A0=

I think some characteristics for this kind of low-level policies=
 are
1.=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 Network level paramete=
rs, need to be distributed by GRASP and ACP
2.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Relatively static compared to intent, perhaps only needed t=
o be configured once
=
3.=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 Unavoid=
able in current network environment, mostly for establishing the network in=
frastructure
4.=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0 Rarely need c=
oordinations with others parameters, usually is closed related to one speci=
fic autonomic function

<jcs>
If we want to talk about tr=
anslating intent, we should
probably look at putting this in either the anima
reference model (or better, men=
tion it there and then
point to a separate draft), or start a new draft.
Note that we have had discussions in=
 SUPA, as well
as=
 in the ONF and TMF, about such a function.

If there is sufficient interest, I could write up an I-D.
</jcs>

regar=
ds,
John



On Tue, Mar 15, 2016 at 6:37 PM, Duzongpeng <<=
a href=3D"mailto:duzongpeng@huawei.com" target=3D"_blank">duzongpeng@huawei=
.com> wrote:











Hello, everyone


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 I understand that Autonomic network should be able to wo=
rk with as less configuration as possible. In the ideal case, an Autonomic =
network should be able to work well while only intent is needed from human =
operators.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 However, I wonder whether Autonomic network may need som=
e intervention from human operators through some low-level policies beyond =
intents.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 For example, in the 
https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it i=
s suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in IP RAN) can be assigned as an "intent". These configuration pa=
rameters need to be distributed in the autonomic
 domain to influence the detail configurations on each autonomic node.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 As intent is described as an abstract, declarative, high=
-level policy used to operate an autonomic domain, such as an enterprise ne=
twork. Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cp=
refix-management ID=E2=80=9D to some
 other words.


=C2=A0


Is =
=E2=80=9CASA parameters" ok here?



<=
/u>=C2=A0


I th=
ink some characteristics for this kind of low-level policies are<=
/u>



1.=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0
Network level parameters, =
need to be distributed by GRASP and ACP



2.=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0
Relatively static compared=
 to intent, perhaps only needed to be configured once<=
/p>


3.=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0
Unavoidable in current net=
work environment, mostly for establishing the network infrastructure=




4.=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0
Rarely need coordinations =
with others parameters, usually is closed related to one specific autonomic=
 function


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 


Welc=
ome for comment. Your suggestions will be appreciated.=



<=
/u>=C2=A0


<=
/u>=C2=A0


Best Regards


Zongpeng Du







_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima





-- 
regards,
John




--001a114016e4d8697e052e96ec5e--


From nobody Mon Mar 21 16:45:01 2016
Return-Path: 
X-Original-To: anima@ietf.org
Delivered-To: anima@ietfa.amsl.com
Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 4158D12D176; Mon, 21 Mar 2016 16:44:59 -0700 (PDT)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
From: internet-drafts@ietf.org
To: 
X-Test-IDTracker: no
X-IETF-IDTracker: 6.17.0
Auto-Submitted: auto-generated
Precedence: bulk
Message-ID: <20160321234459.10893.4388.idtracker@ietfa.amsl.com>
Date: Mon, 21 Mar 2016 16:44:59 -0700
Archived-At: 
Cc: anima@ietf.org
Subject: [Anima] I-D Action: draft-ietf-anima-autonomic-control-plane-02.txt
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Mon, 21 Mar 2016 23:44:59 -0000

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Autonomic Networking Integrated Model and Approach of the IETF.

        Title           : An Autonomic Control Plane
        Authors         : Michael H. Behringer
                          Steinthor Bjarnason
                          Balaji BL
                          Toerless Eckert
	Filename        : draft-ietf-anima-autonomic-control-plane-02.txt
	Pages           : 32
	Date            : 2016-03-21

Abstract:
   Autonomic functions need a control plane to communicate, which
   depends on some addressing and routing.  This Autonomic Control Plane
   should ideally be self-managing, and as independent as possible of
   configuration.  This document defines an "Autonomic Control Plane",
   with the primary use as a control plane for autonomic functions.  It
   also serves as a "virtual out of band channel" for OAM communications
   over a network that is not configured, or mis-configured.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-anima-autonomic-control-plane/

There's also a htmlized version available at:
https://tools.ietf.org/html/draft-ietf-anima-autonomic-control-plane-02

A diff from the previous version is available at:
https://www.ietf.org/rfcdiff?url2=draft-ietf-anima-autonomic-control-plane-02


Please note that it may take a couple of minutes from the time of submission
until the htmlized version and diff are available at tools.ietf.org.

Internet-Drafts are also available by anonymous FTP at:
ftp://ftp.ietf.org/internet-drafts/


From nobody Mon Mar 21 20:36:37 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E263712D5C1 for ; Mon, 21 Mar 2016 20:36:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zme1yvMtqevl for ; Mon, 21 Mar 2016 20:36:22 -0700 (PDT)
Received: from mail-pf0-x22b.google.com (mail-pf0-x22b.google.com [IPv6:2607:f8b0:400e:c00::22b]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0C71B12D685 for ; Mon, 21 Mar 2016 20:36:22 -0700 (PDT)
Received: by mail-pf0-x22b.google.com with SMTP id u190so291812311pfb.3 for ; Mon, 21 Mar 2016 20:36:21 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=M8SX7bzfdZjrFiabJuM+MIt9phGy65iYH8aiaJ79Tr8=; b=LAZI00b6r7wm15LMo4FI/XbYIbE/5ng9mQvxNgQPlX/tD3qVqTGrHgcMvCwg387v8W E60fJtm318FzZdVma/yb9PbwQFkMH4ZiAq9j5+/1zI3qxRyKYvfkTyeb4K1gEFGqPUWz KCbA3sSxkJqCnpBNNL40oFv947RKQ95AW+lPlkAr07mh8Q+5Hgt2Q81juzj5PrGoXYnK AWO1nHs84Xuq1EpkAumBb4m9vFrIXEzxirxFN0Mkc/2hOxkhxhbTgEbYJhkSpF4lsiSo 9c3bTRblIXH7YdG/sFiVfTxfCNq66TtlwDY7QLdAmZ7v1M8uFdwiZOt/xKucL7UpyOmZ hsxg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=M8SX7bzfdZjrFiabJuM+MIt9phGy65iYH8aiaJ79Tr8=; b=l48XkZLXMGIvXmF6jpsu2VtkJbHFjyHOriLSQItpbtLbOI7rHeB8oY132uEC5UEPi9 5wpp6FQWYZqpJI10V1dC48orAMIHCKm4FN3+OaVu/yU7tZ9nSR0Oua9zSQ0tXoGuVDby +jULeL01nnCHg+2YhqxfQkGqlnzh3NlwJeOnR51fXO2PxtpNx8kZIJKeVm0DQ850Kkoq jO1n5w+V1RmUPEqI5VydyeqfixpUSba9QeBvri70uvIB53ogk+tKqk81Sc/zzPM7iWf6 FFquwVaN+y8sXqs8SOu6Ss/l8KBqOjy6WRw8ez7B5VXkzNO1i2W+YMgw4oxOpN2aeTUK NHCg==
X-Gm-Message-State: AD7BkJJiqgAt2T7muvfY7ITb7CHqrQz/CzNaQMTsxI843Ppcf2fbKZzX2T4rTeCnPHNhdw==
X-Received: by 10.66.248.168 with SMTP id yn8mr50381679pac.24.1458617781486; Mon, 21 Mar 2016 20:36:21 -0700 (PDT)
Received: from ?IPv6:2406:e007:4373:1:28cc:dc4c:9703:6781? ([2406:e007:4373:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id v3sm44011299par.17.2016.03.21.20.36.18 (version=TLSv1/SSLv3 cipher=OTHER); Mon, 21 Mar 2016 20:36:19 -0700 (PDT)
To: Michael Richardson , "anima@ietf.org" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com> <485.1458564774@obiwan.sandelman.ca>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F0BDB4.1040904@gmail.com>
Date: Tue, 22 Mar 2016 16:36:20 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <485.1458564774@obiwan.sandelman.ca>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: Re: [Anima] need for some ASA definitions definitions to help bootstrap
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 22 Mar 2016 03:36:35 -0000

A few questions and comments:

On 22/03/2016 01:52, Michael Richardson wrote:
> 
> The ANIMA bootstrap work is coming along.  The question of how, prior to
> secure bootstrap, the New Node (the pledge) finds the proxy has been resolved
> with the use of dnssd/mDNS-service-discovery on the link local addresses.
> This has been selected on the basis of code reuse for devices which need/want
> to bootstrap, but which might never participate in the ACP itself.  Please
> see our draft.
> 
> There remains two other things that need to discover which we do not cover,
> which are more ANIMA general problems.  These are:
>   1) the "discovery" ("revelation"?) of a node that it needs to become
>      a proxy.
> 
>   2) the discovery of the proxy node of the registrar.

Does that mean that there is exactly one registrar per AN?

> These are things that *could* be determined by "power-on-above", i.e. a
> controller in the NOC could push (netconf-style or otherweise) configuration
> into a node to tell it, "You will be a bootstrap proxy on interfaces A,C and
> Q, and the registrar is at address BAR"

That doesn't really seem right, because it assumes that the NOC has
perfect knowledge of the topology. Potentially, any node with multiple
link interfaces could be a proxy, and if the topology is non-redundant,
they would all need to be proxies.

> 
> Or these things could be discovered via ASA.
> 
> number (2) -- discoverying the registrar by the proxy node should be a
>        relatively straight forward application of GRASP over the secured
>        ACP to discover an ASA.

Quite easy. A few Python statements, including having the proxy
register itself with GRASP:

ok, asa_nonce = grasp.register_asa("AN_Proxy")
if not ok:
    exit() #Oops, another proxy is already registered
obj = grasp.objective("AN_Registrar")
obj.loop_count = 4
obj.synch = True
registrar_locators = grasp.discover(asa_nonce, obj, 5000)
if registrar_locators == []:
    exit() #Oops, there is no registrar

> 
> number (1) -- whether to become a proxy on a particular interface ought also
>        to be negotiated via the secured ACP, even though the proxy could
>        observe the mDNS-service-discovery process, the node which replies
>        could in fact be malicious.
>        So this is a less straightforward application of GRASP, it needs to
>        driven by an intent that says something like: "new nodes may join
>        network BAR on interface FOO"  for suitable abstractions of "FOO"
>        being device independant.

ACP nodes that are connected to multiple links will know that fact about
themselves, so they know they are potential proxies. Indeed, they could
always start up an ASA which is a *candidate* proxy and it would then
decode the network intent to figure out what to do next.

> There is also the ASA which would cause a node to turn itself into a
> registrar (or rather, enable a registrar container, process or VM) when it
> realizes that there is no other registrar, but I think that this is a higher
> level of independance of Autonomic than we are ready for yet.

Maybe, but is it all that scary? All the potential registrars could discover
each other and go through an election process. In fact it could even be
an intent option: elect_registrar = true.

   Brian

> Aside from socializing these requirements into the larger group, I think that
> we also looking for someone who might step forward to take the lead on these
> two documents.  Maybe someone new?
> 
> --
> Michael Richardson , Sandelman Software Works
>  -= IPv6 IoT consulting =-
> 
> 
> 
> 
> 
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
> 


From nobody Mon Mar 21 21:08:51 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7006512D122 for ; Mon, 21 Mar 2016 21:08:49 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.222
X-Spam-Level: 
X-Spam-Status: No, score=-4.222 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id R__Hgyeg_99g for ; Mon, 21 Mar 2016 21:08:46 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9C38D12D0F3 for ; Mon, 21 Mar 2016 21:08:45 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml702-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGK73844; Tue, 22 Mar 2016 04:08:43 +0000 (GMT)
Received: from nkgeml407-hub.china.huawei.com (10.98.56.38) by lhreml702-cah.china.huawei.com (10.201.5.99) with Microsoft SMTP Server (TLS) id 14.3.235.1; Tue, 22 Mar 2016 04:08:42 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml407-hub.china.huawei.com ([10.98.56.38]) with mapi id 14.03.0235.001; Tue, 22 Mar 2016 12:08:35 +0800
From: "Liubing (Leo)" 
To: Brian E Carpenter , Laurent Ciavaglia , Michael Behringer , "anima@ietf.org" 
Thread-Topic: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts
Thread-Index: AdGA84Aiy7EwHtVtTUWSCghLdxIQCwCYxtng//+J+gCAAALFgIAAEHEAgAB/GID//vA6sA==
Date: Tue, 22 Mar 2016 04:08:34 +0000
Message-ID: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com>
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com>
In-Reply-To: <56F04B4B.1060703@gmail.com>
Accept-Language: en-US, zh-CN
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.98.117]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020205.56F0C54C.0034, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 7e907c2345d3a058fd35fabae67a2320
Archived-At: 
Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 22 Mar 2016 04:08:49 -0000
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From nobody Mon Mar 21 23:44:42 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 263D412D68C for ; Mon, 21 Mar 2016 23:44:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.62
X-Spam-Level: 
X-Spam-Status: No, score=-2.62 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id f_e-K0dK_Lbv for ; Mon, 21 Mar 2016 23:44:39 -0700 (PDT)
Received: from relay4-d.mail.gandi.net (relay4-d.mail.gandi.net [217.70.183.196]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2C3EB12D1AD for ; Mon, 21 Mar 2016 23:44:39 -0700 (PDT)
Received: from mfilter44-d.gandi.net (mfilter44-d.gandi.net [217.70.178.175]) by relay4-d.mail.gandi.net (Postfix) with ESMTP id F20651720B3; Tue, 22 Mar 2016 07:44:37 +0100 (CET)
X-Virus-Scanned: Debian amavisd-new at mfilter44-d.gandi.net
Received: from relay4-d.mail.gandi.net ([IPv6:::ffff:217.70.183.196]) by mfilter44-d.gandi.net (mfilter44-d.gandi.net [::ffff:10.0.15.180]) (amavisd-new, port 10024) with ESMTP id h-aFYLKozOS6; Tue, 22 Mar 2016 07:44:36 +0100 (CET)
X-Originating-IP: 93.199.229.85
Received: from nar.local (p5DC7E555.dip0.t-ipconnect.de [93.199.229.85]) (Authenticated sender: cabo@cabo.im) by relay4-d.mail.gandi.net (Postfix) with ESMTPSA id 667C8172095; Tue, 22 Mar 2016 07:44:34 +0100 (CET)
Message-ID: <56F0E9CF.7070602@tzi.org>
Date: Tue, 22 Mar 2016 07:44:31 +0100
From: Carsten Bormann 
User-Agent: Postbox 4.0.8 (Macintosh/20151105)
MIME-Version: 1.0
To: "Liubing (Leo)" 
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com>
In-Reply-To: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com>
X-Enigmail-Version: 1.2.3
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Archived-At: 
Cc: Laurent Ciavaglia , "anima@ietf.org" , Michael Behringer 
Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 22 Mar 2016 06:44:41 -0000

YANG is a data modeling language for data at rest (e.g., in a management
data repository); it is not designed for describing protocol data units
(we recently have been using CDDL for the latter).

https://tools.ietf.org/html/draft-greevenbosch-appsawg-cbor-cddl-08

YANG data from (or going into) a repository can be represented on the
wire in XML (native RFC 6020), JSON, and now also in CBOR:

https://tools.ietf.org/html/draft-ietf-netmod-yang-json-09
https://tools.ietf.org/html/draft-veillette-core-yang-cbor-mapping-00

Indeed, the COMI project in the CoRE working group is about enabling
RESTCONF style transactions on YANG-modeled data within constrained
devices, employing CBOR representation as one of the crucial
optimizations to make this possible.

Grüße, Carsten


From nobody Tue Mar 22 12:02:35 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 83B8512D572 for ; Tue, 22 Mar 2016 12:02:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id dg-KlHA4mdgO for ; Tue, 22 Mar 2016 12:02:32 -0700 (PDT)
Received: from mail-pf0-x229.google.com (mail-pf0-x229.google.com [IPv6:2607:f8b0:400e:c00::229]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id ED7A512D8EA for ; Tue, 22 Mar 2016 12:01:49 -0700 (PDT)
Received: by mail-pf0-x229.google.com with SMTP id x3so322587156pfb.1 for ; Tue, 22 Mar 2016 12:01:49 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=mvr7MOnb/vD6lpsd9AqkH87SgdtnWc2Yzj9lijLXcPM=; b=E6z0h4nDtHZlE7tkGC2cgvmZlDpE9PIbdUqR3dfVSV9kcJgGCm8kOLPrE5JgOMXaw2 MUxEi6x2e1cXfDraoKZMVZBvuAZ8r/hE3B5V18Esmo8SbL9pkchFtHvnWsXX7LwlNSMb aD1OIbvg4d+IhpJxbaROU3BoqHGGS6MqBcalH8teUQhF7JK/lri5bM0nmzc1hzcZCuYz sZGK2vumhN7ubO9GoV78OV0nbnSW0NuDDxgfgVKxZVAMz4xNBHQThPLavlkmEbmCantB R0DSTakOA/RXM+13UaBax7tdyvMnWhsPEEVG34k/HI1n2AbAbaa0TVckaGYP3HOsA/01 p8Nw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=mvr7MOnb/vD6lpsd9AqkH87SgdtnWc2Yzj9lijLXcPM=; b=K/Lsu2InCe3ml8dN91sC8vehtU3BlXOWPNQIv31IPrz24KhTpjzQKeOF5cqUccIUw1 mWxvjvP1Y+34nbMzxjD9qgy1XtAcWA3ElvGVywYw0F+aa5LFrx9GZLd4N5qtVyk47cjt TrJvBezzDLOqQpmm6hRQHg+5ESLlFhqpSr693C3CDCgkCn4aag5RH9uLyL2aY+KYJtL5 hzpwk9+vXGsxH6dPzmrLi/cafrmOZB0TWFzF9EGUCJkpKA0a/gNxc3i5sVtpYNa9LazQ D5A+RW36VSqdfZzSOXWskx9O/ZVlE5Jame3kQMCt8tdCqiEXkahDLGKPVmnx4TBRWxeg l6Zw==
X-Gm-Message-State: AD7BkJLjPUUzTK8Co/Iasop93aT+QJ5s3eBred8CNwhGRPjBcv2dnZO7SeSMRNkP9UGYhQ==
X-Received: by 10.66.120.202 with SMTP id le10mr56106334pab.155.1458673309542;  Tue, 22 Mar 2016 12:01:49 -0700 (PDT)
Received: from ?IPv6:2406:e007:589e:1:28cc:dc4c:9703:6781? ([2406:e007:589e:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id 9sm50090611pfm.10.2016.03.22.12.01.46 (version=TLSv1/SSLv3 cipher=OTHER); Tue, 22 Mar 2016 12:01:48 -0700 (PDT)
To: Carsten Bormann , "Liubing (Leo)" 
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com> <56F0E9CF.7070602@tzi.org>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F1969D.1060804@gmail.com>
Date: Wed, 23 Mar 2016 08:01:49 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56F0E9CF.7070602@tzi.org>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Laurent Ciavaglia , "anima@ietf.org" , Michael Behringer 
Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 22 Mar 2016 19:02:34 -0000

Thanks Carsten. draft-veillette-core-yang-cbor-mapping looks like
the correct solution in the GRASP context, if someone provides the
appropriate run-time support code.

IMHO using the XML or JSON formats would be a mistake.

   Brian

On 22/03/2016 19:44, Carsten Bormann wrote:
> YANG is a data modeling language for data at rest (e.g., in a managemen=
t
> data repository); it is not designed for describing protocol data units=

> (we recently have been using CDDL for the latter).
>=20
> https://tools.ietf.org/html/draft-greevenbosch-appsawg-cbor-cddl-08
>=20
> YANG data from (or going into) a repository can be represented on the
> wire in XML (native RFC 6020), JSON, and now also in CBOR:
>=20
> https://tools.ietf.org/html/draft-ietf-netmod-yang-json-09
> https://tools.ietf.org/html/draft-veillette-core-yang-cbor-mapping-00
>=20
> Indeed, the COMI project in the CoRE working group is about enabling
> RESTCONF style transactions on YANG-modeled data within constrained
> devices, employing CBOR representation as one of the crucial
> optimizations to make this possible.
>=20
> Gr=C3=BC=C3=9Fe, Carsten
>=20


From nobody Tue Mar 22 13:32:35 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D522512D99E for ; Tue, 22 Mar 2016 13:32:32 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Rg2qIdCzZ08A for ; Tue, 22 Mar 2016 13:32:30 -0700 (PDT)
Received: from mail-lf0-x22f.google.com (mail-lf0-x22f.google.com [IPv6:2a00:1450:4010:c07::22f]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4B1E312D916 for ; Tue, 22 Mar 2016 13:32:29 -0700 (PDT)
Received: by mail-lf0-x22f.google.com with SMTP id e196so44146512lfg.1 for ; Tue, 22 Mar 2016 13:32:29 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=WkPqC71y+6YfgzkR3lNCLHoz+GRCmOwnLdNX3Cx+O5Q=; b=Q/5FlQ+08PYxf6WfH7WRzrNWujH+WoBGYE4xOB3jFc9WY+ASJHz4BmXKI49/w/Q71d 48iiUCZ5O93xopvPEgmuKuxGvI0Nk9bSfrnVgWAJhxibAs8kiTif1ThnebJMht1BNKnx wGpXgYHKG2mLrJjnPHg2nxK/A/M4FybgAlQBf52a5xzydF8ueAJA+jh0QST3jl358mxu 6TjFCc/Ta4XS1wMfpmpcJTi+lWtaoJemeuRSNbOLKsyXRwMGbyF6z1gSVIb3X4EDNbZE JQF6t7CAfTeI6iC2PAzj6Dr+xRhSPU94DruZkoG95RLgCcgvzEvrcvgi61qpY7HkxrCv IXiQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=WkPqC71y+6YfgzkR3lNCLHoz+GRCmOwnLdNX3Cx+O5Q=; b=hLP3SvVWGaVfEFD1mqGdLPg2ppGwWIn3MSnNV21LLJNvd0FXarIA6DGxMx6EhChq0a mYhtV/GdzLrpAVFnq/nljQlTPwqTk0JVY4KQO6qzQKRJhbCkro7CXdLBVviNZV9UcEVo aMTeEqQ70o4lLWuvDgY1FF5oS+wUD5YtW8M2Wx0pUf8Uesj0ONRb2Xt84axn8rFgvc8j 3XNXtzetU+68aJLizc3+Vxr2HwqO3Wr8RBF5mb41q/O1FY+NGrna+0OVMUR+QuZw63Xm Ds969mS71JORk60qUFFnbWHpPszDgMHash/s3/nkeAVDg/GtVOXlRFcjcCgZHAQbopkf QZVQ==
X-Gm-Message-State: AD7BkJJCHziCNuCITZFjGFZEXKNqBZmmnZ9dK5Cur+qaOXsi+ot157483ekRAT70vmbKI93D6OoFI2Rob7Hivg==
MIME-Version: 1.0
X-Received: by 10.25.205.7 with SMTP id d7mr14288015lfg.70.1458678747396; Tue, 22 Mar 2016 13:32:27 -0700 (PDT)
Received: by 10.25.23.219 with HTTP; Tue, 22 Mar 2016 13:32:27 -0700 (PDT)
In-Reply-To: <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>
Date: Tue, 22 Mar 2016 13:32:27 -0700
Message-ID: 
From: John Strassner 
To: "Michael Behringer (mbehring)" , John Strassner 
Content-Type: multipart/alternative; boundary=001a1142091a948431052ea91cf9
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , Sheng Jiang , "anima@ietf.org" , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 22 Mar 2016 20:32:33 -0000

--001a1142091a948431052ea91cf9
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

In general, I agree with Michael. I'd like to emphasize and expand on a
couple of key points:


First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =
=E2=80=9Csteering=E2=80=9D a network, at
least not initially. There is still traditional config, plus all the
existing and emerging SDN models. My view has always been that ANIMA sets a
=E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are more=
 specific and
can override intent. This is described in RFC7575. Main point here: ANIMA
Intent isn=E2=80=99t the only tool in a network.



Absolutely agree. Note that even in the Congress case I mentioned
earlier, it is NOT pushing config, and it is NOT viewed as the
"only" policy



Intent could perfectly well cover a high level policy such as =E2=80=9Call =
nodes of
type x do this; type y does that=E2=80=9D. (Which I think is what you menti=
on
below.) When we started with Intent, the vision was to keep it very high
level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J



Agreed. Although to me, this means that we need to consider the types of
actors that are using intent. For example, most of my CCIE and JCIE friends
have no interest in using an intent language for pushing configuration. On
the other hand, when I talk to operators and application developers, they
would love to be able to use a high-level intent and have the intent engine
take care of the details for them.


It has become clear that in today=E2=80=99s networking, people will want to=
 push
more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I thin=
k we=E2=80=99ll just have to acknowledge
that.


Sorry Michael, I don't follow here. What do you mean
by "parameters"?



Personally I think it=E2=80=99ll pan out like this: Intent will have sectio=
ns per
autonomic function. An Intent parser will validate overall correctness
(signature, etc), then segment the Intent into its bits: One part will be
the autonomic fabric itself, then there will be bits for each autonomic
function.


Do you mean that the intent writer will **know the difference**
between the fabric and each autonomic function? I hope not,
but perhaps I'm misunderstanding you...



So the Intent parser on a node will grab intent, see there is some Intent
for autonomic function A, will see whether this is running locally, and if
so, it=E2=80=99ll just pass that piece of Intent to that function.


Agreed.



In this way, a function owner defines both the intent itself, and how it is
consumed by his/her function.


I'm not sure what you mean by "function owner". (Sorry to be
picky, I'm likely under-caffeinated.)

The above seemed like an autonomous decision, made by the parser.
I don't think that the intent writer should be cognizant of how many or
what types of autonomic functions are present.



The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can d=
o what they
want, in the worst case even push config, which is clearly NOT what Intent
is meant to do.


And hence, the "HAL 2000" scenario. :-)



I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=80=
=9Cnode 17: here is your
CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D=
. But what I
describe above COULD be abused in this way. RFC7575 says this is NOT
Intent.


Absolutely agree!


 regards,

John





On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
mbehring@cisco.com> wrote:

> Hi Zongpeng,
>
>
>
> Yes, this is a recurring question and topic. Let me give my view on two
> levels:
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J  It has=
 become clear
> that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writing =
Intent like =E2=80=9Cnode
> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here i=
s your
> CLI=E2=80=9D. But what I describe above COULD be abused in this way. RFC7=
575 says
> this is NOT Intent.
>
>
>
> In other words: Let=E2=80=99s not be too prescriptive, and hope that folk=
s will do
> more or less reasonable things.
>
>
>
> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly co=
ntribute,
> I=E2=80=99m just struggling with priorities at the moment (as you can see=
 by my
> relative silence over the past weeks).
>
>
>
> Michael
>
>
>
>
>
>
>
> *From:* Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Duzongpeng
> *Sent:* 16 March 2016 02:38
> *To:* anima@ietf.org
> *Cc:* Laurent Ciavaglia ; J=C3=A9ferson Camp=
os
> Nobre ; Sheng Jiang 
> *Subject:* [Anima] Does Autonomic network need some intervention beyond
> intent
>
>
>
> Hello, everyone
>
>
>
>          I understand that Autonomic network should be able to work with
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
>
>
>
>          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
>
>
>
>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
>
>
>
>          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
>
>
>
> Is =E2=80=9CASA parameters" ok here?
>
>
>
> I think some characteristics for this kind of low-level policies are
>
> 1.       Network level parameters, need to be distributed by GRASP and AC=
P
>
> 2.       Relatively static compared to intent, perhaps only needed to be
> configured once
>
> 3.       Unavoidable in current network environment, mostly for
> establishing the network infrastructure
>
> 4.       Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
>
>
>
> Welcome for comment. Your suggestions will be appreciated.
>
>
>
>
>
> Best Regards
>
> Zongpeng Du
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>


--=20
regards,
John

--001a1142091a948431052ea91cf9
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


In general, I agree with Michael. I'd like to emp=
hasize and expand on a couple of key points:


<=
/div>
First, =
ANIMA isn=E2=80=99t the only thing =
=E2=80=9Cinfluencing=E2=80=9D or =E2=80=9Csteering=E2=80=9D a network, at l=
east not initially. There is still traditional config, plus all the existin=
g and emerging SDN models. My view has always been that ANIMA sets a =E2=80=9Cdefault=E2=80=9D policy in=
 a network, and all other means are more specific and can override intent. =
This is described in RFC7575. Main point here: ANIMA Intent isn=E2=80=99t the only tool in a network. 

<jcs>
Absolutely agre=
e. Note that even in the Congress case I mentioned
earlier, it is=
 NOT pushing config, and it is NOT viewed as the
"only"=
 policy
</jcs>


Intent could per=
fectly well cover a high level policy such as =E2=80=9Call nodes of type x =
do this; type y does that=E2=80=9D. (Which I think is what you mention belo=
w.) When we started with Intent, the vision was to keep it very high level,=
 more along the line =E2=80=9Cdo the right thing=E2=80=9D J=C2=A0

<jcs>
Agreed. Although to me, this =
means that we need to consider the types of actors that are using intent. F=
or example, most of my CCIE and JCIE friends have no interest in using an i=
ntent language for pushing configuration. On the other hand, when I talk to=
 operators and application developers, they would love to be able to use a =
high-level intent and have the intent engine take care of the details for t=
hem.
</jcs>

It has become clear that in today=E2=80=99s net=
working, people will want to push more =E2=80=9Cparameters=E2=80=9D than =
=E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowledg=
e that.

<jcs>
Sorry Michael, I don't follow here. What d=
o you mean
by "parameters"?
</jcs><=
/div>


Personally I think it=E2=80=99ll pan out like this: I=
ntent will have sections per autonomic function. An Intent parser will vali=
date overall correctness (signature, etc), then segment the Intent into its=
 bits: One part will be the autonomic fabric itself, then there will be bit=
s for each autonomic function. 

<jcs>
Do you mean that the intent writer will **know the=
 difference**
between the fabric and each autonomic function? I h=
ope not,
but perhaps I'm misunderstanding you...
</jcs>


So the Intent parser on a node will grab in=
tent, see there is some Intent for autonomic function A, will see whether t=
his is running locally, and if so, it=E2=80=99ll just pass that piece of In=
tent to that function. 

=

<jcs>
Agreed.
</jcs>


In this way, a function owner defines both the intent itself=
, and how it is consumed by his/her function. <=
/p>
<jcs>
I'm not sure what you mean by &=
quot;function owner". (Sorry to be
picky, I'm likely und=
er-caffeinated.)

The above seemed like an autonomo=
us decision, made by the parser.
I don't think that the inten=
t writer should be cognizant of how many or
what types of autonom=
ic functions are present.
</jcs>


The downside is that ther=
e is no =E2=80=9Ccontrol=E2=80=9D and people can do what they want, in the =
worst case even push config, which is clearly NOT what Intent is meant to d=
o. 

<jcs>
And hence, the &q=
uot;HAL 2000" scenario. :-)
</jcs>

<=
/div>

I=E2=
=80=99d hope that people don=E2=80=99t start writing Intent like =E2=80=9Cn=
ode 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here=
 is your CLI=E2=80=9D. But what I describe above COULD be abused in this wa=
y. RFC7575 says this is NOT Intent. 

<jcs>
Absolutely =
agree!
</jcs>

=
=C2=A0regards,
John
=C2=A0


On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (=
mbehring) <mbehring@cisco.com> wrote:











=
Hi Zongpeng,



=
=C2=A0


=
Yes, this is a recurrin=
g question and topic. Let me give my view on two levels:



=
=C2=A0


=
First, ANIMA isn=E2=80=
=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =E2=80=9Csteering=E2=
=80=9D a network, at least not initially. There is still traditional
 config, plus all the existing and emerging SDN models. My view has always =
been that ANIMA sets a =E2=80=9Cdefault=E2=80=9D policy in a network, and a=
ll other means are more specific and can override intent. This is described=
 in RFC7575. Main point here: ANIMA Intent isn=E2=80=99t
 the only tool in a network. 


=
=C2=A0


=
Intent could perfectly =
well cover a high level policy such as =E2=80=9Call nodes of type x do this=
; type y does that=E2=80=9D. (Which
 I think is what you mention below.) When we started with Intent, the visio=
n was to keep it very high level, more along the line =E2=80=9Cdo the right=
 thing=E2=80=9D
J=
=C2=A0
 It has become clear that in today=E2=80=99s networking, people will want t=
o push more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and=
 I think we=E2=80=99ll just have to acknowledge that.



=
=C2=A0


=
Personally I think it=
=E2=80=99ll pan out like this: Intent will have sections per autonomic func=
tion. An Intent parser will validate
 overall correctness (signature, etc), then segment the Intent into its bit=
s: One part will be the autonomic fabric itself, then there will be bits fo=
r each autonomic function.



=
=C2=A0


=
So the Intent parser on=
 a node will grab intent, see there is some Intent for autonomic function A=
, will see whether this
 is running locally, and if so, it=E2=80=99ll just pass that piece of Inten=
t to that function.



=
=C2=A0


=
In this way, a function=
 owner defines both the intent itself, and how it is consumed by his/her fu=
nction.



=
=C2=A0


=
The downside is that th=
ere is no =E2=80=9Ccontrol=E2=80=9D and people can do what they want, in th=
e worst case even push config, which
 is clearly NOT what Intent is meant to do. I=E2=80=99d hope that people do=
n=E2=80=99t start writing Intent like =E2=80=9Cnode 17: here is your CLI; n=
ode 23: here is your CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D. But =
what I describe above COULD be abused in this way. RFC7575 says this is
 NOT Intent. 


=
=C2=A0


=
In other words: Let=E2=
=80=99s not be too prescriptive, and hope that folks will do more or less r=
easonable things.



=
=C2=A0


=
I=E2=80=99ve been meani=
ng to comment on the Intent drafts and possibly contribute, I=E2=80=99m jus=
t struggling with priorities at the
 moment (as you can see by my relative silence over the past weeks). 


=
=C2=A0


=
Michael


=
=C2=A0


=
=C2=A0


=
=C2=A0








From: Anima [mailto:anima-bounces@ietf.org]
On Behalf Of Duzongpeng

Sent: 16 March 2016 02:38
To: anima@ietf.org

Cc: Laurent Ciavaglia <laurent.ciav=
aglia@nokia.com>; J=C3=A9ferson Campos Nobre <jcnobre@inf.ufrgs.br>; Sheng =
Jiang <jiangs=
heng@huawei.com>

Subject: [Anima] Does Autono=
mic network need some intervention beyond intent






=C2=A0


Hello, everyone

=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 I understand that Autonomic network should be able to work with as less=
 configuration as possible. In the ideal case, an Autonomic
 network should be able to work well while only intent is needed from human=
 operators.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 However, I wonder whether Autonomic network may need some intervention =
from human operators through some low-level policies beyond
 intents.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 For example, in the
https://tools.ietf.org/html/draft-ietf-anima-prefix-ma=
nagement-00, it is suggested that the prefix lengths for the CSG, ASG, =
RSG (different roles in IP RAN) can be assigned as an
 "intent". These configuration parameters need to be distributed =
in the autonomic domain to influence the detail configurations on each auto=
nomic node.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 As intent is described as an abstract, declarative, high-level policy u=
sed to operate an autonomic domain, such as an enterprise
 network. Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=
=9Cprefix-management ID=E2=80=9D to some other words.<=
/font>


=C2=A0


Is =E2=80=9CASA pa=
rameters" ok here?



=C2=A0


I think some chara=
cteristics for this kind of low-level policies are



1.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Network leve=
l parameters, need to be distributed by GRASP and ACP<=
/p>


2.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Relatively s=
tatic compared to intent, perhaps only needed to be configured once<=
u>



3.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Unavoidable =
in current network environment, mostly for establishing the network infrast=
ructure



4.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Rarely need =
coordinations with others parameters, usually is closed related to one spec=
ific autonomic function


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0



Welcome for commen=
t. Your suggestions will be appreciated.


=C2=A0


=C2=A0


Best Regards


Zongpeng Du









_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima





-- 
regards,
John




--001a1142091a948431052ea91cf9--


From nobody Tue Mar 22 20:46:45 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 58C3312D56A for ; Tue, 22 Mar 2016 20:46:44 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.222
X-Spam-Level: 
X-Spam-Status: No, score=-4.222 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3YQuuWtM0LDP for ; Tue, 22 Mar 2016 20:46:42 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 84F0D12D159 for ; Tue, 22 Mar 2016 20:46:41 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml701-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGL93330; Wed, 23 Mar 2016 03:46:39 +0000 (GMT)
Received: from NKGEML406-HUB.china.huawei.com (10.98.56.37) by lhreml701-cah.china.huawei.com (10.201.5.93) with Microsoft SMTP Server (TLS) id 14.3.235.1; Wed, 23 Mar 2016 03:46:38 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml406-hub.china.huawei.com ([10.98.56.37]) with mapi id 14.03.0235.001; Wed, 23 Mar 2016 11:46:33 +0800
From: "Liubing (Leo)" 
To: Brian E Carpenter , Carsten Bormann 
Thread-Topic: How we manage Intent?-//RE: [Anima] CBOR for Intent?-//RE: Intent yang model - some thoughts
Thread-Index: AdGA84Aiy7EwHtVtTUWSCghLdxIQCwCYxtng//+J+gCAAALFgIAAEHEAgAB/GID//vA6sIABzL2AgADOAID//wuqEA==
Date: Wed, 23 Mar 2016 03:46:32 +0000
Message-ID: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5E464@nkgeml514-mbx.china.huawei.com>
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com> <56F0E9CF.7070602@tzi.org> <56F1969D.1060804@gmail.com>
In-Reply-To: <56F1969D.1060804@gmail.com>
Accept-Language: en-US, zh-CN
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.98.117]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020201.56F2119F.011C, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 5cd976dcde300bbb4de59d56bc4750c9
Archived-At: 
Cc: Laurent Ciavaglia , "anima@ietf.org" , Michael Behringer 
Subject: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 03:46:44 -0000
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From nobody Wed Mar 23 01:55:28 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B137412DB6E for ; Wed, 23 Mar 2016 01:55:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Hn_HqxEhZnOr for ; Wed, 23 Mar 2016 01:55:25 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 945F012DB6B for ; Wed, 23 Mar 2016 01:55:24 -0700 (PDT)
Received: from fr712umx3.dmz.alcatel-lucent.com (unknown [135.245.210.42]) by Websense Email Security Gateway with ESMTPS id 95636792367EE; Wed, 23 Mar 2016 08:55:20 +0000 (GMT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (fr711usmtp1.zeu.alcatel-lucent.com [135.239.2.122]) by fr712umx3.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2N8tMqf013687 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Wed, 23 Mar 2016 08:55:22 GMT
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id u2N8tIpp027102 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 23 Mar 2016 09:55:22 +0100
Received: from [172.27.205.106] (135.239.27.41) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Wed, 23 Mar 2016 09:55:19 +0100
To: "EXT Liubing (Leo)" , Brian E Carpenter , Carsten Bormann 
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com> <56F0E9CF.7070602@tzi.org> <56F1969D.1060804@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5E464@nkgeml514-mbx.china.huawei.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56F259F6.5000807@alcatel-lucent.com>
Date: Wed, 23 Mar 2016 09:55:18 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5E464@nkgeml514-mbx.china.huawei.com>
Content-Type: multipart/alternative; boundary="------------090302050607060303000203"
X-Originating-IP: [135.239.27.41]
Archived-At: 
Cc: Laurent Ciavaglia , Michael Behringer , "anima@ietf.org" 
Subject: Re: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 08:55:27 -0000

--------------090302050607060303000203
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit

Hi Bong, All,

Please see below for my comments.

On 23/03/2016 04:46, EXT Liubing (Leo) wrote:
> Hi all,
>
> According to Carsten's education (thanks a lot!), and recent discussion, I realize we need to separate two things for Intent:
> 1. Encoding. Seems CBOR is a decent choice as discussed recently.
> 2. Modeling/describing Language.
I would say that we need more (depending on the final definition for 
ANIMA intent).
If ANIMA intent is/becomes an element of a Policy Based Management 
approach, then there are several elements to consider, i.e. the 
classical PBM machinery.

>
> I think No.2 is not only a language choice problem (CDDL vs YANG or other), the essential thing is how we expect the Intent to be managed.
> In my mind, there are two ways:
> a)  Just like flooding TLVs (Intent Distribution). Each node just stored the received Intent, maybe there needs to be some simple update operation according to the Intent content, but there's no restful operation through protocol between nodes. In this case, Intent is just on-wire protocol data, the same as a GRASP message or option. (Then CDDL is the proper choice as Carsten commented. Carsten: please correct me if I got your point wrong).
> b)  Sophisticated operation on Intent, e.g. partial update, deleting, query etc. Then we probably need restful operation through protocol to manipulate Intent repository stored in nodes. In this case, Intent is a model for data construction and manipulating, it's not on-wire. (Just like Netconf/Restconf+YANG style.)
I think both aspects are not excluding each other.
ANIMA intent will need to be "written", translated in various 
formats/encodings, and distributed.
If we refer to the policy continuum (which we need to define for ANIMA 
environments), a) and b) will both be specified.
I don't see how we can manage an ANIMA networks in a comprehensive way 
without considering more than just how the intent could be distributed 
(but this needs to be considered / part of the solution).

One element that might be interesting to consider is if/how the specific 
characteristics of autonomic networks/functions create a special case 
for PBM/intent based networking, e.g. by providing advanced-embedded 
reasoning/processing capabilities. So far we have not explore this 
aspect in the ANIMA WG (to my knowledge). There is a rich literature on 
PBM + autonomics.

As for RESTful "operations", this is a possible approach but we should 
probably first discuss the requirements before choosing a type of 
communication scheme (cf characteristics of autonomic networks).

>
> Apparently, GRASP does not provide any restful operation capability so far. If we want to manage Intent in this way, we either extend GRASP or even introduce another protocol.
> However, since Intent is mostly high-level policy-like content, I personally tend to simply distribute the Intent as choice a). Choice b) seems over complex for me.
I think it is not as simple as 'just" choosing a).
We need a system/functions/mechanisms to allow creating the intent 
(content), process it at the different levels, check for 
conflict/consistency, etc.

Best regards, Laurent.
>
> Thoughts?
>
> Best regards,
> Bing
>
>> -----Original Message-----
>> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
>> Sent: Wednesday, March 23, 2016 3:02 AM
>> To: Carsten Bormann; Liubing (Leo)
>> Cc: Laurent Ciavaglia; Michael Behringer; anima@ietf.org
>> Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some
>> thoughts
>>
>> Thanks Carsten. draft-veillette-core-yang-cbor-mapping looks like the
>> correct solution in the GRASP context, if someone provides the appropriate
>> run-time support code.
>>
>> IMHO using the XML or JSON formats would be a mistake.
>>
>>     Brian
>>
>> On 22/03/2016 19:44, Carsten Bormann wrote:
>>> YANG is a data modeling language for data at rest (e.g., in a
>>> management data repository); it is not designed for describing
>>> protocol data units (we recently have been using CDDL for the latter).
>>>
>>> https://tools.ietf.org/html/draft-greevenbosch-appsawg-cbor-cddl-08
>>>
>>> YANG data from (or going into) a repository can be represented on the
>>> wire in XML (native RFC 6020), JSON, and now also in CBOR:
>>>
>>> https://tools.ietf.org/html/draft-ietf-netmod-yang-json-09
>>> https://tools.ietf.org/html/draft-veillette-core-yang-cbor-mapping-00
>>>
>>> Indeed, the COMI project in the CoRE working group is about enabling
>>> RESTCONF style transactions on YANG-modeled data within constrained
>>> devices, employing CBOR representation as one of the crucial
>>> optimizations to make this possible.
>>>
>>> Grüße, Carsten
>>>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------090302050607060303000203
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Hi Bong, All,

      

      Please see below for my comments.

    

    
On 23/03/2016 04:46, EXT Liubing (Leo)
      wrote:

    

    

      
Hi all,

According to Carsten's education (thanks a lot!), and recent discussion, I realize we need to separate two things for Intent:
1. Encoding. Seems CBOR is a decent choice as discussed recently.
2. Modeling/describing Language. 

    

    I would say that we need more (depending on the final definition for
    ANIMA intent).

    If ANIMA intent is/becomes an element of a Policy Based Management
    approach, then there are several elements to consider, i.e. the
    classical PBM machinery.

    

    

      

I think No.2 is not only a language choice problem (CDDL vs YANG or other), the essential thing is how we expect the Intent to be managed.
In my mind, there are two ways:
a)  Just like flooding TLVs (Intent Distribution). Each node just stored the received Intent, maybe there needs to be some simple update operation according to the Intent content, but there's no restful operation through protocol between nodes. In this case, Intent is just on-wire protocol data, the same as a GRASP message or option. (Then CDDL is the proper choice as Carsten commented. Carsten: please correct me if I got your point wrong).
b)  Sophisticated operation on Intent, e.g. partial update, deleting, query etc. Then we probably need restful operation through protocol to manipulate Intent repository stored in nodes. In this case, Intent is a model for data construction and manipulating, it's not on-wire. (Just like Netconf/Restconf+YANG style.)

    

    I think both aspects are not excluding each other.

    ANIMA intent will need to be "written", translated in various
    formats/encodings, and distributed.

    If we refer to the policy continuum (which we need to define for
    ANIMA environments), a) and b) will both be specified.

    I don't see how we can manage an ANIMA networks in a comprehensive
    way without considering more than just how the intent could be
    distributed (but this needs to be considered / part of the
    solution).

    

    One element that might be interesting to consider is if/how the
    specific characteristics of autonomic networks/functions create a
    special case for PBM/intent based networking, e.g. by providing
    advanced-embedded reasoning/processing capabilities. So far we have
    not explore this aspect in the ANIMA WG (to my knowledge). There is
    a rich literature on PBM + autonomics.

    

    As for RESTful "operations", this is a possible approach but we
    should probably first discuss the requirements before choosing a
    type of communication scheme (cf characteristics of autonomic
    networks). 

    

    

      

Apparently, GRASP does not provide any restful operation capability so far. If we want to manage Intent in this way, we either extend GRASP or even introduce another protocol.
However, since Intent is mostly high-level policy-like content, I personally tend to simply distribute the Intent as choice a). Choice b) seems over complex for me.

    

    I think it is not as simple as 'just" choosing a).

    We need a system/functions/mechanisms to allow creating the intent
    (content), process it at the different levels, check for
    conflict/consistency, etc.

    

    Best regards, Laurent.

    

      

Thoughts?

Best regards,
Bing



      

        
-----Original Message-----
From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
Sent: Wednesday, March 23, 2016 3:02 AM
To: Carsten Bormann; Liubing (Leo)
Cc: Laurent Ciavaglia; Michael Behringer; anima@ietf.org
Subject: Re: [Anima] CBOR for Intent?-//RE: Intent yang model - some
thoughts

Thanks Carsten. draft-veillette-core-yang-cbor-mapping looks like the
correct solution in the GRASP context, if someone provides the appropriate
run-time support code.

IMHO using the XML or JSON formats would be a mistake.

   Brian

On 22/03/2016 19:44, Carsten Bormann wrote:


        

          
YANG is a data modeling language for data at rest (e.g., in a
management data repository); it is not designed for describing
protocol data units (we recently have been using CDDL for the latter).

https://tools.ietf.org/html/draft-greevenbosch-appsawg-cbor-cddl-08

YANG data from (or going into) a repository can be represented on the
wire in XML (native RFC 6020), JSON, and now also in CBOR:

https://tools.ietf.org/html/draft-ietf-netmod-yang-json-09
https://tools.ietf.org/html/draft-veillette-core-yang-cbor-mapping-00

Indeed, the COMI project in the CoRE working group is about enabling
RESTCONF style transactions on YANG-modeled data within constrained
devices, employing CBOR representation as one of the crucial
optimizations to make this possible.

Grüße, Carsten



        

      

      
_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima


    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------090302050607060303000203--


From nobody Wed Mar 23 05:07:32 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5364712D637 for ; Wed, 23 Mar 2016 05:07:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.939
X-Spam-Level: 
X-Spam-Status: No, score=-5.939 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_FONT_FACE_BAD=0.981, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id nLrqD9niw95C for ; Wed, 23 Mar 2016 05:07:22 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 48D2E12D1DA for ; Wed, 23 Mar 2016 05:06:23 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id 94790AF9F95F4; Wed, 23 Mar 2016 11:54:42 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2NBsiRG016539 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Wed, 23 Mar 2016 11:54:45 GMT
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2NBsg63010287 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 23 Mar 2016 12:54:44 +0100
Received: from [172.27.205.106] (135.239.27.38) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Wed, 23 Mar 2016 12:54:44 +0100
To: "EXT Michael Behringer (mbehring)" , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56F28403.6060401@alcatel-lucent.com>
Date: Wed, 23 Mar 2016 12:54:43 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>
Content-Type: multipart/alternative; boundary="------------020901020103090003070801"
X-Originating-IP: [135.239.27.38]
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , Sheng Jiang , "anima@ietf.org" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 12:07:28 -0000

--------------020901020103090003070801
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit

Michael,

There are some aspects I am not sure to fully understand / agree with 
(marked below with [LC]).

Best regards, Laurent.


On 23/03/2016 11:57, EXT Michael Behringer (mbehring) wrote:
>
> John,
>
> /[stripped]/
>
> The “parameter†versus “policy†discussion: My view is that we 
> shouldn’t try to define exactly **what** intent allows and doesn’t 
> allow, since we’ve heard from many sides that the boundary is not 
> clear, and it may be impossible to have a really clear, usable 
> definition. So my point was: Let’s focus on the distribution model: 
> Intent is flooded to all nodes. Let’s NOT do selective flooding, 
> unicast Intent to specific nodes, etc, because that would blur the 
> boundary even worse.
>

[LC]
Before stating that intent policies should be flooded to all nodes, can 
we first compare/evaluate the different options?
-broadcast is simple...but how important is it compared to other 
criteria...?
I see at least two issues:
1. flooding a "big" intent file (cf your current tentative example) may 
be bandwidth hungry, if one considers how frequently such an intent file 
could be distributed (ASA stop/start, device join/leave, policy 
change/update...) compared to distributing only what is required / where 
it is required.
2. security: flooding the entire file reveal information to target 
(possibly undesirable/rogue) entities that do not need it (and thus may 
mis-use it).

More fundamentally, I think it is misleading to imagine that intent 
policies will be contained in one "file" (it is possible to do so but 
impractical/cumbersome to manage). cf also the next comment.
[/LC]

> My suggestion: Intent is flooded to all nodes in a domain.
>
[LC]
It is the "resolution" of the intent clauses that determines the 
domain/set of target "nodes".
So in that order, yes the intent will be flooded to all the relevant 
nodes in the "intent policy domain". This domain may be all nodes in an 
ANIMA network, or a sub-set, only one node, or zero. This is contextual.
There will be _*many*_ intent policies for a given ANIMA network.
[/LC]

> The ASA owners define the content of what they want to push in their 
> bit of the Intent. If we insist of full flooding, we encourage the 
> right behaviour. And if someone really puts into his Intent “node A: 
> do this; node B: do the other; node C: do the third†it is awkward 
> enough for people to take a step back and think whether they’re doing 
> the right thing J
>
[LC]
I think we have a disconnect on what an intent look like and how it can 
be used.
In my "mental model", there is not one "big" intent with sub-parts for 
each ASA, the ANI, etc...
Each ASA designer (owner) will define the content / specificity of its 
ASA functionality or links to the existing elements of the ANIMA intent 
policy model, but it does not mean that we must aggregate all parts into 
a single "file".
What would be the reason to do that?

In addition, there will not only be intent policies per functions, or 
for the ANI.
There will be intent policies conveying the Service level, the Business 
level... these will not resolve/map 1-to-1 in a single ASA "policy".
[/LC]

> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
> Not sure I understand your concern. Yes, in my model there is a 
> section for the fabric, and sections for the various functions. Seems 
> a natural structure?!?
>
[LC]
It is OK to have different intent-policies for the fabric and for the 
functions, but why put them all into one file/policy?
[/LC]
>
> “Function ownerâ€: Yes, I’m suggesting to structure Intent around 
> Autonomic Functions. This is not optimal from a scientific point of 
> view, but aligns very well with the way how things are organised. 
> Probably we need to discuss that in BA. But refer to my YANG model 
> sent a few days ago.
>
[LC]
I fully agree on the focus per autonomic function, but it does not imply 
to have one single intent policy with all parts inside.
[/LC]

> We should schedule some discussion time outside the ANIMA meeting 
> window...
>
[LC]
Yes. And a white board too!
Unfortunately, I cannot attend physically this time.
Can we schedule a face-to-face ad-hoc meeting?
[/LC]


> Michael
>
> *From:*John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* 22 March 2016 21:32
> *To:* Michael Behringer (mbehring) ; John 
> Strassner 
> *Cc:* Duzongpeng ; anima@ietf.org; Laurent 
> Ciavaglia ; Jéferson Campos Nobre 
> ; Sheng Jiang 
> *Subject:* Re: [Anima] Does Autonomic network need some intervention 
> beyond intent
>
> In general, I agree with Michael. I'd like to emphasize and expand on 
> a couple of key points:
>
> First, ANIMA isn’t the only thing “influencing†or “steering†a 
> network, at least not initially. There is still traditional config, 
> plus all the existing and emerging SDN models. My view has always been 
> that ANIMA sets a “default†policy in a network, and all other means 
> are more specific and can override intent. This is described in 
> RFC7575. Main point here: ANIMA Intent isn’t the only tool in a network.
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
> Intent could perfectly well cover a high level policy such as “all 
> nodes of type x do this; type y does thatâ€. (Which I think is what you 
> mention below.) When we started with Intent, the vision was to keep it 
> very high level, more along the line “do the right thing†J
>
> 
>
> Agreed. Although to me, this means that we need to consider the types 
> of actors that are using intent. For example, most of my CCIE and JCIE 
> friends have no interest in using an intent language for pushing 
> configuration. On the other hand, when I talk to operators and 
> application developers, they would love to be able to use a high-level 
> intent and have the intent engine take care of the details for them.
>
> 
>
> It has become clear that in today’s networking, people will want to 
> push more “parameters†than “policyâ€, and I think we’ll just have to 
> acknowledge that.
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
> Personally I think it’ll pan out like this: Intent will have sections 
> per autonomic function. An Intent parser will validate overall 
> correctness (signature, etc), then segment the Intent into its bits: 
> One part will be the autonomic fabric itself, then there will be bits 
> for each autonomic function.
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
> So the Intent parser on a node will grab intent, see there is some 
> Intent for autonomic function A, will see whether this is running 
> locally, and if so, it’ll just pass that piece of Intent to that 
> function.
>
> 
>
> Agreed.
>
> 
>
> In this way, a function owner defines both the intent itself, and how 
> it is consumed by his/her function.
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
> The downside is that there is no “control†and people can do what they 
> want, in the worst case even push config, which is clearly NOT what 
> Intent is meant to do.
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
> I’d hope that people don’t start writing Intent like “node 17: here is 
> your CLI; node 23: here is your CLIâ€; node 37: here is your CLIâ€. But 
> what I describe above COULD be abused in this way. RFC7575 says this 
> is NOT Intent.
>
> 
>
> Absolutely agree!
>
> 
>
>  regards,
>
> John
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) 
> > wrote:
>
>     Hi Zongpeng,
>
>     Yes, this is a recurring question and topic. Let me give my view
>     on two levels:
>
>     First, ANIMA isn’t the only thing “influencing†or “steering†a
>     network, at least not initially. There is still traditional
>     config, plus all the existing and emerging SDN models. My view has
>     always been that ANIMA sets a “default†policy in a network, and
>     all other means are more specific and can override intent. This is
>     described in RFC7575. Main point here: ANIMA Intent isn’t the only
>     tool in a network.
>
>     Intent could perfectly well cover a high level policy such as “all
>     nodes of type x do this; type y does thatâ€. (Which I think is what
>     you mention below.) When we started with Intent, the vision was to
>     keep it very high level, more along the line “do the right thingâ€
>     J  It has become clear that in today’s networking, people will
>     want to push more “parameters†than “policyâ€, and I think we’ll
>     just have to acknowledge that.
>
>     Personally I think it’ll pan out like this: Intent will have
>     sections per autonomic function. An Intent parser will validate
>     overall correctness (signature, etc), then segment the Intent into
>     its bits: One part will be the autonomic fabric itself, then there
>     will be bits for each autonomic function.
>
>     So the Intent parser on a node will grab intent, see there is some
>     Intent for autonomic function A, will see whether this is running
>     locally, and if so, it’ll just pass that piece of Intent to that
>     function.
>
>     In this way, a function owner defines both the intent itself, and
>     how it is consumed by his/her function.
>
>     The downside is that there is no “control†and people can do what
>     they want, in the worst case even push config, which is clearly
>     NOT what Intent is meant to do. I’d hope that people don’t start
>     writing Intent like “node 17: here is your CLI; node 23: here is
>     your CLIâ€; node 37: here is your CLIâ€. But what I describe above
>     COULD be abused in this way. RFC7575 says this is NOT Intent.
>
>     In other words: Let’s not be too prescriptive, and hope that folks
>     will do more or less reasonable things.
>
>     I’ve been meaning to comment on the Intent drafts and possibly
>     contribute, I’m just struggling with priorities at the moment (as
>     you can see by my relative silence over the past weeks).
>
>     Michael
>
>     *From:*Anima [mailto:anima-bounces@ietf.org
>     ] *On Behalf Of *Duzongpeng
>     *Sent:* 16 March 2016 02:38
>     *To:* anima@ietf.org 
>     *Cc:* Laurent Ciavaglia      >; Jéferson Campos Nobre
>     >; Sheng Jiang
>     >
>     *Subject:* [Anima] Does Autonomic network need some intervention
>     beyond intent
>
>     Hello, everyone
>
>              I understand that Autonomic network should be able to
>     work with as less configuration as possible. In the ideal case, an
>     Autonomic network should be able to work well while only intent is
>     needed from human operators.
>
>              However, I wonder whether Autonomic network may need some
>     intervention from human operators through some low-level policies
>     beyond intents.
>
>              For example, in the
>     https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,
>     it is suggested that the prefix lengths for the CSG, ASG, RSG
>     (different roles in IP RAN) can be assigned as an "intent". These
>     configuration parameters need to be distributed in the autonomic
>     domain to influence the detail configurations on each autonomic node.
>
>              As intent is described as an abstract, declarative,
>     high-level policy used to operate an autonomic domain, such as an
>     enterprise network. Perhaps we can rename the “intent†in the
>     “prefix-management ID†to some other words.
>
>     Is “ASA parameters" ok here?
>
>     I think some characteristics for this kind of low-level policies are
>
>     1.Network level parameters, need to be distributed by GRASP and ACP
>
>     2.Relatively static compared to intent, perhaps only needed to be
>     configured once
>
>     3.Unavoidable in current network environment, mostly for
>     establishing the network infrastructure
>
>     4.Rarely need coordinations with others parameters, usually is
>     closed related to one specific autonomic function
>
>     Welcome for comment. Your suggestions will be appreciated.
>
>     Best Regards
>
>     Zongpeng Du
>
>
>     _______________________________________________
>     Anima mailing list
>     Anima@ietf.org 
>     https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> -- 
>
> regards,
>
> John
>

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------020901020103090003070801
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Michael,

      

      There are some aspects I am not sure to fully understand / agree
      with (marked below with [LC]).

      

      Best regards, Laurent.

      

    

    
On 23/03/2016 11:57, EXT Michael
      Behringer (mbehring) wrote:

    

    

      
      
      
      

        
John,
              

        [stripped]

        

        
The
              “parameter†versus “policy†discussion: My view is that we
              shouldn’t try to define exactly *what* intent
              allows and doesn’t allow, since we’ve heard from many
              sides that the boundary is not clear, and it may be
              impossible to have a really clear, usable definition. So
              my point was: Let’s focus on the distribution model:
              Intent is flooded to all nodes. Let’s NOT do selective
              flooding, unicast Intent to specific nodes, etc, because
              that would blur the boundary even worse.
            

      

    

    

    [LC]

    Before stating that intent policies should be flooded to all nodes,
    can we first compare/evaluate the different options?

    -broadcast is simple...but how important is it compared to other
    criteria...?

    I see at least two issues:

    1. flooding a "big" intent file (cf your current tentative example)
    may be bandwidth hungry, if one considers how frequently such an
    intent file could be distributed (ASA stop/start, device join/leave,
    policy change/update...) compared to distributing only what is
    required / where it is required.

    2. security: flooding the entire file reveal information to target
    (possibly undesirable/rogue) entities that do not need it (and thus
    may mis-use it).

    

    More fundamentally, I think it is misleading to imagine that intent
    policies will be contained in one "file" (it is possible to do so
    but impractical/cumbersome to manage). cf also the next comment.

    [/LC]

    

    

      

        

        
My
              suggestion: Intent is flooded to all nodes in a domain.

      

    

    [LC]

    It is the "resolution" of the intent clauses that determines the
    domain/set of target "nodes".

    So in that order, yes the intent will be flooded to all the relevant
    nodes in the "intent policy domain". This domain may be all nodes in
    an ANIMA network, or a sub-set, only one node, or zero. This is
    contextual.

    There will be many intent policies for a given ANIMA
    network.

    [/LC]

    

    

      

        

              The ASA owners define the content of what they want to
              push in their bit of the Intent. If we insist of full
              flooding, we encourage the right behaviour. And if someone
              really puts into his Intent “node A: do this; node B: do
              the other; node C: do the third†it is awkward enough for
              people to take a step back and think whether they’re doing
              the right thing 
          J

      

    

    [LC]

    I think we have a disconnect on what an intent look like and how it
    can be used.

    In my "mental model", there is not one "big" intent with sub-parts
    for each ASA, the ANI, etc... 

    Each ASA designer (owner) will define the content / specificity of
    its ASA functionality or links to the existing elements of the ANIMA
    intent policy model, but it does not mean that we must aggregate all
    parts into a single "file".

    What would be the reason to do that?

    

    In addition, there will not only be intent policies per functions,
    or for the ANI.

    There will be intent policies conveying the Service level, the
    Business level... these will not resolve/map 1-to-1 in a single ASA
    "policy".

    [/LC]

    

    

      

        
<jcs>

        
Do
              you mean that the intent writer will **know the
              difference**

        
between
              the fabric and each autonomic function? I hope not,

        
but
              perhaps I'm misunderstanding you...

        
</jcs>

        
 

        
Not
              sure I understand your concern. Yes, in my model there is
              a section for the fabric, and sections for the various
              functions. Seems a natural structure?!? 

      

    

    [LC]

    It is OK to have different intent-policies for the fabric and for
    the functions, but why put them all into one file/policy?

    [/LC]

    

      

        

        
 

        
“Function
              ownerâ€: Yes, I’m suggesting to structure Intent around
              Autonomic Functions. This is not optimal from a scientific
              point of view, but aligns very well with the way how
              things are organised. Probably we need to discuss that in
              BA. But refer to my YANG model sent a few days ago.
            

      

    

    [LC]

    I fully agree on the focus per autonomic function, but it does not
    imply to have one single intent policy with all parts inside.

    [/LC]

    

    

      

        

        
 

        
We
              should schedule some discussion time outside the ANIMA
              meeting window...

        
 

      

    

    [LC]

    Yes. And a white board too!

    Unfortunately, I cannot attend physically this time.

    Can we schedule a face-to-face ad-hoc meeting?

    [/LC]

    

    

    

      

        
Michael

        
 

        
 

        

          

            

              
From: John Strassner
                    [mailto:strazpdj@gmail.com] 

                    Sent:
                    22 March 2016 21:32

                    To:
                    Michael Behringer (mbehring)
                    <mbehring@cisco.com>; John Strassner
                    <strazpdj@gmail.com>

                    Cc:
                    Duzongpeng <duzongpeng@huawei.com>;
                    anima@ietf.org; Laurent Ciavaglia
                    <laurent.ciavaglia@nokia.com>; Jéferson Campos
                    Nobre <jcnobre@inf.ufrgs.br>; Sheng Jiang
                    <jiangsheng@huawei.com>

                    Subject:
                    Re: [Anima] Does Autonomic network need some
                    intervention beyond intent

            

          

          
 

          

            

              
In general, I agree with
                    Michael. I'd like to emphasize and expand on a
                    couple of key points:

            

            

              
 

            

            

              
 

            

            

              
First,
                    ANIMA isn’t the only thing “influencing†or
                    “steering†a network, at least not initially. There
                    is still traditional config, plus all the existing
                    and emerging SDN models. My view has always been
                    that ANIMA sets a “default†policy in a network, and
                    all other means are more specific and can override
                    intent. This is described in RFC7575. Main point
                    here: ANIMA Intent isn’t the only tool in a network.
                  

              
 

            

            

              
<jcs>

            

            

              
Absolutely agree. Note that
                    even in the Congress case I mentioned

            

            

              
earlier, it is NOT pushing
                    config, and it is NOT viewed as the

            

            

              
"only" policy

            

            

              

                
</jcs>

              

            

            

              
 

            

            

              
 

            

            

              
Intent
                    could perfectly well cover a high level policy such
                    as “all nodes of type x do this; type y does thatâ€.
                    (Which I think is what you mention below.) When we
                    started with Intent, the vision was to keep it very
                    high level, more along the line “do the right thingâ€
                  J 

              
 

            

            

              

                
<jcs>

              

              

                
Agreed.
                      Although to me, this means that we need to
                      consider the types of actors that are using
                      intent. For example, most of my CCIE and JCIE
                      friends have no interest in using an intent
                      language for pushing configuration. On the other
                      hand, when I talk to operators and application
                      developers, they would love to be able to use a
                      high-level intent and have the intent engine take
                      care of the details for them.

              

              

                
</jcs>

              

              

                
 

              

            

            

              

                
It has
                      become clear that in today’s networking, people
                      will want to push more “parameters†than “policyâ€,
                      and I think we’ll just have to acknowledge that.

              

              

                
 

              

              

                
<jcs>

              

              

                
Sorry Michael,
                      I don't follow here. What do you mean

              

              

                
by
                      "parameters"?

              

              

                
</jcs>

              

              

                
 

              

              

                

                  
 

                

                

                  
Personally
                        I think it’ll pan out like this: Intent will
                        have sections per autonomic function. An Intent
                        parser will validate overall correctness
                        (signature, etc), then segment the Intent into
                        its bits: One part will be the autonomic fabric
                        itself, then there will be bits for each
                        autonomic function.
                      

                  
 

                

                

                  
<jcs>

                

                

                  
Do you mean
                        that the intent writer will **know the
                        difference**

                

                

                  
between the
                        fabric and each autonomic function? I hope not,

                

                

                  
but perhaps
                        I'm misunderstanding you...

                

                

                  
</jcs>

                

                

                  
 

                

                

                  
 

                

                

                  
So
                        the Intent parser on a node will grab intent,
                        see there is some Intent for autonomic function
                        A, will see whether this is running locally, and
                        if so, it’ll just pass that piece of Intent to
                        that function.
                      

                

                

                  

                    
 

                  

                  

                    
<jcs>

                  

                  

                    
Agreed.

                  

                  

                    
</jcs>

                  

                  

                    
 

                  

                  

                    

                      
 

                    

                    

                      
In
                            this way, a function owner defines both the
                            intent itself, and how it is consumed by
                            his/her function. 

                      
 

                    

                    

                      
<jcs>

                    

                    

                      
I'm
                            not sure what you mean by "function owner".
                            (Sorry to be

                    

                    

                      
picky,
                            I'm likely under-caffeinated.)

                    

                    

                      
 

                    

                    

                      
The
                            above seemed like an autonomous decision,
                            made by the parser.

                    

                    

                      
I
                            don't think that the intent writer should be
                            cognizant of how many or

                    

                    

                      
what
                            types of autonomic functions are present.

                    

                    

                      
</jcs>

                    

                    

                      
 

                    

                    

                      
 

                    

                    

                      

                        
The
                              downside is that there is no “control†and
                              people can do what they want, in the worst
                              case even push config, which is clearly
                              NOT what Intent is meant to do.
                            

                        
 

                      

                      

                        
<jcs>

                      

                      

                        
And
                              hence, the "HAL 2000" scenario. :-)

                      

                      

                        
</jcs>

                      

                      

                        
 

                      

                      

                        

                          
 

                        

                        

                          
I’d
                                hope that people don’t start writing
                                Intent like “node 17: here is your CLI;
                                node 23: here is your CLIâ€; node 37:
                                here is your CLIâ€. But what I describe
                                above COULD be abused in this way.
                                RFC7575 says this is NOT Intent.
                              

                        

                        

                          
 

                        

                        

                          
<jcs>

                        

                        

                          
Absolutely
                                agree!

                        

                        

                          
</jcs>

                        

                        

                          
 

                        

                      

                    

                  

                

              

            

            

              
 regards,

              
John

              
 

              
 

            

          

          

            
 

            

              
On Wed, Mar 16, 2016 at
                    1:16 AM, Michael Behringer (mbehring) <mbehring@cisco.com>
                    wrote:

              

                

                  

                    
Hi
                          Zongpeng,
                        

                    
 

                    
Yes,
                          this is a recurring question and topic. Let me
                          give my view on two levels: 

                    
 

                    
First,
                          ANIMA isn’t the only thing “influencing†or
                          “steering†a network, at least not initially.
                          There is still traditional config, plus all
                          the existing and emerging SDN models. My view
                          has always been that ANIMA sets a “defaultâ€
                          policy in a network, and all other means are
                          more specific and can override intent. This is
                          described in RFC7575. Main point here: ANIMA
                          Intent isn’t the only tool in a network.
                        

                    
 

                    
Intent
                          could perfectly well cover a high level policy
                          such as “all nodes of type x do this; type y
                          does thatâ€. (Which I think is what you mention
                          below.) When we started with Intent, the
                          vision was to keep it very high level, more
                          along the line “do the right thingâ€
                        J  It
                          has become clear that in today’s networking,
                          people will want to push more “parametersâ€
                          than “policyâ€, and I think we’ll just have to
                          acknowledge that.
                        

                    
 

                    
Personally
                          I think it’ll pan out like this: Intent will
                          have sections per autonomic function. An
                          Intent parser will validate overall
                          correctness (signature, etc), then segment the
                          Intent into its bits: One part will be the
                          autonomic fabric itself, then there will be
                          bits for each autonomic function.
                        

                    
 

                    
So
                          the Intent parser on a node will grab intent,
                          see there is some Intent for autonomic
                          function A, will see whether this is running
                          locally, and if so, it’ll just pass that piece
                          of Intent to that function.
                        

                    
 

                    
In
                          this way, a function owner defines both the
                          intent itself, and how it is consumed by
                          his/her function. 

                    
 

                    
The
                          downside is that there is no “control†and
                          people can do what they want, in the worst
                          case even push config, which is clearly NOT
                          what Intent is meant to do. I’d hope that
                          people don’t start writing Intent like “node
                          17: here is your CLI; node 23: here is your
                          CLIâ€; node 37: here is your CLIâ€. But what I
                          describe above COULD be abused in this way.
                          RFC7575 says this is NOT Intent. 
                      

                    
 

                    
In
                          other words: Let’s not be too prescriptive,
                          and hope that folks will do more or less
                          reasonable things. 

                    
 

                    
I’ve
                          been meaning to comment on the Intent drafts
                          and possibly contribute, I’m just struggling
                          with priorities at the moment (as you can see
                          by my relative silence over the past weeks).
                        

                    
 

                    
Michael

                    
 

                    
 

                    
 

                    

                      

                        

                          
From: Anima [mailto:anima-bounces@ietf.org]
                                On
                                    Behalf Of Duzongpeng

                                Sent:
                                16 March 2016 02:38

                                To:
                              anima@ietf.org

                                Cc:
                                Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; Jéferson Campos Nobre
                                <jcnobre@inf.ufrgs.br>; Sheng Jiang <jiangsheng@huawei.com>

                                Subject:
                                [Anima] Does Autonomic network need some
                                intervention beyond intent

                        

                      

                      
 

                      
Hello, everyone

                      
 

                      
         I understand that
                            Autonomic network should be able to work
                            with as less configuration as possible. In
                            the ideal case, an Autonomic network should
                            be able to work well while only intent is
                            needed from human operators.

                      
 

                      
         However, I wonder
                            whether Autonomic network may need some
                            intervention from human operators through
                            some low-level policies beyond intents.

                      
 

                      
         For example, in the
                          https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is suggested that the
                            prefix lengths for the CSG, ASG, RSG
                            (different roles in IP RAN) can be assigned
                            as an "intent". These configuration
                            parameters need to be distributed in the
                            autonomic domain to influence the detail
                            configurations on each autonomic node.

                      
 

                      
         As intent is described
                            as an abstract, declarative, high-level
                            policy used to operate an autonomic domain,
                            such as an enterprise network. Perhaps we
                            can rename the “intent†in the
                            “prefix-management ID†to some other words.

                      
 

                      
Is “ASA parameters" ok here?
                          

                      
 

                      
I think some characteristics
                            for this kind of low-level policies are

                      
1.      
                          Network level
                          parameters, need to be distributed by GRASP
                          and ACP

                      
2.      
                          Relatively
                          static compared to intent, perhaps only needed
                          to be configured once

                      
3.      
                          Unavoidable
                          in current network environment, mostly for
                          establishing the network infrastructure

                      
4.      
                          Rarely need
                          coordinations with others parameters, usually
                          is closed related to one specific autonomic
                          function

                      
        
                          

                      
Welcome for comment. Your
                            suggestions will be appreciated.

                      
 

                      
 

                      
Best Regards

                      
Zongpeng Du

                    

                  

                

                


                      _______________________________________________

                      Anima mailing list

                      Anima@ietf.org

                      https://www.ietf.org/mailman/listinfo/anima

              

            

            


                  

                  

                  -- 

            

              

                
regards,

              

              

                
John

              

            

          

        

      

    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------020901020103090003070801--


From nobody Wed Mar 23 05:33:13 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C1B2D12DCC4 for ; Wed, 23 Mar 2016 05:33:09 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7UizfZS73yFD for ; Wed, 23 Mar 2016 05:33:02 -0700 (PDT)
Received: from rcdn-iport-7.cisco.com (rcdn-iport-7.cisco.com [173.37.86.78]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0C87312DD11 for ; Wed, 23 Mar 2016 05:23:02 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=9941; q=dns/txt; s=iport; t=1458735781; x=1459945381; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=fp/bLcfPWI/+ocL0xiQ33TYAghz/INYDGnv+oZjV3KI=; b=fZpR0ckM1mUxxV64fYMDMO1IXOc5rdWUHcVrjZAx0INoTKH8XgJBHsj3 xDy0+RED276YsK4u1WfNW9lkn8M35wz1dHbclphFTfvS6Wv/B8cMaLBur ovqEc2KMbytSDBLPp8x4AkQNAYeYDfFIkehHtmrElpIPT2pyAMN4oIsGM g=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AlAgDJevJW/5JdJa1egzNTegavDoZbh?= =?us-ascii?q?HMBDYFwFwqFbAKBPzgUAQEBAQEBAWQnhEEBAQEDAQEBATcxAxAHBAIBCBEEAQE?= =?us-ascii?q?BHgkHIQYLFAkIAQEEARIIiAoDCggOuxsNhHEBAQEBAQEBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QERBIYehESCPoFZhXsFh2KFWYVJhCUxAYcdhHGBboFthEyIWIcyh1QBHgEBQhy?= =?us-ascii?q?BYoFnagGIUDx+AQEB?=
X-IronPort-AV: E=Sophos;i="5.24,382,1454976000"; d="scan'208";a="84004822"
Received: from rcdn-core-10.cisco.com ([173.37.93.146]) by rcdn-iport-7.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 23 Mar 2016 11:19:56 +0000
Received: from XCH-RCD-010.cisco.com (xch-rcd-010.cisco.com [173.37.102.20]) by rcdn-core-10.cisco.com (8.14.5/8.14.5) with ESMTP id u2NBJuOG023469 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Wed, 23 Mar 2016 11:19:56 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-010.cisco.com (173.37.102.20) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Wed, 23 Mar 2016 06:19:56 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Wed, 23 Mar 2016 06:19:56 -0500
From: "Michael Behringer (mbehring)" 
To: "Joel M. Halpern" , Brian E Carpenter , "anima@ietf.org" 
Thread-Topic: [Anima] Intent yang model - some thoughts
Thread-Index: AdGA84Aiy7EwHtVtTUWSCghLdxIQCwCFPbeAAADOYgAAFVahIA==
Date: Wed, 23 Mar 2016 11:19:55 +0000
Message-ID: <612d4aefef0d46378a6617ae6d803219@XCH-RCD-006.cisco.com>
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <56EEF9A7.90709@gmail.com> <56EEFF10.1050808@joelhalpern.com>
In-Reply-To: <56EEFF10.1050808@joelhalpern.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.61.106.202]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Subject: Re: [Anima] Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 12:33:10 -0000

> -----Original Message-----
> From: Joel M. Halpern [mailto:jmh@joelhalpern.com]
> Sent: 20 March 2016 20:51
> To: Brian E Carpenter ; Michael Behringer
> (mbehring) ; anima@ietf.org
> Subject: Re: [Anima] Intent yang model - some thoughts
>=20
> There are two aspects of this model that have me very confused.
>=20
> First, in most policy work I have seen one specifies a set of policies, a=
nd then
> specifies what they apply to.  This model seems to say "Here are a bunch =
of
> domains", and then for each domain it lists the policies.
>   Structuring it this way means that reuse is harder.

That is a good discussion to have. I don't have a good answer at the moment=
.=20

> Somewhat related to that, I find the given metadata confusing:
> 1) There apparently is some notion that a domain has sub-domains, but tha=
t
> there may be things in the higher domain?  And that you may or may not
> want the policy to recurse downwards.  But if you want it to recurse
> downwards, you want it to recurse forever?  (This looks like an optimizat=
ion
> for sharing policies, but only across all sub-domains of a
> domain.)

We need to find the right way to do this. But some form of "inheritance" of=
 intent from parent to child domain seem a good idea. There are various way=
s to do that, we could turn the definition around and just list the (sub-)d=
omains, as you suggested; we could recurse indefinitely until another domai=
n definition explicitly stops it, we could limit the number of recursions, =
to be seen.=20

> 2) Fail-open, maintenance-window, and other such items seem to be treated
> as being applicable and relevant to all policies in the domain.
>   That is not at all obvious.

And it may not be in the right spot; again, this is really a brainstorming =
exercise; the main point is that there exist parameters (not a good word; i=
ntent-bits?)  on each of those levels: One for the intent file itself, one =
per domain, etc.=20

> 2a) For a given ASA / function, fail-open may be meaningful, and for an
> individual policy the goal may be fail-open or fail-closed.  Or we may no=
t even
> know what "fail" means.

Here, I'm referring to the AN infrastructure. There will be cases where sec=
urity checks fail, but communication is still possible; for those, we need =
to allow Intent "fail-open" or "fail-close".=20

> 2b) Even for policies where a maintenance window makes sense, it is unlik=
ely
> that all policies will have the same maintenance window.
> Usually, since maintenance can be unpredictable in practice, policies hav=
e on-
> off switches individually.

The proposed structure is not cast in stone, needs discussion ;-)

> Second, am confused by the modeling for "functions".  I will assume that
> these functions are ASAs.  So they are network-wide targets of policy.
> But then I am confused as to what is being modeled here to help the
> function modeling.  I see a boolean for whether the function applies to a=
ll
> devices.  It seems unlikely that the choices are all / one, so a boolean =
seems
> very wrong.  I think the idea is taht you can specify "all" or a single r=
ole.=20

indeed.

> I
> would think a list of roles would actually be more useful.
> And then I see dynamic-role and tag.  I have no idea what those are
> supposed to mean in an ASA-independent fashion.

Sorry, not explained yet, and the details are for future discussion.=20
=20
> I do think that there are some ASA-policy attributes that are useful and
> common.  Things like whether the policy is active.
>=20
> As a minor confusion, on the scale of things, I presume that function_int=
ent
> is actually the placeholder for the function parameters?

yes.=20

Sorry for causing all the confusion, again, let's focus on the overall stru=
cture so far, not on the detailed parameters. The parameters shown are exam=
ples only.=20

Michael
=20
> Yours,
> Joel
>=20
>=20
> On 3/20/16 3:27 PM, Brian E Carpenter wrote:
> > Hi Michael,
> >
> > At a glance, from someone who is a yang beginner, this looks
> > plausible, although I think we need some analysis of the purpose of
> > many of the items. Just to pick a random example, consider "maintenance=
-
> window".
> >   - is this fixed or does it change?
> >   - if it changes, how long in advance does the new intent get sent?
> >   - what are the semantics (must all ASAs stop operation, etc.)?
> >   - what happens if maintenance ends early?
> >   - what happens if maintenance ends late?
> > In other words these items are not self-defining; each one needs work.
> >
> > A beginner's question: how is yang serialised on the wire?
> >
> > And another: can we send partial yang, i.e. only send the bits that
> changed?
> >
> > Regards
> >     Brian
> >
> >
> > On 18/03/2016 22:52, Michael Behringer (mbehring) wrote:
> >> WRT the current discussion on what should be in Intent and what
> shouldn't, as mentioned before, I think we should leave this to the owner=
 of
> an autonomic function to decide. This risks "abuse" in a way that wasn't
> foreseen, but it gives maximum agility, which is what we need to get ANIM=
A
> off the ground, IMO. If we are in the nice position to find that it gets =
adopted
> a lot, and if there are things that need fixing then, we can work on vers=
ion 2.
> >>
> >> Several months ago I started with a yang model for Intent, but never
> finished. Maybe this helps with this discussion. Note I'm not at all a ya=
ng
> expert, so please look more at the general structure for now. Not sure if=
 it
> helps the discussion...
> >>
> >> +--rw intent
> >>     +--rw meta-data
> >>     |  +--rw intent-version  uint32   // the version of the intent spe=
cification
> >>     |  +--rw timestamp 		uint32   // latest time stamp wins
> >>     |  +--rw administrator 	string   // who created that intent
> >>     |  +--rw device?   		string   // on which node was it created
> >>     |  +--rw execution-time	uint32   // plus other parameters on when =
to
> execute
> >>     |  +--rw expiration    	uint32   // when does this intent cease to=
 be
> valid
> >>     |  +--rw feedback                 // how to let the operator know =
about
> success/failure
> >>     |  +--rw signature		??       // intent must be signed
> >>     |
> >>     +--rw scope
> >>        +--rw domains
> >>           +--rw domain* [name]
> >>              +--rw name		string    // is there a fqdn type defined?
> >>              +--rw meta-data           // general about the domain
> >>              |  +--rw include-subdomains	boolean  // should sub-domain=
s
> follow this policy?
> >>              |  +--rw fail-open  boolean  // fail-open or fail-closed =
policy
> >>              |  +--rw maintenance-window  // when can critical operati=
ons take
> place
> >>              |  +--rw ...                 // other general policies on=
 domain level
> >> |        ...
> >>              |
> >>              +--rw ani-intent
> >>              +-- (contains all policies which the ANI should follow)
> >>              |  ex: ACP policies, coordination between AN functions, s=
ecurity
> policies, etc
> >>              |  for ani-intent, scope is always "all" (ie all nodes mu=
st process)
> >>              |  ...
> >> ...
> >> All of the above MUST be standardised, because it refers to the AN
> infrastructure itself.
> >> Of the below we need to standardise the structure, but not content
> >> for each function Conceptually, we define Intent for a domain.
> >> ...
> >>              |
> >>              +--rw intent-for-autonomic-functions
> >>                 +--rw function_1 [name]
> >>                 |  +--rw scope                    // which devices in =
the domain need to
> listen.
> >>                 |  |  +--rw all    boolean        // all devices
> >>                 |  |  +--rw role   string         // as defined in cer=
tificate
> >>                 |  |  +--rw dynamic_role  string  // as determined dyn=
amically
> >>                 |  |  +--rw tag    string         // can be used for l=
ocation
> >>                 |  +-- function_intent            // to be defined by =
function owner
> >>                 |
> >>                 +--rw function_2 [name]
> >>                 |  +--rw scope                    // which devices in =
the domain need to
> listen.
> >>                 |  |  +--rw all    boolean        // all devices
> >>                 |  |  +--rw role   string         // as defined in cer=
tificate
> >>                 |  |  +--rw dynamic_role  string  // as determined dyn=
amically
> >>                 |  |  +--rw tag    string         // can be used for l=
ocation
> >>                 |  +-- function_intent            // to be defined by =
function owner
> >>                 |
> >> ...
> >>
> >> If a node receives Intent, it checks:
> >> 1) meta data: Mostly for forwarding and processing of Intent
> >> 2) domain info: domain specific policy
> >> 3) ANI Intent: Intent for the ANI itself; every node must be able to
> >> process that
> >> 4) For each of the "function" blocks:
> >>     - should I run this function?
> >>     - do I run this function?
> >>     - if so, forward the rest of that block to the corresponding handl=
er.
> >>
> >> I think we want a clean architectural separation between the ANI and t=
he
> functions. This way every function can do what it wants.
> >>
> >> Thoughts?
> >> Michael
> >>
> >> _______________________________________________
> >> Anima mailing list
> >> Anima@ietf.org
> >> https://www.ietf.org/mailman/listinfo/anima
> >>
> >
> > _______________________________________________
> > Anima mailing list
> > Anima@ietf.org
> > https://www.ietf.org/mailman/listinfo/anima
> >


From nobody Wed Mar 23 05:51:29 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 606F812D0C2 for ; Wed, 23 Mar 2016 05:51:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.53
X-Spam-Level: 
X-Spam-Status: No, score=-14.53 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mcgHLY-NAc-t for ; Wed, 23 Mar 2016 05:51:21 -0700 (PDT)
Received: from rcdn-iport-9.cisco.com (rcdn-iport-9.cisco.com [173.37.86.80]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5845912D16C for ; Wed, 23 Mar 2016 05:41:09 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=78024; q=dns/txt; s=iport; t=1458736869; x=1459946469; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=uyqHozAci5BwSTNxf4ZY8U+VTjX9ltiKBcnm/9DcgAQ=; b=YTHnZ3w9Q2kXyZaAzFklR+qLjR9vCp6zNUa7EJ1VNkmn7lNHRSzpCmM7 2aXQ3M/pantnyXyIglFghHlvInsuSO3a3I6vA1ZMeh52QHI2FGcKDsUGt mIoTGpRuWdAvZj4WjrVKuP2ZH9Jfyoc28gfATQ2XskHECTIIFAX4hMgT/ Q=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AKAgCedfJW/49dJa1egmdMU3oGrw6LT?= =?us-ascii?q?gENgW0DFwEJhWwCHIEjOBQBAQEBAQEBZCeEQQEBAQQBAQEgCkELEAIBBgIOAwQ?= =?us-ascii?q?BASEBBgMCAgIfBgsUCQgBAQQOBQiICgMSDpIGnReMBA2EcQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEBAQEBAQEBAREEhh6ERII+gisJgkqCVgWTBIQlMQGFcIYegW6PEYcyh1QBHgE?= =?us-ascii?q?BQoNlaohSJRZ+AQEB?=
X-IronPort-AV: E=Sophos; i="5.24,382,1454976000"; d="scan'208,217"; a="83683319"
Received: from rcdn-core-7.cisco.com ([173.37.93.143]) by rcdn-iport-9.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 23 Mar 2016 10:57:03 +0000
Received: from XCH-RCD-006.cisco.com (xch-rcd-006.cisco.com [173.37.102.16]) by rcdn-core-7.cisco.com (8.14.5/8.14.5) with ESMTP id u2NAv3I3032129 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Wed, 23 Mar 2016 10:57:03 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-006.cisco.com (173.37.102.16) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Wed, 23 Mar 2016 05:57:02 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Wed, 23 Mar 2016 05:57:02 -0500
From: "Michael Behringer (mbehring)" 
To: John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAVKENoAADnV0sA==
Date: Wed, 23 Mar 2016 10:57:02 +0000
Message-ID: <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> 
In-Reply-To: 
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.61.106.202]
Content-Type: multipart/alternative; boundary="_000_8368235337ea4adfab5c428d57c3bfb2XCHRCD006ciscocom_"
MIME-Version: 1.0
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , Sheng Jiang , "anima@ietf.org" , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 12:51:27 -0000

--_000_8368235337ea4adfab5c428d57c3bfb2XCHRCD006ciscocom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64

Sm9obiwNCg0KSSB0aGluayB3ZeKAmXJlIG9uIHRoZSBzYW1lIHBhZ2UuIENvdXBsZSBvZiBwb2lu
dHMsIHRvcCBwb3N0aW5nIGhlcmUsIG90aGVyd2lzZSBpdCBnZXRzIHRvbyBtZXNzeS4NCg0KV2hp
bGUgKkNJRXMgZG9u4oCZdCBsaWtlIG1hZ2ljLCB0aGV5IGRvIHVuZGVyc3RhbmQgdGhlIGNvbmNl
cHQgb2YgYSBkZWZhdWx0IGJlaGF2aW91ciB3aGljaCB0aGV5IGNhbiBpbmZsdWVuY2UgaWYgbmVl
ZGVkLiBBbnkgcmVhc29uYWJsZSBuZXR3b3JrIG1hbmFnZW1lbnQgYXBwcm9hY2ggdHJpZXMgdG8g
ZGVmaW5lIOKAnGRlZmF1bHTigJ0gYmVoYXZpb3VyIGFuZCDigJxleGNlcHRpb25z4oCdLiBUaGUg
QU4gc3RvcnkgZmlsbHMgdGhlIOKAnGRlZmF1bHTigJ0gcGFydCB2ZXJ5IG5pY2VseSwgYW5kICpD
SUVzIGdldCB0aGF0Lg0KDQpUaGUg4oCccGFyYW1ldGVy4oCdIHZlcnN1cyDigJxwb2xpY3nigJ0g
ZGlzY3Vzc2lvbjogTXkgdmlldyBpcyB0aGF0IHdlIHNob3VsZG7igJl0IHRyeSB0byBkZWZpbmUg
ZXhhY3RseSAqd2hhdCogaW50ZW50IGFsbG93cyBhbmQgZG9lc27igJl0IGFsbG93LCBzaW5jZSB3
ZeKAmXZlIGhlYXJkIGZyb20gbWFueSBzaWRlcyB0aGF0IHRoZSBib3VuZGFyeSBpcyBub3QgY2xl
YXIsIGFuZCBpdCBtYXkgYmUgaW1wb3NzaWJsZSB0byBoYXZlIGEgcmVhbGx5IGNsZWFyLCB1c2Fi
bGUgZGVmaW5pdGlvbi4gU28gbXkgcG9pbnQgd2FzOiBMZXTigJlzIGZvY3VzIG9uIHRoZSBkaXN0
cmlidXRpb24gbW9kZWw6IEludGVudCBpcyBmbG9vZGVkIHRvIGFsbCBub2Rlcy4gTGV04oCZcyBO
T1QgZG8gc2VsZWN0aXZlIGZsb29kaW5nLCB1bmljYXN0IEludGVudCB0byBzcGVjaWZpYyBub2Rl
cywgZXRjLCBiZWNhdXNlIHRoYXQgd291bGQgYmx1ciB0aGUgYm91bmRhcnkgZXZlbiB3b3JzZS4N
Cg0KTXkgc3VnZ2VzdGlvbjogSW50ZW50IGlzIGZsb29kZWQgdG8gYWxsIG5vZGVzIGluIGEgZG9t
YWluLiBUaGUgQVNBIG93bmVycyBkZWZpbmUgdGhlIGNvbnRlbnQgb2Ygd2hhdCB0aGV5IHdhbnQg
dG8gcHVzaCBpbiB0aGVpciBiaXQgb2YgdGhlIEludGVudC4gSWYgd2UgaW5zaXN0IG9mIGZ1bGwg
Zmxvb2RpbmcsIHdlIGVuY291cmFnZSB0aGUgcmlnaHQgYmVoYXZpb3VyLiBBbmQgaWYgc29tZW9u
ZSByZWFsbHkgcHV0cyBpbnRvIGhpcyBJbnRlbnQg4oCcbm9kZSBBOiBkbyB0aGlzOyBub2RlIEI6
IGRvIHRoZSBvdGhlcjsgbm9kZSBDOiBkbyB0aGUgdGhpcmTigJ0gaXQgaXMgYXdrd2FyZCBlbm91
Z2ggZm9yIHBlb3BsZSB0byB0YWtlIGEgc3RlcCBiYWNrIGFuZCB0aGluayB3aGV0aGVyIHRoZXni
gJlyZSBkb2luZyB0aGUgcmlnaHQgdGhpbmcg4pi6DQoNCjxqY3M+DQpEbyB5b3UgbWVhbiB0aGF0
IHRoZSBpbnRlbnQgd3JpdGVyIHdpbGwgKiprbm93IHRoZSBkaWZmZXJlbmNlKioNCmJldHdlZW4g
dGhlIGZhYnJpYyBhbmQgZWFjaCBhdXRvbm9taWMgZnVuY3Rpb24/IEkgaG9wZSBub3QsDQpidXQg
cGVyaGFwcyBJJ20gbWlzdW5kZXJzdGFuZGluZyB5b3UuLi4NCjwvamNzPg0KDQpOb3Qgc3VyZSBJ
IHVuZGVyc3RhbmQgeW91ciBjb25jZXJuLiBZZXMsIGluIG15IG1vZGVsIHRoZXJlIGlzIGEgc2Vj
dGlvbiBmb3IgdGhlIGZhYnJpYywgYW5kIHNlY3Rpb25zIGZvciB0aGUgdmFyaW91cyBmdW5jdGlv
bnMuIFNlZW1zIGEgbmF0dXJhbCBzdHJ1Y3R1cmU/IT8NCg0K4oCcRnVuY3Rpb24gb3duZXLigJ06
IFllcywgSeKAmW0gc3VnZ2VzdGluZyB0byBzdHJ1Y3R1cmUgSW50ZW50IGFyb3VuZCBBdXRvbm9t
aWMgRnVuY3Rpb25zLiBUaGlzIGlzIG5vdCBvcHRpbWFsIGZyb20gYSBzY2llbnRpZmljIHBvaW50
IG9mIHZpZXcsIGJ1dCBhbGlnbnMgdmVyeSB3ZWxsIHdpdGggdGhlIHdheSBob3cgdGhpbmdzIGFy
ZSBvcmdhbmlzZWQuIFByb2JhYmx5IHdlIG5lZWQgdG8gZGlzY3VzcyB0aGF0IGluIEJBLiBCdXQg
cmVmZXIgdG8gbXkgWUFORyBtb2RlbCBzZW50IGEgZmV3IGRheXMgYWdvLg0KDQpXZSBzaG91bGQg
c2NoZWR1bGUgc29tZSBkaXNjdXNzaW9uIHRpbWUgb3V0c2lkZSB0aGUgQU5JTUEgbWVldGluZyB3
aW5kb3cuLi4NCg0KTWljaGFlbA0KDQoNCkZyb206IEpvaG4gU3RyYXNzbmVyIFttYWlsdG86c3Ry
YXpwZGpAZ21haWwuY29tXQ0KU2VudDogMjIgTWFyY2ggMjAxNiAyMTozMg0KVG86IE1pY2hhZWwg
QmVocmluZ2VyIChtYmVocmluZykgPG1iZWhyaW5nQGNpc2NvLmNvbT47IEpvaG4gU3RyYXNzbmVy
IDxzdHJhenBkakBnbWFpbC5jb20+DQpDYzogRHV6b25ncGVuZyA8ZHV6b25ncGVuZ0BodWF3ZWku
Y29tPjsgYW5pbWFAaWV0Zi5vcmc7IExhdXJlbnQgQ2lhdmFnbGlhIDxsYXVyZW50LmNpYXZhZ2xp
YUBub2tpYS5jb20+OyBKw6lmZXJzb24gQ2FtcG9zIE5vYnJlIDxqY25vYnJlQGluZi51ZnJncy5i
cj47IFNoZW5nIEppYW5nIDxqaWFuZ3NoZW5nQGh1YXdlaS5jb20+DQpTdWJqZWN0OiBSZTogW0Fu
aW1hXSBEb2VzIEF1dG9ub21pYyBuZXR3b3JrIG5lZWQgc29tZSBpbnRlcnZlbnRpb24gYmV5b25k
IGludGVudA0KDQpJbiBnZW5lcmFsLCBJIGFncmVlIHdpdGggTWljaGFlbC4gSSdkIGxpa2UgdG8g
ZW1waGFzaXplIGFuZCBleHBhbmQgb24gYSBjb3VwbGUgb2Yga2V5IHBvaW50czoNCg0KDQpGaXJz
dCwgQU5JTUEgaXNu4oCZdCB0aGUgb25seSB0aGluZyDigJxpbmZsdWVuY2luZ+KAnSBvciDigJxz
dGVlcmluZ+KAnSBhIG5ldHdvcmssIGF0IGxlYXN0IG5vdCBpbml0aWFsbHkuIFRoZXJlIGlzIHN0
aWxsIHRyYWRpdGlvbmFsIGNvbmZpZywgcGx1cyBhbGwgdGhlIGV4aXN0aW5nIGFuZCBlbWVyZ2lu
ZyBTRE4gbW9kZWxzLiBNeSB2aWV3IGhhcyBhbHdheXMgYmVlbiB0aGF0IEFOSU1BIHNldHMgYSDi
gJxkZWZhdWx04oCdIHBvbGljeSBpbiBhIG5ldHdvcmssIGFuZCBhbGwgb3RoZXIgbWVhbnMgYXJl
IG1vcmUgc3BlY2lmaWMgYW5kIGNhbiBvdmVycmlkZSBpbnRlbnQuIFRoaXMgaXMgZGVzY3JpYmVk
IGluIFJGQzc1NzUuIE1haW4gcG9pbnQgaGVyZTogQU5JTUEgSW50ZW50IGlzbuKAmXQgdGhlIG9u
bHkgdG9vbCBpbiBhIG5ldHdvcmsuDQoNCjxqY3M+DQpBYnNvbHV0ZWx5IGFncmVlLiBOb3RlIHRo
YXQgZXZlbiBpbiB0aGUgQ29uZ3Jlc3MgY2FzZSBJIG1lbnRpb25lZA0KZWFybGllciwgaXQgaXMg
Tk9UIHB1c2hpbmcgY29uZmlnLCBhbmQgaXQgaXMgTk9UIHZpZXdlZCBhcyB0aGUNCiJvbmx5IiBw
b2xpY3kNCjwvamNzPg0KDQoNCkludGVudCBjb3VsZCBwZXJmZWN0bHkgd2VsbCBjb3ZlciBhIGhp
Z2ggbGV2ZWwgcG9saWN5IHN1Y2ggYXMg4oCcYWxsIG5vZGVzIG9mIHR5cGUgeCBkbyB0aGlzOyB0
eXBlIHkgZG9lcyB0aGF04oCdLiAoV2hpY2ggSSB0aGluayBpcyB3aGF0IHlvdSBtZW50aW9uIGJl
bG93LikgV2hlbiB3ZSBzdGFydGVkIHdpdGggSW50ZW50LCB0aGUgdmlzaW9uIHdhcyB0byBrZWVw
IGl0IHZlcnkgaGlnaCBsZXZlbCwgbW9yZSBhbG9uZyB0aGUgbGluZSDigJxkbyB0aGUgcmlnaHQg
dGhpbmfigJ0g4pi6DQoNCjxqY3M+DQpBZ3JlZWQuIEFsdGhvdWdoIHRvIG1lLCB0aGlzIG1lYW5z
IHRoYXQgd2UgbmVlZCB0byBjb25zaWRlciB0aGUgdHlwZXMgb2YgYWN0b3JzIHRoYXQgYXJlIHVz
aW5nIGludGVudC4gRm9yIGV4YW1wbGUsIG1vc3Qgb2YgbXkgQ0NJRSBhbmQgSkNJRSBmcmllbmRz
IGhhdmUgbm8gaW50ZXJlc3QgaW4gdXNpbmcgYW4gaW50ZW50IGxhbmd1YWdlIGZvciBwdXNoaW5n
IGNvbmZpZ3VyYXRpb24uIE9uIHRoZSBvdGhlciBoYW5kLCB3aGVuIEkgdGFsayB0byBvcGVyYXRv
cnMgYW5kIGFwcGxpY2F0aW9uIGRldmVsb3BlcnMsIHRoZXkgd291bGQgbG92ZSB0byBiZSBhYmxl
IHRvIHVzZSBhIGhpZ2gtbGV2ZWwgaW50ZW50IGFuZCBoYXZlIHRoZSBpbnRlbnQgZW5naW5lIHRh
a2UgY2FyZSBvZiB0aGUgZGV0YWlscyBmb3IgdGhlbS4NCjwvamNzPg0KDQpJdCBoYXMgYmVjb21l
IGNsZWFyIHRoYXQgaW4gdG9kYXnigJlzIG5ldHdvcmtpbmcsIHBlb3BsZSB3aWxsIHdhbnQgdG8g
cHVzaCBtb3JlIOKAnHBhcmFtZXRlcnPigJ0gdGhhbiDigJxwb2xpY3nigJ0sIGFuZCBJIHRoaW5r
IHdl4oCZbGwganVzdCBoYXZlIHRvIGFja25vd2xlZGdlIHRoYXQuDQoNCjxqY3M+DQpTb3JyeSBN
aWNoYWVsLCBJIGRvbid0IGZvbGxvdyBoZXJlLiBXaGF0IGRvIHlvdSBtZWFuDQpieSAicGFyYW1l
dGVycyI/DQo8L2pjcz4NCg0KDQpQZXJzb25hbGx5IEkgdGhpbmsgaXTigJlsbCBwYW4gb3V0IGxp
a2UgdGhpczogSW50ZW50IHdpbGwgaGF2ZSBzZWN0aW9ucyBwZXIgYXV0b25vbWljIGZ1bmN0aW9u
LiBBbiBJbnRlbnQgcGFyc2VyIHdpbGwgdmFsaWRhdGUgb3ZlcmFsbCBjb3JyZWN0bmVzcyAoc2ln
bmF0dXJlLCBldGMpLCB0aGVuIHNlZ21lbnQgdGhlIEludGVudCBpbnRvIGl0cyBiaXRzOiBPbmUg
cGFydCB3aWxsIGJlIHRoZSBhdXRvbm9taWMgZmFicmljIGl0c2VsZiwgdGhlbiB0aGVyZSB3aWxs
IGJlIGJpdHMgZm9yIGVhY2ggYXV0b25vbWljIGZ1bmN0aW9uLg0KDQo8amNzPg0KRG8geW91IG1l
YW4gdGhhdCB0aGUgaW50ZW50IHdyaXRlciB3aWxsICoqa25vdyB0aGUgZGlmZmVyZW5jZSoqDQpi
ZXR3ZWVuIHRoZSBmYWJyaWMgYW5kIGVhY2ggYXV0b25vbWljIGZ1bmN0aW9uPyBJIGhvcGUgbm90
LA0KYnV0IHBlcmhhcHMgSSdtIG1pc3VuZGVyc3RhbmRpbmcgeW91Li4uDQo8L2pjcz4NCg0KDQpT
byB0aGUgSW50ZW50IHBhcnNlciBvbiBhIG5vZGUgd2lsbCBncmFiIGludGVudCwgc2VlIHRoZXJl
IGlzIHNvbWUgSW50ZW50IGZvciBhdXRvbm9taWMgZnVuY3Rpb24gQSwgd2lsbCBzZWUgd2hldGhl
ciB0aGlzIGlzIHJ1bm5pbmcgbG9jYWxseSwgYW5kIGlmIHNvLCBpdOKAmWxsIGp1c3QgcGFzcyB0
aGF0IHBpZWNlIG9mIEludGVudCB0byB0aGF0IGZ1bmN0aW9uLg0KDQo8amNzPg0KQWdyZWVkLg0K
PC9qY3M+DQoNCg0KSW4gdGhpcyB3YXksIGEgZnVuY3Rpb24gb3duZXIgZGVmaW5lcyBib3RoIHRo
ZSBpbnRlbnQgaXRzZWxmLCBhbmQgaG93IGl0IGlzIGNvbnN1bWVkIGJ5IGhpcy9oZXIgZnVuY3Rp
b24uDQoNCjxqY3M+DQpJJ20gbm90IHN1cmUgd2hhdCB5b3UgbWVhbiBieSAiZnVuY3Rpb24gb3du
ZXIiLiAoU29ycnkgdG8gYmUNCnBpY2t5LCBJJ20gbGlrZWx5IHVuZGVyLWNhZmZlaW5hdGVkLikN
Cg0KVGhlIGFib3ZlIHNlZW1lZCBsaWtlIGFuIGF1dG9ub21vdXMgZGVjaXNpb24sIG1hZGUgYnkg
dGhlIHBhcnNlci4NCkkgZG9uJ3QgdGhpbmsgdGhhdCB0aGUgaW50ZW50IHdyaXRlciBzaG91bGQg
YmUgY29nbml6YW50IG9mIGhvdyBtYW55IG9yDQp3aGF0IHR5cGVzIG9mIGF1dG9ub21pYyBmdW5j
dGlvbnMgYXJlIHByZXNlbnQuDQo8L2pjcz4NCg0KDQpUaGUgZG93bnNpZGUgaXMgdGhhdCB0aGVy
ZSBpcyBubyDigJxjb250cm9s4oCdIGFuZCBwZW9wbGUgY2FuIGRvIHdoYXQgdGhleSB3YW50LCBp
biB0aGUgd29yc3QgY2FzZSBldmVuIHB1c2ggY29uZmlnLCB3aGljaCBpcyBjbGVhcmx5IE5PVCB3
aGF0IEludGVudCBpcyBtZWFudCB0byBkby4NCg0KPGpjcz4NCkFuZCBoZW5jZSwgdGhlICJIQUwg
MjAwMCIgc2NlbmFyaW8uIDotKQ0KPC9qY3M+DQoNCg0KSeKAmWQgaG9wZSB0aGF0IHBlb3BsZSBk
b27igJl0IHN0YXJ0IHdyaXRpbmcgSW50ZW50IGxpa2Ug4oCcbm9kZSAxNzogaGVyZSBpcyB5b3Vy
IENMSTsgbm9kZSAyMzogaGVyZSBpcyB5b3VyIENMSeKAnTsgbm9kZSAzNzogaGVyZSBpcyB5b3Vy
IENMSeKAnS4gQnV0IHdoYXQgSSBkZXNjcmliZSBhYm92ZSBDT1VMRCBiZSBhYnVzZWQgaW4gdGhp
cyB3YXkuIFJGQzc1NzUgc2F5cyB0aGlzIGlzIE5PVCBJbnRlbnQuDQoNCjxqY3M+DQpBYnNvbHV0
ZWx5IGFncmVlIQ0KPC9qY3M+DQoNCiByZWdhcmRzLA0KSm9obg0KDQoNCg0KT24gV2VkLCBNYXIg
MTYsIDIwMTYgYXQgMToxNiBBTSwgTWljaGFlbCBCZWhyaW5nZXIgKG1iZWhyaW5nKSA8bWJlaHJp
bmdAY2lzY28uY29tPG1haWx0bzptYmVocmluZ0BjaXNjby5jb20+PiB3cm90ZToNCkhpIFpvbmdw
ZW5nLA0KDQpZZXMsIHRoaXMgaXMgYSByZWN1cnJpbmcgcXVlc3Rpb24gYW5kIHRvcGljLiBMZXQg
bWUgZ2l2ZSBteSB2aWV3IG9uIHR3byBsZXZlbHM6DQoNCkZpcnN0LCBBTklNQSBpc27igJl0IHRo
ZSBvbmx5IHRoaW5nIOKAnGluZmx1ZW5jaW5n4oCdIG9yIOKAnHN0ZWVyaW5n4oCdIGEgbmV0d29y
aywgYXQgbGVhc3Qgbm90IGluaXRpYWxseS4gVGhlcmUgaXMgc3RpbGwgdHJhZGl0aW9uYWwgY29u
ZmlnLCBwbHVzIGFsbCB0aGUgZXhpc3RpbmcgYW5kIGVtZXJnaW5nIFNETiBtb2RlbHMuIE15IHZp
ZXcgaGFzIGFsd2F5cyBiZWVuIHRoYXQgQU5JTUEgc2V0cyBhIOKAnGRlZmF1bHTigJ0gcG9saWN5
IGluIGEgbmV0d29yaywgYW5kIGFsbCBvdGhlciBtZWFucyBhcmUgbW9yZSBzcGVjaWZpYyBhbmQg
Y2FuIG92ZXJyaWRlIGludGVudC4gVGhpcyBpcyBkZXNjcmliZWQgaW4gUkZDNzU3NS4gTWFpbiBw
b2ludCBoZXJlOiBBTklNQSBJbnRlbnQgaXNu4oCZdCB0aGUgb25seSB0b29sIGluIGEgbmV0d29y
ay4NCg0KSW50ZW50IGNvdWxkIHBlcmZlY3RseSB3ZWxsIGNvdmVyIGEgaGlnaCBsZXZlbCBwb2xp
Y3kgc3VjaCBhcyDigJxhbGwgbm9kZXMgb2YgdHlwZSB4IGRvIHRoaXM7IHR5cGUgeSBkb2VzIHRo
YXTigJ0uIChXaGljaCBJIHRoaW5rIGlzIHdoYXQgeW91IG1lbnRpb24gYmVsb3cuKSBXaGVuIHdl
IHN0YXJ0ZWQgd2l0aCBJbnRlbnQsIHRoZSB2aXNpb24gd2FzIHRvIGtlZXAgaXQgdmVyeSBoaWdo
IGxldmVsLCBtb3JlIGFsb25nIHRoZSBsaW5lIOKAnGRvIHRoZSByaWdodCB0aGluZ+KAnSDimLog
IEl0IGhhcyBiZWNvbWUgY2xlYXIgdGhhdCBpbiB0b2RheeKAmXMgbmV0d29ya2luZywgcGVvcGxl
IHdpbGwgd2FudCB0byBwdXNoIG1vcmUg4oCccGFyYW1ldGVyc+KAnSB0aGFuIOKAnHBvbGljeeKA
nSwgYW5kIEkgdGhpbmsgd2XigJlsbCBqdXN0IGhhdmUgdG8gYWNrbm93bGVkZ2UgdGhhdC4NCg0K
UGVyc29uYWxseSBJIHRoaW5rIGl04oCZbGwgcGFuIG91dCBsaWtlIHRoaXM6IEludGVudCB3aWxs
IGhhdmUgc2VjdGlvbnMgcGVyIGF1dG9ub21pYyBmdW5jdGlvbi4gQW4gSW50ZW50IHBhcnNlciB3
aWxsIHZhbGlkYXRlIG92ZXJhbGwgY29ycmVjdG5lc3MgKHNpZ25hdHVyZSwgZXRjKSwgdGhlbiBz
ZWdtZW50IHRoZSBJbnRlbnQgaW50byBpdHMgYml0czogT25lIHBhcnQgd2lsbCBiZSB0aGUgYXV0
b25vbWljIGZhYnJpYyBpdHNlbGYsIHRoZW4gdGhlcmUgd2lsbCBiZSBiaXRzIGZvciBlYWNoIGF1
dG9ub21pYyBmdW5jdGlvbi4NCg0KU28gdGhlIEludGVudCBwYXJzZXIgb24gYSBub2RlIHdpbGwg
Z3JhYiBpbnRlbnQsIHNlZSB0aGVyZSBpcyBzb21lIEludGVudCBmb3IgYXV0b25vbWljIGZ1bmN0
aW9uIEEsIHdpbGwgc2VlIHdoZXRoZXIgdGhpcyBpcyBydW5uaW5nIGxvY2FsbHksIGFuZCBpZiBz
bywgaXTigJlsbCBqdXN0IHBhc3MgdGhhdCBwaWVjZSBvZiBJbnRlbnQgdG8gdGhhdCBmdW5jdGlv
bi4NCg0KSW4gdGhpcyB3YXksIGEgZnVuY3Rpb24gb3duZXIgZGVmaW5lcyBib3RoIHRoZSBpbnRl
bnQgaXRzZWxmLCBhbmQgaG93IGl0IGlzIGNvbnN1bWVkIGJ5IGhpcy9oZXIgZnVuY3Rpb24uDQoN
ClRoZSBkb3duc2lkZSBpcyB0aGF0IHRoZXJlIGlzIG5vIOKAnGNvbnRyb2zigJ0gYW5kIHBlb3Bs
ZSBjYW4gZG8gd2hhdCB0aGV5IHdhbnQsIGluIHRoZSB3b3JzdCBjYXNlIGV2ZW4gcHVzaCBjb25m
aWcsIHdoaWNoIGlzIGNsZWFybHkgTk9UIHdoYXQgSW50ZW50IGlzIG1lYW50IHRvIGRvLiBJ4oCZ
ZCBob3BlIHRoYXQgcGVvcGxlIGRvbuKAmXQgc3RhcnQgd3JpdGluZyBJbnRlbnQgbGlrZSDigJxu
b2RlIDE3OiBoZXJlIGlzIHlvdXIgQ0xJOyBub2RlIDIzOiBoZXJlIGlzIHlvdXIgQ0xJ4oCdOyBu
b2RlIDM3OiBoZXJlIGlzIHlvdXIgQ0xJ4oCdLiBCdXQgd2hhdCBJIGRlc2NyaWJlIGFib3ZlIENP
VUxEIGJlIGFidXNlZCBpbiB0aGlzIHdheS4gUkZDNzU3NSBzYXlzIHRoaXMgaXMgTk9UIEludGVu
dC4NCg0KSW4gb3RoZXIgd29yZHM6IExldOKAmXMgbm90IGJlIHRvbyBwcmVzY3JpcHRpdmUsIGFu
ZCBob3BlIHRoYXQgZm9sa3Mgd2lsbCBkbyBtb3JlIG9yIGxlc3MgcmVhc29uYWJsZSB0aGluZ3Mu
DQoNCknigJl2ZSBiZWVuIG1lYW5pbmcgdG8gY29tbWVudCBvbiB0aGUgSW50ZW50IGRyYWZ0cyBh
bmQgcG9zc2libHkgY29udHJpYnV0ZSwgSeKAmW0ganVzdCBzdHJ1Z2dsaW5nIHdpdGggcHJpb3Jp
dGllcyBhdCB0aGUgbW9tZW50IChhcyB5b3UgY2FuIHNlZSBieSBteSByZWxhdGl2ZSBzaWxlbmNl
IG92ZXIgdGhlIHBhc3Qgd2Vla3MpLg0KDQpNaWNoYWVsDQoNCg0KDQpGcm9tOiBBbmltYSBbbWFp
bHRvOmFuaW1hLWJvdW5jZXNAaWV0Zi5vcmc8bWFpbHRvOmFuaW1hLWJvdW5jZXNAaWV0Zi5vcmc+
XSBPbiBCZWhhbGYgT2YgRHV6b25ncGVuZw0KU2VudDogMTYgTWFyY2ggMjAxNiAwMjozOA0KVG86
IGFuaW1hQGlldGYub3JnPG1haWx0bzphbmltYUBpZXRmLm9yZz4NCkNjOiBMYXVyZW50IENpYXZh
Z2xpYSA8bGF1cmVudC5jaWF2YWdsaWFAbm9raWEuY29tPG1haWx0bzpsYXVyZW50LmNpYXZhZ2xp
YUBub2tpYS5jb20+PjsgSsOpZmVyc29uIENhbXBvcyBOb2JyZSA8amNub2JyZUBpbmYudWZyZ3Mu
YnI8bWFpbHRvOmpjbm9icmVAaW5mLnVmcmdzLmJyPj47IFNoZW5nIEppYW5nIDxqaWFuZ3NoZW5n
QGh1YXdlaS5jb208bWFpbHRvOmppYW5nc2hlbmdAaHVhd2VpLmNvbT4+DQpTdWJqZWN0OiBbQW5p
bWFdIERvZXMgQXV0b25vbWljIG5ldHdvcmsgbmVlZCBzb21lIGludGVydmVudGlvbiBiZXlvbmQg
aW50ZW50DQoNCkhlbGxvLCBldmVyeW9uZQ0KDQogICAgICAgICBJIHVuZGVyc3RhbmQgdGhhdCBB
dXRvbm9taWMgbmV0d29yayBzaG91bGQgYmUgYWJsZSB0byB3b3JrIHdpdGggYXMgbGVzcyBjb25m
aWd1cmF0aW9uIGFzIHBvc3NpYmxlLiBJbiB0aGUgaWRlYWwgY2FzZSwgYW4gQXV0b25vbWljIG5l
dHdvcmsgc2hvdWxkIGJlIGFibGUgdG8gd29yayB3ZWxsIHdoaWxlIG9ubHkgaW50ZW50IGlzIG5l
ZWRlZCBmcm9tIGh1bWFuIG9wZXJhdG9ycy4NCg0KICAgICAgICAgSG93ZXZlciwgSSB3b25kZXIg
d2hldGhlciBBdXRvbm9taWMgbmV0d29yayBtYXkgbmVlZCBzb21lIGludGVydmVudGlvbiBmcm9t
IGh1bWFuIG9wZXJhdG9ycyB0aHJvdWdoIHNvbWUgbG93LWxldmVsIHBvbGljaWVzIGJleW9uZCBp
bnRlbnRzLg0KDQogICAgICAgICBGb3IgZXhhbXBsZSwgaW4gdGhlIGh0dHBzOi8vdG9vbHMuaWV0
Zi5vcmcvaHRtbC9kcmFmdC1pZXRmLWFuaW1hLXByZWZpeC1tYW5hZ2VtZW50LTAwLCBpdCBpcyBz
dWdnZXN0ZWQgdGhhdCB0aGUgcHJlZml4IGxlbmd0aHMgZm9yIHRoZSBDU0csIEFTRywgUlNHIChk
aWZmZXJlbnQgcm9sZXMgaW4gSVAgUkFOKSBjYW4gYmUgYXNzaWduZWQgYXMgYW4gImludGVudCIu
IFRoZXNlIGNvbmZpZ3VyYXRpb24gcGFyYW1ldGVycyBuZWVkIHRvIGJlIGRpc3RyaWJ1dGVkIGlu
IHRoZSBhdXRvbm9taWMgZG9tYWluIHRvIGluZmx1ZW5jZSB0aGUgZGV0YWlsIGNvbmZpZ3VyYXRp
b25zIG9uIGVhY2ggYXV0b25vbWljIG5vZGUuDQoNCiAgICAgICAgIEFzIGludGVudCBpcyBkZXNj
cmliZWQgYXMgYW4gYWJzdHJhY3QsIGRlY2xhcmF0aXZlLCBoaWdoLWxldmVsIHBvbGljeSB1c2Vk
IHRvIG9wZXJhdGUgYW4gYXV0b25vbWljIGRvbWFpbiwgc3VjaCBhcyBhbiBlbnRlcnByaXNlIG5l
dHdvcmsuIFBlcmhhcHMgd2UgY2FuIHJlbmFtZSB0aGUg4oCcaW50ZW504oCdIGluIHRoZSDigJxw
cmVmaXgtbWFuYWdlbWVudCBJROKAnSB0byBzb21lIG90aGVyIHdvcmRzLg0KDQpJcyDigJxBU0Eg
cGFyYW1ldGVycyIgb2sgaGVyZT8NCg0KSSB0aGluayBzb21lIGNoYXJhY3RlcmlzdGljcyBmb3Ig
dGhpcyBraW5kIG9mIGxvdy1sZXZlbCBwb2xpY2llcyBhcmUNCg0KMS4gICAgICAgTmV0d29yayBs
ZXZlbCBwYXJhbWV0ZXJzLCBuZWVkIHRvIGJlIGRpc3RyaWJ1dGVkIGJ5IEdSQVNQIGFuZCBBQ1AN
Cg0KMi4gICAgICAgUmVsYXRpdmVseSBzdGF0aWMgY29tcGFyZWQgdG8gaW50ZW50LCBwZXJoYXBz
IG9ubHkgbmVlZGVkIHRvIGJlIGNvbmZpZ3VyZWQgb25jZQ0KDQozLiAgICAgICBVbmF2b2lkYWJs
ZSBpbiBjdXJyZW50IG5ldHdvcmsgZW52aXJvbm1lbnQsIG1vc3RseSBmb3IgZXN0YWJsaXNoaW5n
IHRoZSBuZXR3b3JrIGluZnJhc3RydWN0dXJlDQoNCjQuICAgICAgIFJhcmVseSBuZWVkIGNvb3Jk
aW5hdGlvbnMgd2l0aCBvdGhlcnMgcGFyYW1ldGVycywgdXN1YWxseSBpcyBjbG9zZWQgcmVsYXRl
ZCB0byBvbmUgc3BlY2lmaWMgYXV0b25vbWljIGZ1bmN0aW9uDQoNCldlbGNvbWUgZm9yIGNvbW1l
bnQuIFlvdXIgc3VnZ2VzdGlvbnMgd2lsbCBiZSBhcHByZWNpYXRlZC4NCg0KDQpCZXN0IFJlZ2Fy
ZHMNClpvbmdwZW5nIER1DQoNCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fDQpBbmltYSBtYWlsaW5nIGxpc3QNCkFuaW1hQGlldGYub3JnPG1haWx0bzpBbmlt
YUBpZXRmLm9yZz4NCmh0dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vYW5pbWEN
Cg0KDQoNCi0tDQpyZWdhcmRzLA0KSm9obg0K

--_000_8368235337ea4adfab5c428d57c3bfb2XCHRCD006ciscocom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_8368235337ea4adfab5c428d57c3bfb2XCHRCD006ciscocom_--


From nobody Wed Mar 23 06:09:37 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EFC4F12DD41 for ; Wed, 23 Mar 2016 06:09:32 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.52
X-Spam-Level: 
X-Spam-Status: No, score=-14.52 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_KAM_HTML_FONT_INVALID=0.01, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0QBwGLFyW61G for ; Wed, 23 Mar 2016 06:09:24 -0700 (PDT)
Received: from alln-iport-1.cisco.com (alln-iport-1.cisco.com [173.37.142.88]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9BF9912D63B for ; Wed, 23 Mar 2016 05:58:28 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=97002; q=dns/txt; s=iport; t=1458737908; x=1459947508; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=MXiRw/sAmCXP56+bfL9Rtd29yq3Mz/7x6TakXfrKk3k=; b=aW1jmA2ZI2LX3BhE02f2L2Zt+1GNzN+9DWkEFxtdq0ijmiq/h2Yl8jIy 49Sh3Ceci6hE/tazQvTtbEXhqYvxfBMCQogypACUnbPQ4fNaV6cfioHkM YdlT91R4f9a1PW5upbmUq5Xw8Yd14t/vnalqTT0eNpxEZsT/2nasm7PKu g=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AKAgC6kfJW/5ldJa1egmdMU3oGrw6LT?= =?us-ascii?q?gENgW0DFwEJhWwCHIEfOBQBAQEBAQEBZCeEQQEBAQQBAQEXCQpBCxACAQYCEQQ?= =?us-ascii?q?BASEBBgMCAgIfBgsUCQgBAQQBDQUIF4dzAxIOkkWdF4wGDYRxAQEBAQEBAQEBA?= =?us-ascii?q?QEBAQEBAQEBAQEBFYYehESCPoIUFwmCSoJWBZMEhCUxAYVwgnKDLIFugW2NJIY?= =?us-ascii?q?OgSSHVAEPDwEBQoNlaohSJRZ+AQEB?=
X-IronPort-AV: E=Sophos;i="5.24,382,1454976000";  d="scan'208,217";a="252880226"
Received: from rcdn-core-2.cisco.com ([173.37.93.153]) by alln-iport-1.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 23 Mar 2016 12:58:26 +0000
Received: from XCH-ALN-006.cisco.com (xch-aln-006.cisco.com [173.36.7.16]) by rcdn-core-2.cisco.com (8.14.5/8.14.5) with ESMTP id u2NCwQGj003422 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Wed, 23 Mar 2016 12:58:26 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-006.cisco.com (173.36.7.16) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Wed, 23 Mar 2016 07:58:25 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Wed, 23 Mar 2016 07:58:25 -0500
From: "Michael Behringer (mbehring)" 
To: Laurent Ciavaglia , John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAVKENoAADnV0sAARwEGAAAiQQxA=
Date: Wed, 23 Mar 2016 12:58:25 +0000
Message-ID: <0ce0e3160cf746d3b03724075c42040f@XCH-RCD-006.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <56F28403.6060401@alcatel-lucent.com>
In-Reply-To: <56F28403.6060401@alcatel-lucent.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.61.106.202]
Content-Type: multipart/alternative; boundary="_000_0ce0e3160cf746d3b03724075c42040fXCHRCD006ciscocom_"
MIME-Version: 1.0
Archived-At: 
Cc: Duzongpeng , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 13:09:33 -0000

--_000_0ce0e3160cf746d3b03724075c42040fXCHRCD006ciscocom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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=

--_000_0ce0e3160cf746d3b03724075c42040fXCHRCD006ciscocom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_0ce0e3160cf746d3b03724075c42040fXCHRCD006ciscocom_--


From nobody Wed Mar 23 10:57:03 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4AA0012D0EF for ; Wed, 23 Mar 2016 10:57:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.53
X-Spam-Level: 
X-Spam-Status: No, score=-14.53 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fpHCBQGjmPfe for ; Wed, 23 Mar 2016 10:56:59 -0700 (PDT)
Received: from alln-iport-4.cisco.com (alln-iport-4.cisco.com [173.37.142.91]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4A86612D1A6 for ; Wed, 23 Mar 2016 10:56:58 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=72872; q=dns/txt; s=iport; t=1458755818; x=1459965418; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=6edqeM2X/GiuyYKWDv8dr9qHAZcyAGiSm858mWYXu8o=; b=EMUuG53sgrTD/VclGNrATs/wK8JHuq9vkRdx7RjAIlQrN+1219fqt352 tRzctSmj2FcsMB+OlWza/XhJRuMWswxUfdvpiFgKGaYn99p+yqDb/MawG CCLQvAQCewQAZVRSdoO5qDKcTrtRExZCLSSwZIF1m407lSq3V24fVyVex c=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0A1AgBO2PJW/5tdJa1egmdMU3oGrxSLT?= =?us-ascii?q?gENgW0DFwEJhWwCHIEiOBQBAQEBAQEBZCeEQQEBAQQBAQEXCQpBCxACAQYCEQQ?= =?us-ascii?q?BASEBBgMCAgIfBgsUCQgBAQQBDQUIiAoDEg6TPp0Xi3wNhGABAQEBAQEBAQEBA?= =?us-ascii?q?QEBAQEBAQEBAQERBIpigj6CKwmCSoJWBZMEhCUxAYVwhh6Bbo8RhzKHVAEeAQF?= =?us-ascii?q?Cg2VqiFIlFgF9AQEB?=
X-IronPort-AV: E=Sophos;i="5.24,383,1454976000";  d="scan'208,217";a="252040143"
Received: from rcdn-core-4.cisco.com ([173.37.93.155]) by alln-iport-4.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 23 Mar 2016 17:56:56 +0000
Received: from XCH-RTP-009.cisco.com (xch-rtp-009.cisco.com [64.101.220.149]) by rcdn-core-4.cisco.com (8.14.5/8.14.5) with ESMTP id u2NHuuKB005100 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Wed, 23 Mar 2016 17:56:56 GMT
Received: from xch-rtp-001.cisco.com (64.101.220.141) by XCH-RTP-009.cisco.com (64.101.220.149) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Wed, 23 Mar 2016 13:56:54 -0400
Received: from xch-rtp-001.cisco.com ([64.101.220.141]) by XCH-RTP-001.cisco.com ([64.101.220.141]) with mapi id 15.00.1104.009; Wed, 23 Mar 2016 13:56:55 -0400
From: "Alexander Clemm (alex)" 
To: "Michael Behringer (mbehring)" , John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAVBrxYAAHjH7AAAF5yGg
Date: Wed, 23 Mar 2016 17:56:55 +0000
Message-ID: <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>
In-Reply-To: <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [128.107.151.23]
Content-Type: multipart/alternative; boundary="_000_6aca826a56994d5898481b70e0cb5196XCHRTP001ciscocom_"
MIME-Version: 1.0
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 17:57:02 -0000

--_000_6aca826a56994d5898481b70e0cb5196XCHRTP001ciscocom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_6aca826a56994d5898481b70e0cb5196XCHRTP001ciscocom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_6aca826a56994d5898481b70e0cb5196XCHRTP001ciscocom_--


From nobody Wed Mar 23 11:43:52 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 156F612D6F7 for ; Wed, 23 Mar 2016 11:43:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.702
X-Spam-Level: 
X-Spam-Status: No, score=-2.702 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EZdNBzoxKDah for ; Wed, 23 Mar 2016 11:43:43 -0700 (PDT)
Received: from maila2.tigertech.net (maila2.tigertech.net [208.80.4.152]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A68DC12D597 for ; Wed, 23 Mar 2016 11:43:43 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by maila2.tigertech.net (Postfix) with ESMTP id 5DBD9241309; Wed, 23 Mar 2016 11:43:43 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1458758623; bh=qlUFk1XQnZoW7Tyh1AaANjH68Lif7IgzkJbgpNvuVrU=; h=Subject:To:References:Cc:From:Date:In-Reply-To:From; b=N6S6EMi2IIRecwnfW/9292Bk62hULPpFCUrjgVDz2xFeyO3Sn47su0TUsiPPRe5+T u9v+Dy7xSDsf2Ax90SY/lFokweN+FWqFNvNWtRdCzAbNCbgzo8m12alPgxkl8ajE7z fBFvzIsMSTnpKLvl6LoeCKhCRx673zeo+RscjnmY=
X-Virus-Scanned: Debian amavisd-new at maila2.tigertech.net
Received: from Joels-MacBook-Pro.local (unknown [88.131.67.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by maila2.tigertech.net (Postfix) with ESMTPSA id 2022625096B; Wed, 23 Mar 2016 11:43:40 -0700 (PDT)
To: "Alexander Clemm (alex)" , "Michael Behringer (mbehring)" , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>
From: "Joel M. Halpern" 
Message-ID: <56F2E3DA.9030204@joelhalpern.com>
Date: Wed, 23 Mar 2016 14:43:38 -0400
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 8bit
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , Sheng Jiang , "anima@ietf.org" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 18:43:51 -0000

I think the quesiton is what the scope of an intent is.
If, as seems likely, a given intent is for processing by a given objective,
and given that the objectives each form a connected graph over the domain,
it would seem sensible to flood the intent to the participants in the 
objective.

If intents are not mapped (by the time they get into ANIMA) to a single 
objective, then it probably makes more sense to flood more generally.

To some degree this also depend on how often we expect intents to be 
introduced, modified, or deleted.  If that rate is low, then flooding 
everywhere is probably cheap.  If that rate is high, I would want to be 
able to constrain the propagation.

Yours,
Joel

On 3/23/16 1:56 PM, Alexander Clemm (alex) wrote:
> I really like your suggestion of flooding intent to all nodes (even if
> there may be optimizations later).  It provides a level of conceptual
> clarity.  Every node should be aware of overall intent (and need to
> decide for itself if there are actions for it to take).
>
> If there is some intent that applies only to some parts of the network
> and not others, that can be addressed by using different Autonomic
> Domains.  And, of course, Autonomic Domains may be associated with
> certain Autonomic Functions.  Intervention beyond intent should NOT be
> allowed for the Autonomic Function(s) being served an Autonomic Domain,
> but of course other interventions (for different functions, hence
> outside the scope) are still possible.
>
> --- Alex
>
> *From:*Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Michael
> Behringer (mbehring)
> *Sent:* Wednesday, March 23, 2016 3:57 AM
> *To:* John Strassner 
> *Cc:* Duzongpeng ; Laurent Ciavaglia
> ; Sheng Jiang ;
> anima@ietf.org; Jéferson Campos Nobre 
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
> John,
>
> I think we’re on the same page. Couple of points, top posting here,
> otherwise it gets too messy.
>
> While *CIEs don’t like magic, they do understand the concept of a
> default behaviour which they can influence if needed. Any reasonable
> network management approach tries to define “default” behaviour and
> “exceptions”. The AN story fills the “default” part very nicely, and
> *CIEs get that.
>
> The “parameter” versus “policy” discussion: My view is that we shouldn’t
> try to define exactly **what** intent allows and doesn’t allow, since
> we’ve heard from many sides that the boundary is not clear, and it may
> be impossible to have a really clear, usable definition. So my point
> was: Let’s focus on the distribution model: Intent is flooded to all
> nodes. Let’s NOT do selective flooding, unicast Intent to specific
> nodes, etc, because that would blur the boundary even worse.
>
> My suggestion: Intent is flooded to all nodes in a domain. The ASA
> owners define the content of what they want to push in their bit of the
> Intent. If we insist of full flooding, we encourage the right behaviour.
> And if someone really puts into his Intent “node A: do this; node B: do
> the other; node C: do the third” it is awkward enough for people to take
> a step back and think whether they’re doing the right thing J
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
> Not sure I understand your concern. Yes, in my model there is a section
> for the fabric, and sections for the various functions. Seems a natural
> structure?!?
>
> “Function owner”: Yes, I’m suggesting to structure Intent around
> Autonomic Functions. This is not optimal from a scientific point of
> view, but aligns very well with the way how things are organised.
> Probably we need to discuss that in BA. But refer to my YANG model sent
> a few days ago.
>
> We should schedule some discussion time outside the ANIMA meeting window...
>
> Michael
>
> *From:*John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* 22 March 2016 21:32
> *To:* Michael Behringer (mbehring)  >; John Strassner  >
> *Cc:* Duzongpeng >;
> anima@ietf.org ; Laurent Ciavaglia
> >;
> Jéferson Campos Nobre  >; Sheng Jiang  >
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
> In general, I agree with Michael. I'd like to emphasize and expand on a
> couple of key points:
>
> First, ANIMA isn’t the only thing “influencing” or “steering” a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA
> sets a “default” policy in a network, and all other means are more
> specific and can override intent. This is described in RFC7575. Main
> point here: ANIMA Intent isn’t the only tool in a network.
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
> Intent could perfectly well cover a high level policy such as “all nodes
> of type x do this; type y does that”. (Which I think is what you mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line “do the right thing” J
>
> 
>
> Agreed. Although to me, this means that we need to consider the types of
> actors that are using intent. For example, most of my CCIE and JCIE
> friends have no interest in using an intent language for pushing
> configuration. On the other hand, when I talk to operators and
> application developers, they would love to be able to use a high-level
> intent and have the intent engine take care of the details for them.
>
> 
>
> It has become clear that in today’s networking, people will want to push
> more “parameters” than “policy”, and I think we’ll just have to
> acknowledge that.
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
> Personally I think it’ll pan out like this: Intent will have sections
> per autonomic function. An Intent parser will validate overall
> correctness (signature, etc), then segment the Intent into its bits: One
> part will be the autonomic fabric itself, then there will be bits for
> each autonomic function.
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
> So the Intent parser on a node will grab intent, see there is some
> Intent for autonomic function A, will see whether this is running
> locally, and if so, it’ll just pass that piece of Intent to that function.
>
> 
>
> Agreed.
>
> 
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
> The downside is that there is no “control” and people can do what they
> want, in the worst case even push config, which is clearly NOT what
> Intent is meant to do.
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
> I’d hope that people don’t start writing Intent like “node 17: here is
> your CLI; node 23: here is your CLI”; node 37: here is your CLI”. But
> what I describe above COULD be abused in this way. RFC7575 says this is
> NOT Intent.
>
> 
>
> Absolutely agree!
>
> 
>
>   regards,
>
> John
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring)
> > wrote:
>
>     Hi Zongpeng,
>
>     Yes, this is a recurring question and topic. Let me give my view on
>     two levels:
>
>     First, ANIMA isn’t the only thing “influencing” or “steering” a
>     network, at least not initially. There is still traditional config,
>     plus all the existing and emerging SDN models. My view has always
>     been that ANIMA sets a “default” policy in a network, and all other
>     means are more specific and can override intent. This is described
>     in RFC7575. Main point here: ANIMA Intent isn’t the only tool in a
>     network.
>
>     Intent could perfectly well cover a high level policy such as “all
>     nodes of type x do this; type y does that”. (Which I think is what
>     you mention below.) When we started with Intent, the vision was to
>     keep it very high level, more along the line “do the right thing” J
>     It has become clear that in today’s networking, people will want to
>     push more “parameters” than “policy”, and I think we’ll just have to
>     acknowledge that.
>
>     Personally I think it’ll pan out like this: Intent will have
>     sections per autonomic function. An Intent parser will validate
>     overall correctness (signature, etc), then segment the Intent into
>     its bits: One part will be the autonomic fabric itself, then there
>     will be bits for each autonomic function.
>
>     So the Intent parser on a node will grab intent, see there is some
>     Intent for autonomic function A, will see whether this is running
>     locally, and if so, it’ll just pass that piece of Intent to that
>     function.
>
>     In this way, a function owner defines both the intent itself, and
>     how it is consumed by his/her function.
>
>     The downside is that there is no “control” and people can do what
>     they want, in the worst case even push config, which is clearly NOT
>     what Intent is meant to do. I’d hope that people don’t start writing
>     Intent like “node 17: here is your CLI; node 23: here is your CLI”;
>     node 37: here is your CLI”. But what I describe above COULD be
>     abused in this way. RFC7575 says this is NOT Intent.
>
>     In other words: Let’s not be too prescriptive, and hope that folks
>     will do more or less reasonable things.
>
>     I’ve been meaning to comment on the Intent drafts and possibly
>     contribute, I’m just struggling with priorities at the moment (as
>     you can see by my relative silence over the past weeks).
>
>     Michael
>
>     *From:*Anima [mailto:anima-bounces@ietf.org
>     ] *On Behalf Of *Duzongpeng
>     *Sent:* 16 March 2016 02:38
>     *To:* anima@ietf.org 
>     *Cc:* Laurent Ciavaglia      >; Jéferson Campos Nobre
>     >; Sheng Jiang
>     >
>     *Subject:* [Anima] Does Autonomic network need some intervention
>     beyond intent
>
>     Hello, everyone
>
>               I understand that Autonomic network should be able to work
>     with as less configuration as possible. In the ideal case, an
>     Autonomic network should be able to work well while only intent is
>     needed from human operators.
>
>               However, I wonder whether Autonomic network may need some
>     intervention from human operators through some low-level policies
>     beyond intents.
>
>               For example, in the
>     https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,
>     it is suggested that the prefix lengths for the CSG, ASG, RSG
>     (different roles in IP RAN) can be assigned as an "intent". These
>     configuration parameters need to be distributed in the autonomic
>     domain to influence the detail configurations on each autonomic node.
>
>               As intent is described as an abstract, declarative,
>     high-level policy used to operate an autonomic domain, such as an
>     enterprise network. Perhaps we can rename the “intent” in the
>     “prefix-management ID” to some other words.
>
>     Is “ASA parameters" ok here?
>
>     I think some characteristics for this kind of low-level policies are
>
>     1.Network level parameters, need to be distributed by GRASP and ACP
>
>     2.Relatively static compared to intent, perhaps only needed to be
>     configured once
>
>     3.Unavoidable in current network environment, mostly for
>     establishing the network infrastructure
>
>     4.Rarely need coordinations with others parameters, usually is
>     closed related to one specific autonomic function
>
>     Welcome for comment. Your suggestions will be appreciated.
>
>     Best Regards
>
>     Zongpeng Du
>
>
>     _______________________________________________
>     Anima mailing list
>     Anima@ietf.org 
>     https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> --
>
> regards,
>
> John
>
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>


From nobody Wed Mar 23 12:11:23 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C7C6412D5A2 for ; Wed, 23 Mar 2016 12:11:21 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ixrRiuzSKxoq for ; Wed, 23 Mar 2016 12:11:18 -0700 (PDT)
Received: from alln-iport-6.cisco.com (alln-iport-6.cisco.com [173.37.142.93]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9642F12D52B for ; Wed, 23 Mar 2016 12:11:18 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=15241; q=dns/txt; s=iport; t=1458760278; x=1459969878; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=Qf3LTiWE0//q/UBhM5pNEpyBRAs6/ZLyElWG1szLisI=; b=b8emeeV+g6I7O6krKWKkhJCMTHWuSWNGyk6A4JCZ6Vz1OsXg7mUbiz+q ZkLoKHUn96m0ql7+IDucSHyx8YRB97BqfFFFs7xabp4S8+0BD7OlK1LhQ TlMXCVFIOgy2tiyWXb5Nnje0FM731pvQD2XpcNXvofMqLN7vbwYZN7xs6 4=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AJAgDf6fJW/51dJa1egzNTegavB4tOA?= =?us-ascii?q?Q2BcBcKhWwCgT44FAEBAQEBAQFkJ4RBAQEBBAEBAWsLDAQCAQgRAgIBAQEnByE?= =?us-ascii?q?GCxQJCAIEAQ0FCIgKAxIOvEQNhGIBAQEBAQEBAQEBAQEBAQEBAQEBAQEVimKCP?= =?us-ascii?q?odUBY1zhGyESjEBhXCGHoFujxGHModUAR4BAUKDZWqIUiUWAX0BAQE?=
X-IronPort-AV: E=Sophos;i="5.24,383,1454976000"; d="scan'208";a="252917085"
Received: from rcdn-core-6.cisco.com ([173.37.93.157]) by alln-iport-6.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 23 Mar 2016 19:11:16 +0000
Received: from XCH-RTP-006.cisco.com (xch-rtp-006.cisco.com [64.101.220.146]) by rcdn-core-6.cisco.com (8.14.5/8.14.5) with ESMTP id u2NJBGhT013900 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Wed, 23 Mar 2016 19:11:16 GMT
Received: from xch-rtp-001.cisco.com (64.101.220.141) by XCH-RTP-006.cisco.com (64.101.220.146) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Wed, 23 Mar 2016 15:11:15 -0400
Received: from xch-rtp-001.cisco.com ([64.101.220.141]) by XCH-RTP-001.cisco.com ([64.101.220.141]) with mapi id 15.00.1104.009; Wed, 23 Mar 2016 15:11:15 -0400
From: "Alexander Clemm (alex)" 
To: "Joel M. Halpern" , "Michael Behringer (mbehring)" , John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAVBrxYAAHjH7AAAF5yGgAApkmQAAB5UTAA==
Date: Wed, 23 Mar 2016 19:11:15 +0000
Message-ID: <2d7e20df6d6045389be8d58baca6dc6e@XCH-RTP-001.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com>
In-Reply-To: <56F2E3DA.9030204@joelhalpern.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [128.107.151.23]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , Sheng Jiang , "anima@ietf.org" , =?iso-8859-1?Q?J=E9ferson_Campos_Nobre?= 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 19:11:22 -0000

One question is how you would form a connected graph of objectives and how =
that would be resolved?  Isn't it simpler to simply assert for a given Auto=
nomic Domain, what its intent is?  If you choose to form multiple Autonomic=
 Domains, then so be it, but if you find yourself fine tuning / optimizing =
things too much, perhaps that is also an indication that you are going from=
 intent to something that is lower level.  Likewise for intent change rate.=
  If that rate becomes too rapid, it may be an indication as well that you =
go from intent and "managing by objectives" into traditional configuration =
and "micromanagement". =20
--- Alex

-----Original Message-----
From: Joel M. Halpern [mailto:jmh@joelhalpern.com]=20
Sent: Wednesday, March 23, 2016 11:44 AM
To: Alexander Clemm (alex) ; Michael Behringer (mbehring) <=
mbehring@cisco.com>; John Strassner 
Cc: Duzongpeng ; Laurent Ciavaglia ; J=E9ferson Campos Nobre ; anima@ietf.or=
g; Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond i=
ntent

I think the quesiton is what the scope of an intent is.
If, as seems likely, a given intent is for processing by a given objective,=
 and given that the objectives each form a connected graph over the domain,=
 it would seem sensible to flood the intent to the participants in the obje=
ctive.

If intents are not mapped (by the time they get into ANIMA) to a single obj=
ective, then it probably makes more sense to flood more generally.

To some degree this also depend on how often we expect intents to be introd=
uced, modified, or deleted.  If that rate is low, then flooding everywhere =
is probably cheap.  If that rate is high, I would want to be able to constr=
ain the propagation.

Yours,
Joel

On 3/23/16 1:56 PM, Alexander Clemm (alex) wrote:
> I really like your suggestion of flooding intent to all nodes (even if=20
> there may be optimizations later).  It provides a level of conceptual=20
> clarity.  Every node should be aware of overall intent (and need to=20
> decide for itself if there are actions for it to take).
>
> If there is some intent that applies only to some parts of the network=20
> and not others, that can be addressed by using different Autonomic=20
> Domains.  And, of course, Autonomic Domains may be associated with=20
> certain Autonomic Functions.  Intervention beyond intent should NOT be=20
> allowed for the Autonomic Function(s) being served an Autonomic=20
> Domain, but of course other interventions (for different functions,=20
> hence outside the scope) are still possible.
>
> --- Alex
>
> *From:*Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Michael=20
> Behringer (mbehring)
> *Sent:* Wednesday, March 23, 2016 3:57 AM
> *To:* John Strassner 
> *Cc:* Duzongpeng ; Laurent Ciavaglia=20
> ; Sheng Jiang ;=20
> anima@ietf.org; J=E9ferson Campos Nobre 
> *Subject:* Re: [Anima] Does Autonomic network need some intervention=20
> beyond intent
>
> John,
>
> I think we're on the same page. Couple of points, top posting here,=20
> otherwise it gets too messy.
>
> While *CIEs don't like magic, they do understand the concept of a=20
> default behaviour which they can influence if needed. Any reasonable=20
> network management approach tries to define "default" behaviour and=20
> "exceptions". The AN story fills the "default" part very nicely, and=20
> *CIEs get that.
>
> The "parameter" versus "policy" discussion: My view is that we=20
> shouldn't try to define exactly **what** intent allows and doesn't=20
> allow, since we've heard from many sides that the boundary is not=20
> clear, and it may be impossible to have a really clear, usable=20
> definition. So my point
> was: Let's focus on the distribution model: Intent is flooded to all=20
> nodes. Let's NOT do selective flooding, unicast Intent to specific=20
> nodes, etc, because that would blur the boundary even worse.
>
> My suggestion: Intent is flooded to all nodes in a domain. The ASA=20
> owners define the content of what they want to push in their bit of=20
> the Intent. If we insist of full flooding, we encourage the right behavio=
ur.
> And if someone really puts into his Intent "node A: do this; node B:=20
> do the other; node C: do the third" it is awkward enough for people to=20
> take a step back and think whether they're doing the right thing J
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
> Not sure I understand your concern. Yes, in my model there is a=20
> section for the fabric, and sections for the various functions. Seems=20
> a natural structure?!?
>
> "Function owner": Yes, I'm suggesting to structure Intent around=20
> Autonomic Functions. This is not optimal from a scientific point of=20
> view, but aligns very well with the way how things are organised.
> Probably we need to discuss that in BA. But refer to my YANG model=20
> sent a few days ago.
>
> We should schedule some discussion time outside the ANIMA meeting window.=
..
>
> Michael
>
> *From:*John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* 22 March 2016 21:32
> *To:* Michael Behringer (mbehring)  >; John Strassner  >
> *Cc:* Duzongpeng  >; anima@ietf.org=20
> ; Laurent Ciavaglia=20
> >;
> J=E9ferson Campos Nobre  >; Sheng Jiang  >
> *Subject:* Re: [Anima] Does Autonomic network need some intervention=20
> beyond intent
>
> In general, I agree with Michael. I'd like to emphasize and expand on=20
> a couple of key points:
>
> First, ANIMA isn't the only thing "influencing" or "steering" a=20
> network, at least not initially. There is still traditional config,=20
> plus all the existing and emerging SDN models. My view has always been=20
> that ANIMA sets a "default" policy in a network, and all other means=20
> are more specific and can override intent. This is described in=20
> RFC7575. Main point here: ANIMA Intent isn't the only tool in a network.
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
> Intent could perfectly well cover a high level policy such as "all=20
> nodes of type x do this; type y does that". (Which I think is what you=20
> mention
> below.) When we started with Intent, the vision was to keep it very=20
> high level, more along the line "do the right thing" J
>
> 
>
> Agreed. Although to me, this means that we need to consider the types=20
> of actors that are using intent. For example, most of my CCIE and JCIE=20
> friends have no interest in using an intent language for pushing=20
> configuration. On the other hand, when I talk to operators and=20
> application developers, they would love to be able to use a high-level=20
> intent and have the intent engine take care of the details for them.
>
> 
>
> It has become clear that in today's networking, people will want to=20
> push more "parameters" than "policy", and I think we'll just have to=20
> acknowledge that.
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
> Personally I think it'll pan out like this: Intent will have sections=20
> per autonomic function. An Intent parser will validate overall=20
> correctness (signature, etc), then segment the Intent into its bits:=20
> One part will be the autonomic fabric itself, then there will be bits=20
> for each autonomic function.
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
> So the Intent parser on a node will grab intent, see there is some=20
> Intent for autonomic function A, will see whether this is running=20
> locally, and if so, it'll just pass that piece of Intent to that function=
.
>
> 
>
> Agreed.
>
> 
>
> In this way, a function owner defines both the intent itself, and how=20
> it is consumed by his/her function.
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many=20
> or
>
> what types of autonomic functions are present.
>
> 
>
> The downside is that there is no "control" and people can do what they=20
> want, in the worst case even push config, which is clearly NOT what=20
> Intent is meant to do.
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
> I'd hope that people don't start writing Intent like "node 17: here is=20
> your CLI; node 23: here is your CLI"; node 37: here is your CLI". But=20
> what I describe above COULD be abused in this way. RFC7575 says this=20
> is NOT Intent.
>
> 
>
> Absolutely agree!
>
> 
>
>   regards,
>
> John
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring)=20
> > wrote:
>
>     Hi Zongpeng,
>
>     Yes, this is a recurring question and topic. Let me give my view on
>     two levels:
>
>     First, ANIMA isn't the only thing "influencing" or "steering" a
>     network, at least not initially. There is still traditional config,
>     plus all the existing and emerging SDN models. My view has always
>     been that ANIMA sets a "default" policy in a network, and all other
>     means are more specific and can override intent. This is described
>     in RFC7575. Main point here: ANIMA Intent isn't the only tool in a
>     network.
>
>     Intent could perfectly well cover a high level policy such as "all
>     nodes of type x do this; type y does that". (Which I think is what
>     you mention below.) When we started with Intent, the vision was to
>     keep it very high level, more along the line "do the right thing" J
>     It has become clear that in today's networking, people will want to
>     push more "parameters" than "policy", and I think we'll just have to
>     acknowledge that.
>
>     Personally I think it'll pan out like this: Intent will have
>     sections per autonomic function. An Intent parser will validate
>     overall correctness (signature, etc), then segment the Intent into
>     its bits: One part will be the autonomic fabric itself, then there
>     will be bits for each autonomic function.
>
>     So the Intent parser on a node will grab intent, see there is some
>     Intent for autonomic function A, will see whether this is running
>     locally, and if so, it'll just pass that piece of Intent to that
>     function.
>
>     In this way, a function owner defines both the intent itself, and
>     how it is consumed by his/her function.
>
>     The downside is that there is no "control" and people can do what
>     they want, in the worst case even push config, which is clearly NOT
>     what Intent is meant to do. I'd hope that people don't start writing
>     Intent like "node 17: here is your CLI; node 23: here is your CLI";
>     node 37: here is your CLI". But what I describe above COULD be
>     abused in this way. RFC7575 says this is NOT Intent.
>
>     In other words: Let's not be too prescriptive, and hope that folks
>     will do more or less reasonable things.
>
>     I've been meaning to comment on the Intent drafts and possibly
>     contribute, I'm just struggling with priorities at the moment (as
>     you can see by my relative silence over the past weeks).
>
>     Michael
>
>     *From:*Anima [mailto:anima-bounces@ietf.org
>     ] *On Behalf Of *Duzongpeng
>     *Sent:* 16 March 2016 02:38
>     *To:* anima@ietf.org 
>     *Cc:* Laurent Ciavaglia      >; J=E9ferson Campos Nobre
>     >; Sheng Jiang
>     >
>     *Subject:* [Anima] Does Autonomic network need some intervention
>     beyond intent
>
>     Hello, everyone
>
>               I understand that Autonomic network should be able to work
>     with as less configuration as possible. In the ideal case, an
>     Autonomic network should be able to work well while only intent is
>     needed from human operators.
>
>               However, I wonder whether Autonomic network may need some
>     intervention from human operators through some low-level policies
>     beyond intents.
>
>               For example, in the
>     https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,
>     it is suggested that the prefix lengths for the CSG, ASG, RSG
>     (different roles in IP RAN) can be assigned as an "intent". These
>     configuration parameters need to be distributed in the autonomic
>     domain to influence the detail configurations on each autonomic node.
>
>               As intent is described as an abstract, declarative,
>     high-level policy used to operate an autonomic domain, such as an
>     enterprise network. Perhaps we can rename the "intent" in the
>     "prefix-management ID" to some other words.
>
>     Is "ASA parameters" ok here?
>
>     I think some characteristics for this kind of low-level policies=20
> are
>
>     1.Network level parameters, need to be distributed by GRASP and=20
> ACP
>
>     2.Relatively static compared to intent, perhaps only needed to be
>     configured once
>
>     3.Unavoidable in current network environment, mostly for
>     establishing the network infrastructure
>
>     4.Rarely need coordinations with others parameters, usually is
>     closed related to one specific autonomic function
>
>     Welcome for comment. Your suggestions will be appreciated.
>
>     Best Regards
>
>     Zongpeng Du
>
>
>     _______________________________________________
>     Anima mailing list
>     Anima@ietf.org 
>     https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> --
>
> regards,
>
> John
>
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>


From nobody Wed Mar 23 12:27:43 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B6C5612D516 for ; Wed, 23 Mar 2016 12:27:42 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QmOn6x63UPyl for ; Wed, 23 Mar 2016 12:27:35 -0700 (PDT)
Received: from mail-pa0-x233.google.com (mail-pa0-x233.google.com [IPv6:2607:f8b0:400e:c03::233]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 11B4012D540 for ; Wed, 23 Mar 2016 12:27:32 -0700 (PDT)
Received: by mail-pa0-x233.google.com with SMTP id td3so2750354pab.2 for ; Wed, 23 Mar 2016 12:27:32 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=+oHUMNPjmsTfPUrngaMOwxdq84T3/cqHoI/iPRkhOyE=; b=eJusc9zszz244wUwZNd5nyYfJpbKBDpY7EgDzksKLuHfL4CO0zRZz4mL8eBz43VhmP Rm+dG9btgy/sgo4IcPxKi2Qt2C+EHCRtEVeV2Pw/AScu5Gv2Y3salga54gWGQ9THjB/E AcwpeCfGYh3BHohQwEnGM7Cx66UnXahj/TWjEg7fdWxfx7RuCZgBaG1M6dtZg6D6/Y6L Bef+3fARgROeU+a9CKdypSgH1pl5oHVbbOclJHN4bD2msK2Ef2VYpalB9qq5ai/AYaUG z/CRVh2gNP+OTOgOPk9uLUYB0khbRYb7NW8NYQRaoHmYYZ9q6ejnJ6Uscm8cg7mojt/V Q0Ag==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=+oHUMNPjmsTfPUrngaMOwxdq84T3/cqHoI/iPRkhOyE=; b=I+bt55orWfRDCB2rPVODKdHKJB84cehglSYGX+w/qQdwLj0s73yJezeTtgqDh+fdCO 4GMbhOyhnfkkNu/QHQ90ZURUrYTg/cUwTnQv5s2CrON9ie2MJuc0tXVpv+k/wPc+94Lw mHpnRnE4XBAZybYGb5AbmwG0D1KbhQNIoYd2deJwXw7FPVVYbhPpQgtiqT+XCeCLSuR3 HxdqiR5/whVwWbOmPK2HZHLVuvglNQfzOcnvF9V7gDaMuwnQLsYYeZYHdmC25pzKHAqn Sr+LW3XaAMoPoFLtC6MZ3dZVfLbAuwnRjypBRW7PhFMq71UMilCDu20JHyRHCuEZ73n+ HWmw==
X-Gm-Message-State: AD7BkJJsWUw0ShbE6/pcOnNv2gQbyvC+ZrrLDt5JlDEyUBrvkwXYHv5RcvbOoPjBSLYBaw==
X-Received: by 10.66.171.169 with SMTP id av9mr6734612pac.36.1458761251669; Wed, 23 Mar 2016 12:27:31 -0700 (PDT)
Received: from ?IPv6:2406:e007:4457:1:28cc:dc4c:9703:6781? ([2406:e007:4457:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id q85sm5931705pfq.81.2016.03.23.12.27.28 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 23 Mar 2016 12:27:30 -0700 (PDT)
To: Laurent Ciavaglia , "Liubing (Leo)" , Carsten Bormann 
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com> <56F0E9CF.7070602@tzi.org> <56F1969D.1060804@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5E464@nkgeml514-mbx.china.huawei.com> <56F259F6.5000807@alcatel-lucent.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F2EE26.9030107@gmail.com>
Date: Thu, 24 Mar 2016 08:27:34 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56F259F6.5000807@alcatel-lucent.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Michael Behringer , "anima@ietf.org" 
Subject: Re: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 19:27:42 -0000

On 23/03/2016 21:55, Laurent Ciavaglia wrote:
...
>> b)  Sophisticated operation on Intent, e.g. partial update, deleting, query etc.

I'm pretty sure we will need partial update and deletion. I don't expect we
will need query.

...
> As for RESTful "operations", this is a possible approach but we should probably first discuss the requirements before choosing a type of communication scheme (cf characteristics of autonomic networks).

I agree. RESTful has become a sort of religion recently but it's really just a
way of almost obtaining transactional integrity across proxies and caches.
It may not be what we need (for example, we might need semantics more like DNCP).
It may be that simple flooding (already supported by GRASP) will be sufficient,
if the update and deletion operations are atomic.

   Brian


From nobody Wed Mar 23 13:05:19 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4C98512D80E for ; Wed, 23 Mar 2016 13:05:18 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.702
X-Spam-Level: 
X-Spam-Status: No, score=-2.702 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HmABoTCIo5_T for ; Wed, 23 Mar 2016 13:05:16 -0700 (PDT)
Received: from maila2.tigertech.net (maila2.tigertech.net [208.80.4.152]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id EF0D312D85C for ; Wed, 23 Mar 2016 13:05:15 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by maila2.tigertech.net (Postfix) with ESMTP id D134E24EAC5; Wed, 23 Mar 2016 13:05:15 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1458763515; bh=I9tGHRm7kzM2f2JFZvYVFNHi7d1oK3x+ms8W5k4OiEk=; h=Subject:To:References:Cc:From:Date:In-Reply-To:From; b=cC7k5yq8YSbkw2o7YVHM8aAXINRCo6hJOsYxSGiaD7FE9olFlyfGOgr2r0Qe4pr9R lSGNWKIYcRDPqe6+40ZFdekjTJAFdQcXfMhGmTUsw6M99CHkO8/hlHjk6sS335F/q0 UTJDcJCGPg33aNGi3Bfrxj83MQo3YOP37jtmFglU=
X-Virus-Scanned: Debian amavisd-new at maila2.tigertech.net
Received: from Joels-MacBook-Pro.local (unknown [88.131.67.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by maila2.tigertech.net (Postfix) with ESMTPSA id 929872412BF; Wed, 23 Mar 2016 13:05:13 -0700 (PDT)
To: "Alexander Clemm (alex)" , "Michael Behringer (mbehring)" , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <2d7e20df6d6045389be8d58baca6dc6e@XCH-RTP-001.cisco.com>
From: "Joel M. Halpern" 
Message-ID: <56F2F6F6.2060607@joelhalpern.com>
Date: Wed, 23 Mar 2016 16:05:10 -0400
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <2d7e20df6d6045389be8d58baca6dc6e@XCH-RTP-001.cisco.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 8bit
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , Sheng Jiang , "anima@ietf.org" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 20:05:18 -0000

As I read it, GRASP already forms that connected network of ASAs which 
support a given objective.  Maybe I missed something in the design of 
the discovery mechanism, but I thought that was the point of it.

Yours,
Joel

On 3/23/16 3:11 PM, Alexander Clemm (alex) wrote:
> One question is how you would form a connected graph of objectives and how that would be resolved?  Isn't it simpler to simply assert for a given Autonomic Domain, what its intent is?  If you choose to form multiple Autonomic Domains, then so be it, but if you find yourself fine tuning / optimizing things too much, perhaps that is also an indication that you are going from intent to something that is lower level.  Likewise for intent change rate.  If that rate becomes too rapid, it may be an indication as well that you go from intent and "managing by objectives" into traditional configuration and "micromanagement".
> --- Alex
>
> -----Original Message-----
> From: Joel M. Halpern [mailto:jmh@joelhalpern.com]
> Sent: Wednesday, March 23, 2016 11:44 AM
> To: Alexander Clemm (alex) ; Michael Behringer (mbehring) ; John Strassner 
> Cc: Duzongpeng ; Laurent Ciavaglia ; Jéferson Campos Nobre ; anima@ietf.org; Sheng Jiang 
> Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
>
> I think the quesiton is what the scope of an intent is.
> If, as seems likely, a given intent is for processing by a given objective, and given that the objectives each form a connected graph over the domain, it would seem sensible to flood the intent to the participants in the objective.
>
> If intents are not mapped (by the time they get into ANIMA) to a single objective, then it probably makes more sense to flood more generally.
>
> To some degree this also depend on how often we expect intents to be introduced, modified, or deleted.  If that rate is low, then flooding everywhere is probably cheap.  If that rate is high, I would want to be able to constrain the propagation.
>
> Yours,
> Joel
>
> On 3/23/16 1:56 PM, Alexander Clemm (alex) wrote:
>> I really like your suggestion of flooding intent to all nodes (even if
>> there may be optimizations later).  It provides a level of conceptual
>> clarity.  Every node should be aware of overall intent (and need to
>> decide for itself if there are actions for it to take).
>>
>> If there is some intent that applies only to some parts of the network
>> and not others, that can be addressed by using different Autonomic
>> Domains.  And, of course, Autonomic Domains may be associated with
>> certain Autonomic Functions.  Intervention beyond intent should NOT be
>> allowed for the Autonomic Function(s) being served an Autonomic
>> Domain, but of course other interventions (for different functions,
>> hence outside the scope) are still possible.
>>
>> --- Alex
>>
>> *From:*Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Michael
>> Behringer (mbehring)
>> *Sent:* Wednesday, March 23, 2016 3:57 AM
>> *To:* John Strassner 
>> *Cc:* Duzongpeng ; Laurent Ciavaglia
>> ; Sheng Jiang ;
>> anima@ietf.org; Jéferson Campos Nobre 
>> *Subject:* Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>
>> John,
>>
>> I think we're on the same page. Couple of points, top posting here,
>> otherwise it gets too messy.
>>
>> While *CIEs don't like magic, they do understand the concept of a
>> default behaviour which they can influence if needed. Any reasonable
>> network management approach tries to define "default" behaviour and
>> "exceptions". The AN story fills the "default" part very nicely, and
>> *CIEs get that.
>>
>> The "parameter" versus "policy" discussion: My view is that we
>> shouldn't try to define exactly **what** intent allows and doesn't
>> allow, since we've heard from many sides that the boundary is not
>> clear, and it may be impossible to have a really clear, usable
>> definition. So my point
>> was: Let's focus on the distribution model: Intent is flooded to all
>> nodes. Let's NOT do selective flooding, unicast Intent to specific
>> nodes, etc, because that would blur the boundary even worse.
>>
>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>> owners define the content of what they want to push in their bit of
>> the Intent. If we insist of full flooding, we encourage the right behaviour.
>> And if someone really puts into his Intent "node A: do this; node B:
>> do the other; node C: do the third" it is awkward enough for people to
>> take a step back and think whether they're doing the right thing J
>>
>> 
>>
>> Do you mean that the intent writer will **know the difference**
>>
>> between the fabric and each autonomic function? I hope not,
>>
>> but perhaps I'm misunderstanding you...
>>
>> 
>>
>> Not sure I understand your concern. Yes, in my model there is a
>> section for the fabric, and sections for the various functions. Seems
>> a natural structure?!?
>>
>> "Function owner": Yes, I'm suggesting to structure Intent around
>> Autonomic Functions. This is not optimal from a scientific point of
>> view, but aligns very well with the way how things are organised.
>> Probably we need to discuss that in BA. But refer to my YANG model
>> sent a few days ago.
>>
>> We should schedule some discussion time outside the ANIMA meeting window...
>>
>> Michael
>>
>> *From:*John Strassner [mailto:strazpdj@gmail.com]
>> *Sent:* 22 March 2016 21:32
>> *To:* Michael Behringer (mbehring) > >; John Strassner > >
>> *Cc:* Duzongpeng > >; anima@ietf.org
>> ; Laurent Ciavaglia
>> >;
>> Jéferson Campos Nobre > >; Sheng Jiang > >
>> *Subject:* Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>
>> In general, I agree with Michael. I'd like to emphasize and expand on
>> a couple of key points:
>>
>> First, ANIMA isn't the only thing "influencing" or "steering" a
>> network, at least not initially. There is still traditional config,
>> plus all the existing and emerging SDN models. My view has always been
>> that ANIMA sets a "default" policy in a network, and all other means
>> are more specific and can override intent. This is described in
>> RFC7575. Main point here: ANIMA Intent isn't the only tool in a network.
>>
>> 
>>
>> Absolutely agree. Note that even in the Congress case I mentioned
>>
>> earlier, it is NOT pushing config, and it is NOT viewed as the
>>
>> "only" policy
>>
>> 
>>
>> Intent could perfectly well cover a high level policy such as "all
>> nodes of type x do this; type y does that". (Which I think is what you
>> mention
>> below.) When we started with Intent, the vision was to keep it very
>> high level, more along the line "do the right thing" J
>>
>> 
>>
>> Agreed. Although to me, this means that we need to consider the types
>> of actors that are using intent. For example, most of my CCIE and JCIE
>> friends have no interest in using an intent language for pushing
>> configuration. On the other hand, when I talk to operators and
>> application developers, they would love to be able to use a high-level
>> intent and have the intent engine take care of the details for them.
>>
>> 
>>
>> It has become clear that in today's networking, people will want to
>> push more "parameters" than "policy", and I think we'll just have to
>> acknowledge that.
>>
>> 
>>
>> Sorry Michael, I don't follow here. What do you mean
>>
>> by "parameters"?
>>
>> 
>>
>> Personally I think it'll pan out like this: Intent will have sections
>> per autonomic function. An Intent parser will validate overall
>> correctness (signature, etc), then segment the Intent into its bits:
>> One part will be the autonomic fabric itself, then there will be bits
>> for each autonomic function.
>>
>> 
>>
>> Do you mean that the intent writer will **know the difference**
>>
>> between the fabric and each autonomic function? I hope not,
>>
>> but perhaps I'm misunderstanding you...
>>
>> 
>>
>> So the Intent parser on a node will grab intent, see there is some
>> Intent for autonomic function A, will see whether this is running
>> locally, and if so, it'll just pass that piece of Intent to that function.
>>
>> 
>>
>> Agreed.
>>
>> 
>>
>> In this way, a function owner defines both the intent itself, and how
>> it is consumed by his/her function.
>>
>> 
>>
>> I'm not sure what you mean by "function owner". (Sorry to be
>>
>> picky, I'm likely under-caffeinated.)
>>
>> The above seemed like an autonomous decision, made by the parser.
>>
>> I don't think that the intent writer should be cognizant of how many
>> or
>>
>> what types of autonomic functions are present.
>>
>> 
>>
>> The downside is that there is no "control" and people can do what they
>> want, in the worst case even push config, which is clearly NOT what
>> Intent is meant to do.
>>
>> 
>>
>> And hence, the "HAL 2000" scenario. :-)
>>
>> 
>>
>> I'd hope that people don't start writing Intent like "node 17: here is
>> your CLI; node 23: here is your CLI"; node 37: here is your CLI". But
>> what I describe above COULD be abused in this way. RFC7575 says this
>> is NOT Intent.
>>
>> 
>>
>> Absolutely agree!
>>
>> 
>>
>>    regards,
>>
>> John
>>
>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring)
>> > wrote:
>>
>>      Hi Zongpeng,
>>
>>      Yes, this is a recurring question and topic. Let me give my view on
>>      two levels:
>>
>>      First, ANIMA isn't the only thing "influencing" or "steering" a
>>      network, at least not initially. There is still traditional config,
>>      plus all the existing and emerging SDN models. My view has always
>>      been that ANIMA sets a "default" policy in a network, and all other
>>      means are more specific and can override intent. This is described
>>      in RFC7575. Main point here: ANIMA Intent isn't the only tool in a
>>      network.
>>
>>      Intent could perfectly well cover a high level policy such as "all
>>      nodes of type x do this; type y does that". (Which I think is what
>>      you mention below.) When we started with Intent, the vision was to
>>      keep it very high level, more along the line "do the right thing" J
>>      It has become clear that in today's networking, people will want to
>>      push more "parameters" than "policy", and I think we'll just have to
>>      acknowledge that.
>>
>>      Personally I think it'll pan out like this: Intent will have
>>      sections per autonomic function. An Intent parser will validate
>>      overall correctness (signature, etc), then segment the Intent into
>>      its bits: One part will be the autonomic fabric itself, then there
>>      will be bits for each autonomic function.
>>
>>      So the Intent parser on a node will grab intent, see there is some
>>      Intent for autonomic function A, will see whether this is running
>>      locally, and if so, it'll just pass that piece of Intent to that
>>      function.
>>
>>      In this way, a function owner defines both the intent itself, and
>>      how it is consumed by his/her function.
>>
>>      The downside is that there is no "control" and people can do what
>>      they want, in the worst case even push config, which is clearly NOT
>>      what Intent is meant to do. I'd hope that people don't start writing
>>      Intent like "node 17: here is your CLI; node 23: here is your CLI";
>>      node 37: here is your CLI". But what I describe above COULD be
>>      abused in this way. RFC7575 says this is NOT Intent.
>>
>>      In other words: Let's not be too prescriptive, and hope that folks
>>      will do more or less reasonable things.
>>
>>      I've been meaning to comment on the Intent drafts and possibly
>>      contribute, I'm just struggling with priorities at the moment (as
>>      you can see by my relative silence over the past weeks).
>>
>>      Michael
>>
>>      *From:*Anima [mailto:anima-bounces@ietf.org
>>      ] *On Behalf Of *Duzongpeng
>>      *Sent:* 16 March 2016 02:38
>>      *To:* anima@ietf.org 
>>      *Cc:* Laurent Ciavaglia >      >; Jéferson Campos Nobre
>>      >; Sheng Jiang
>>      >
>>      *Subject:* [Anima] Does Autonomic network need some intervention
>>      beyond intent
>>
>>      Hello, everyone
>>
>>                I understand that Autonomic network should be able to work
>>      with as less configuration as possible. In the ideal case, an
>>      Autonomic network should be able to work well while only intent is
>>      needed from human operators.
>>
>>                However, I wonder whether Autonomic network may need some
>>      intervention from human operators through some low-level policies
>>      beyond intents.
>>
>>                For example, in the
>>      https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,
>>      it is suggested that the prefix lengths for the CSG, ASG, RSG
>>      (different roles in IP RAN) can be assigned as an "intent". These
>>      configuration parameters need to be distributed in the autonomic
>>      domain to influence the detail configurations on each autonomic node.
>>
>>                As intent is described as an abstract, declarative,
>>      high-level policy used to operate an autonomic domain, such as an
>>      enterprise network. Perhaps we can rename the "intent" in the
>>      "prefix-management ID" to some other words.
>>
>>      Is "ASA parameters" ok here?
>>
>>      I think some characteristics for this kind of low-level policies
>> are
>>
>>      1.Network level parameters, need to be distributed by GRASP and
>> ACP
>>
>>      2.Relatively static compared to intent, perhaps only needed to be
>>      configured once
>>
>>      3.Unavoidable in current network environment, mostly for
>>      establishing the network infrastructure
>>
>>      4.Rarely need coordinations with others parameters, usually is
>>      closed related to one specific autonomic function
>>
>>      Welcome for comment. Your suggestions will be appreciated.
>>
>>      Best Regards
>>
>>      Zongpeng Du
>>
>>
>>      _______________________________________________
>>      Anima mailing list
>>      Anima@ietf.org 
>>      https://www.ietf.org/mailman/listinfo/anima
>>
>>
>>
>>
>> --
>>
>> regards,
>>
>> John
>>
>>
>>
>> _______________________________________________
>> Anima mailing list
>> Anima@ietf.org
>> https://www.ietf.org/mailman/listinfo/anima
>>
>


From nobody Wed Mar 23 17:38:28 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B0B8512D112 for ; Wed, 23 Mar 2016 17:38:25 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7DKQbNrAAXz5 for ; Wed, 23 Mar 2016 17:38:22 -0700 (PDT)
Received: from mail-pa0-x22d.google.com (mail-pa0-x22d.google.com [IPv6:2607:f8b0:400e:c03::22d]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9FD8512D0FD for ; Wed, 23 Mar 2016 17:38:22 -0700 (PDT)
Received: by mail-pa0-x22d.google.com with SMTP id td3so7986729pab.2 for ; Wed, 23 Mar 2016 17:38:22 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=9735BglwxSyLJLHiBTqXebqP9cwy5osbT2ssOa5/GUo=; b=TX5RprokslJW0K+zAZ+gTwo3KUUKn+HMm8UY6vgVJFctSTDHW1gTL1KUDe1/jQtZuo mAQFhZ3VeZb/D+7fcbUEPQUtvBzXJaMJJvjeBTorNDXZn6vEKp66I9d47szyLSnapt7s bARhgCyJqXH9RmDv94RLwgJ423uzvcbGXWkqSicsTdPgAHZHudc1L7RY7h18By5Ifbp+ R+hB/UH8WuYU3HzJ1RtdGG1UdL3077Tn2IRJJgmwq0iVGoVf8YV7Tv5fUEngblVw2+Zz Pu+u2W3IAjLWUziTp5eBMHQQxIeeQw9MttgybK1gJv9thDi80Lfvbc/YgQzdD6RkmE6Z cZzA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=9735BglwxSyLJLHiBTqXebqP9cwy5osbT2ssOa5/GUo=; b=GBl54uggTvOeHYsQ2TwBQSN3aJKmcRAj1mesxE5s6TaE0F/336XuidwWmWvN62NgJF zEa1JNbGbnnTxi0Ys0dWE1G6f+mYR3FUfahIPGZTodjNLiLQ9WOjv+Gz9K25ghczFtqM gvDPbJfhzYI4CbsGpqAXd5Rehum3Hs90X7BwJj17pOFzBD0Maovt2iF/Ap9cLOZcADiR ITriNkHfd1UNSkCpo++CVqYlEt8HZHKIxuIgPX17UsYF7cyWxrs6LHafXaS4EAkhBsdC yaZOdX1je24OYb8VvDfiTolQ3jv/J6V0t5qHG0tVMCOcGxDaaLziMnhuHkX44QyIfijd deRw==
X-Gm-Message-State: AD7BkJJdfTE0QuVxbAs0Oiyph7snfz7+XkAOHOv3mcrplzBmB3xtEWxvBF4yKb6kEuaJ4Q==
X-Received: by 10.66.55.6 with SMTP id n6mr8540761pap.35.1458779902130; Wed, 23 Mar 2016 17:38:22 -0700 (PDT)
Received: from ?IPv6:2406:e007:4457:1:28cc:dc4c:9703:6781? ([2406:e007:4457:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id tb10sm6737838pab.22.2016.03.23.17.38.16 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 23 Mar 2016 17:38:20 -0700 (PDT)
To: "Joel M. Halpern" , "Alexander Clemm (alex)" , "Michael Behringer (mbehring)" , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <2d7e20df6d6045389be8d58baca6dc6e@XCH-RTP-001.cisco.com> <56F2F6F6.2060607@joelhalpern.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F336FF.6020900@gmail.com>
Date: Thu, 24 Mar 2016 13:38:23 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56F2F6F6.2060607@joelhalpern.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 00:38:25 -0000

On 24/03/2016 09:05, Joel M. Halpern wrote:
> As I read it, GRASP already forms that connected network of ASAs which =
support a given objective.  Maybe I missed something in
> the design of the discovery mechanism, but I thought that was the point=
 of it.

True, but the knowledge is distributed across multiple discovery caches. =
We'd have to add
a protocol feature to provide the equivalent of pub/sub, but I don't thin=
k it would
be hard to do. (There's a model sketched in draft-liu-anima-grasp-distrib=
ution,
which I haven't thought about much yet.)

   Brian

> Yours,
> Joel
>=20
> On 3/23/16 3:11 PM, Alexander Clemm (alex) wrote:
>> One question is how you would form a connected graph of objectives and=
 how that would be resolved?  Isn't it simpler to simply
>> assert for a given Autonomic Domain, what its intent is?  If you choos=
e to form multiple Autonomic Domains, then so be it, but
>> if you find yourself fine tuning / optimizing things too much, perhaps=
 that is also an indication that you are going from
>> intent to something that is lower level.  Likewise for intent change r=
ate.  If that rate becomes too rapid, it may be an
>> indication as well that you go from intent and "managing by objectives=
" into traditional configuration and "micromanagement".
>> --- Alex
>>
>> -----Original Message-----
>> From: Joel M. Halpern [mailto:jmh@joelhalpern.com]
>> Sent: Wednesday, March 23, 2016 11:44 AM
>> To: Alexander Clemm (alex) ; Michael Behringer (mbehri=
ng) ; John Strassner
>> 
>> Cc: Duzongpeng ; Laurent Ciavaglia ; J=C3=A9ferson Campos Nobre
>> ; anima@ietf.org; Sheng Jiang 
>> Subject: Re: [Anima] Does Autonomic network need some intervention bey=
ond intent
>>
>> I think the quesiton is what the scope of an intent is.
>> If, as seems likely, a given intent is for processing by a given objec=
tive, and given that the objectives each form a
>> connected graph over the domain, it would seem sensible to flood the i=
ntent to the participants in the objective.
>>
>> If intents are not mapped (by the time they get into ANIMA) to a singl=
e objective, then it probably makes more sense to flood
>> more generally.
>>
>> To some degree this also depend on how often we expect intents to be i=
ntroduced, modified, or deleted.  If that rate is low,
>> then flooding everywhere is probably cheap.  If that rate is high, I w=
ould want to be able to constrain the propagation.
>>
>> Yours,
>> Joel
>>
>> On 3/23/16 1:56 PM, Alexander Clemm (alex) wrote:
>>> I really like your suggestion of flooding intent to all nodes (even i=
f
>>> there may be optimizations later).  It provides a level of conceptual=

>>> clarity.  Every node should be aware of overall intent (and need to
>>> decide for itself if there are actions for it to take).
>>>
>>> If there is some intent that applies only to some parts of the networ=
k
>>> and not others, that can be addressed by using different Autonomic
>>> Domains.  And, of course, Autonomic Domains may be associated with
>>> certain Autonomic Functions.  Intervention beyond intent should NOT b=
e
>>> allowed for the Autonomic Function(s) being served an Autonomic
>>> Domain, but of course other interventions (for different functions,
>>> hence outside the scope) are still possible.
>>>
>>> --- Alex
>>>
>>> *From:*Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Michael
>>> Behringer (mbehring)
>>> *Sent:* Wednesday, March 23, 2016 3:57 AM
>>> *To:* John Strassner 
>>> *Cc:* Duzongpeng ; Laurent Ciavaglia
>>> ; Sheng Jiang ;
>>> anima@ietf.org; J=C3=A9ferson Campos Nobre 
>>> *Subject:* Re: [Anima] Does Autonomic network need some intervention
>>> beyond intent
>>>
>>> John,
>>>
>>> I think we're on the same page. Couple of points, top posting here,
>>> otherwise it gets too messy.
>>>
>>> While *CIEs don't like magic, they do understand the concept of a
>>> default behaviour which they can influence if needed. Any reasonable
>>> network management approach tries to define "default" behaviour and
>>> "exceptions". The AN story fills the "default" part very nicely, and
>>> *CIEs get that.
>>>
>>> The "parameter" versus "policy" discussion: My view is that we
>>> shouldn't try to define exactly **what** intent allows and doesn't
>>> allow, since we've heard from many sides that the boundary is not
>>> clear, and it may be impossible to have a really clear, usable
>>> definition. So my point
>>> was: Let's focus on the distribution model: Intent is flooded to all
>>> nodes. Let's NOT do selective flooding, unicast Intent to specific
>>> nodes, etc, because that would blur the boundary even worse.
>>>
>>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>>> owners define the content of what they want to push in their bit of
>>> the Intent. If we insist of full flooding, we encourage the right beh=
aviour.
>>> And if someone really puts into his Intent "node A: do this; node B:
>>> do the other; node C: do the third" it is awkward enough for people t=
o
>>> take a step back and think whether they're doing the right thing J
>>>
>>> 
>>>
>>> Do you mean that the intent writer will **know the difference**
>>>
>>> between the fabric and each autonomic function? I hope not,
>>>
>>> but perhaps I'm misunderstanding you...
>>>
>>> 
>>>
>>> Not sure I understand your concern. Yes, in my model there is a
>>> section for the fabric, and sections for the various functions. Seems=

>>> a natural structure?!?
>>>
>>> "Function owner": Yes, I'm suggesting to structure Intent around
>>> Autonomic Functions. This is not optimal from a scientific point of
>>> view, but aligns very well with the way how things are organised.
>>> Probably we need to discuss that in BA. But refer to my YANG model
>>> sent a few days ago.
>>>
>>> We should schedule some discussion time outside the ANIMA meeting win=
dow...
>>>
>>> Michael
>>>
>>> *From:*John Strassner [mailto:strazpdj@gmail.com]
>>> *Sent:* 22 March 2016 21:32
>>> *To:* Michael Behringer (mbehring) >> >; John Strassner >> >
>>> *Cc:* Duzongpeng >> >; anima@ietf.org
>>> ; Laurent Ciavaglia
>>> >;
>>> J=C3=A9ferson Campos Nobre >> >; Sheng Jiang >> >
>>> *Subject:* Re: [Anima] Does Autonomic network need some intervention
>>> beyond intent
>>>
>>> In general, I agree with Michael. I'd like to emphasize and expand on=

>>> a couple of key points:
>>>
>>> First, ANIMA isn't the only thing "influencing" or "steering" a
>>> network, at least not initially. There is still traditional config,
>>> plus all the existing and emerging SDN models. My view has always bee=
n
>>> that ANIMA sets a "default" policy in a network, and all other means
>>> are more specific and can override intent. This is described in
>>> RFC7575. Main point here: ANIMA Intent isn't the only tool in a netwo=
rk.
>>>
>>> 
>>>
>>> Absolutely agree. Note that even in the Congress case I mentioned
>>>
>>> earlier, it is NOT pushing config, and it is NOT viewed as the
>>>
>>> "only" policy
>>>
>>> 
>>>
>>> Intent could perfectly well cover a high level policy such as "all
>>> nodes of type x do this; type y does that". (Which I think is what yo=
u
>>> mention
>>> below.) When we started with Intent, the vision was to keep it very
>>> high level, more along the line "do the right thing" J
>>>
>>> 
>>>
>>> Agreed. Although to me, this means that we need to consider the types=

>>> of actors that are using intent. For example, most of my CCIE and JCI=
E
>>> friends have no interest in using an intent language for pushing
>>> configuration. On the other hand, when I talk to operators and
>>> application developers, they would love to be able to use a high-leve=
l
>>> intent and have the intent engine take care of the details for them.
>>>
>>> 
>>>
>>> It has become clear that in today's networking, people will want to
>>> push more "parameters" than "policy", and I think we'll just have to
>>> acknowledge that.
>>>
>>> 
>>>
>>> Sorry Michael, I don't follow here. What do you mean
>>>
>>> by "parameters"?
>>>
>>> 
>>>
>>> Personally I think it'll pan out like this: Intent will have sections=

>>> per autonomic function. An Intent parser will validate overall
>>> correctness (signature, etc), then segment the Intent into its bits:
>>> One part will be the autonomic fabric itself, then there will be bits=

>>> for each autonomic function.
>>>
>>> 
>>>
>>> Do you mean that the intent writer will **know the difference**
>>>
>>> between the fabric and each autonomic function? I hope not,
>>>
>>> but perhaps I'm misunderstanding you...
>>>
>>> 
>>>
>>> So the Intent parser on a node will grab intent, see there is some
>>> Intent for autonomic function A, will see whether this is running
>>> locally, and if so, it'll just pass that piece of Intent to that func=
tion.
>>>
>>> 
>>>
>>> Agreed.
>>>
>>> 
>>>
>>> In this way, a function owner defines both the intent itself, and how=

>>> it is consumed by his/her function.
>>>
>>> 
>>>
>>> I'm not sure what you mean by "function owner". (Sorry to be
>>>
>>> picky, I'm likely under-caffeinated.)
>>>
>>> The above seemed like an autonomous decision, made by the parser.
>>>
>>> I don't think that the intent writer should be cognizant of how many
>>> or
>>>
>>> what types of autonomic functions are present.
>>>
>>> 
>>>
>>> The downside is that there is no "control" and people can do what the=
y
>>> want, in the worst case even push config, which is clearly NOT what
>>> Intent is meant to do.
>>>
>>> 
>>>
>>> And hence, the "HAL 2000" scenario. :-)
>>>
>>> 
>>>
>>> I'd hope that people don't start writing Intent like "node 17: here i=
s
>>> your CLI; node 23: here is your CLI"; node 37: here is your CLI". But=

>>> what I describe above COULD be abused in this way. RFC7575 says this
>>> is NOT Intent.
>>>
>>> 
>>>
>>> Absolutely agree!
>>>
>>> 
>>>
>>>    regards,
>>>
>>> John
>>>
>>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring)
>>> > wrote:
>>>
>>>      Hi Zongpeng,
>>>
>>>      Yes, this is a recurring question and topic. Let me give my view=
 on
>>>      two levels:
>>>
>>>      First, ANIMA isn't the only thing "influencing" or "steering" a
>>>      network, at least not initially. There is still traditional conf=
ig,
>>>      plus all the existing and emerging SDN models. My view has alway=
s
>>>      been that ANIMA sets a "default" policy in a network, and all ot=
her
>>>      means are more specific and can override intent. This is describ=
ed
>>>      in RFC7575. Main point here: ANIMA Intent isn't the only tool in=
 a
>>>      network.
>>>
>>>      Intent could perfectly well cover a high level policy such as "a=
ll
>>>      nodes of type x do this; type y does that". (Which I think is wh=
at
>>>      you mention below.) When we started with Intent, the vision was =
to
>>>      keep it very high level, more along the line "do the right thing=
" J
>>>      It has become clear that in today's networking, people will want=
 to
>>>      push more "parameters" than "policy", and I think we'll just hav=
e to
>>>      acknowledge that.
>>>
>>>      Personally I think it'll pan out like this: Intent will have
>>>      sections per autonomic function. An Intent parser will validate
>>>      overall correctness (signature, etc), then segment the Intent in=
to
>>>      its bits: One part will be the autonomic fabric itself, then the=
re
>>>      will be bits for each autonomic function.
>>>
>>>      So the Intent parser on a node will grab intent, see there is so=
me
>>>      Intent for autonomic function A, will see whether this is runnin=
g
>>>      locally, and if so, it'll just pass that piece of Intent to that=

>>>      function.
>>>
>>>      In this way, a function owner defines both the intent itself, an=
d
>>>      how it is consumed by his/her function.
>>>
>>>      The downside is that there is no "control" and people can do wha=
t
>>>      they want, in the worst case even push config, which is clearly =
NOT
>>>      what Intent is meant to do. I'd hope that people don't start wri=
ting
>>>      Intent like "node 17: here is your CLI; node 23: here is your CL=
I";
>>>      node 37: here is your CLI". But what I describe above COULD be
>>>      abused in this way. RFC7575 says this is NOT Intent.
>>>
>>>      In other words: Let's not be too prescriptive, and hope that fol=
ks
>>>      will do more or less reasonable things.
>>>
>>>      I've been meaning to comment on the Intent drafts and possibly
>>>      contribute, I'm just struggling with priorities at the moment (a=
s
>>>      you can see by my relative silence over the past weeks).
>>>
>>>      Michael
>>>
>>>      *From:*Anima [mailto:anima-bounces@ietf.org
>>>      ] *On Behalf Of *Duzongpeng
>>>      *Sent:* 16 March 2016 02:38
>>>      *To:* anima@ietf.org 
>>>      *Cc:* Laurent Ciavaglia >>      >; J=C3=A9ferson Campos Nobr=
e
>>>      >; Sheng Jian=
g
>>>      >
>>>      *Subject:* [Anima] Does Autonomic network need some intervention=

>>>      beyond intent
>>>
>>>      Hello, everyone
>>>
>>>                I understand that Autonomic network should be able to =
work
>>>      with as less configuration as possible. In the ideal case, an
>>>      Autonomic network should be able to work well while only intent =
is
>>>      needed from human operators.
>>>
>>>                However, I wonder whether Autonomic network may need s=
ome
>>>      intervention from human operators through some low-level policie=
s
>>>      beyond intents.
>>>
>>>                For example, in the
>>>      https://tools.ietf.org/html/draft-ietf-anima-prefix-management-0=
0,
>>>      it is suggested that the prefix lengths for the CSG, ASG, RSG
>>>      (different roles in IP RAN) can be assigned as an "intent". Thes=
e
>>>      configuration parameters need to be distributed in the autonomic=

>>>      domain to influence the detail configurations on each autonomic =
node.
>>>
>>>                As intent is described as an abstract, declarative,
>>>      high-level policy used to operate an autonomic domain, such as a=
n
>>>      enterprise network. Perhaps we can rename the "intent" in the
>>>      "prefix-management ID" to some other words.
>>>
>>>      Is "ASA parameters" ok here?
>>>
>>>      I think some characteristics for this kind of low-level policies=

>>> are
>>>
>>>      1.Network level parameters, need to be distributed by GRASP and
>>> ACP
>>>
>>>      2.Relatively static compared to intent, perhaps only needed to b=
e
>>>      configured once
>>>
>>>      3.Unavoidable in current network environment, mostly for
>>>      establishing the network infrastructure
>>>
>>>      4.Rarely need coordinations with others parameters, usually is
>>>      closed related to one specific autonomic function
>>>
>>>      Welcome for comment. Your suggestions will be appreciated.
>>>
>>>      Best Regards
>>>
>>>      Zongpeng Du
>>>
>>>
>>>      _______________________________________________
>>>      Anima mailing list
>>>      Anima@ietf.org 
>>>      https://www.ietf.org/mailman/listinfo/anima
>>>
>>>
>>>
>>>
>>> --=20
>>>
>>> regards,
>>>
>>> John
>>>
>>>
>>>
>>> _______________________________________________
>>> Anima mailing list
>>> Anima@ietf.org
>>> https://www.ietf.org/mailman/listinfo/anima
>>>
>>
>=20
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>=20


From nobody Wed Mar 23 17:43:35 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1E69912D91C for ; Wed, 23 Mar 2016 17:43:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wjtcvnIS0Z-j for ; Wed, 23 Mar 2016 17:43:32 -0700 (PDT)
Received: from mail-pf0-x230.google.com (mail-pf0-x230.google.com [IPv6:2607:f8b0:400e:c00::230]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4C63F12D8D1 for ; Wed, 23 Mar 2016 17:43:32 -0700 (PDT)
Received: by mail-pf0-x230.google.com with SMTP id 4so43467285pfd.0 for ; Wed, 23 Mar 2016 17:43:32 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=ADor/xBbXJWt2hmM3c78ZRXsZGjZAwX00BwdI1frevA=; b=vNnyUCTUpOG7ATbLGvBb6J3XyLmM8nzDumiPuIG0VQtWYCr/6Ulj/NPsjctdhV0+Vp vj8GOcGncMXVQ2cU2tLzXvxHk3B6tv+smdGe9xtGiOjYztI3DZuKvqFdCs/n86Fpv/LW fA3NneOTogqIELIRPqmXFsI0+UH81aTDu8S58IUiW6jLsZCeB/2goLddpTlmaT+qw6R0 tNTr+qQFCGGGg/XnKPjTRTVRGS/WRuFXnJyhFsUcKdd9BYmbHUWtD2G8TsypWkp35HjL jR8NUUuAJy4afOCIbv94rjSbop+GgtKfU3pDK6YEMGvDsGzJM89t+C5iTBohU9gjrLNk FZFg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=ADor/xBbXJWt2hmM3c78ZRXsZGjZAwX00BwdI1frevA=; b=AXNIpBSAVtzPn0583mas/jo9mwEbETkJYEE5AamOwxqrcLTZvXmmyU5sD+64gtYqfV GtMdmsth0xPy1zZvXyVE97fLuCslxBLfz/avlmWYf/5TgIafXmy2cGMWEdGXLGPFRCdy xpjxF3j2gNv9mSkjRXa00Y6//0mKCzxrNJdxUArbmCMojKn0x+2qwrK8hx8aKv0GBGBg E6Elaj4jrohZCdlnkmOpW7YonOPNnbkVPuPOVWIMO76yD5OGHtJ476dfbLX1CsqQn95C yGz4pLcJEKjVkZsKxY8jqbeJFQhLLuCp+UjfXM9jUy7XXATjuqJ7fBngpwtSl8MTCebv xM4A==
X-Gm-Message-State: AD7BkJJ5nd1UkBPszTEk8EqEEovOmXDFOJNOwFHR7xlM5sun7qEwb16OpGzkZr7akTlT8g==
X-Received: by 10.98.10.136 with SMTP id 8mr8482662pfk.67.1458780211894; Wed, 23 Mar 2016 17:43:31 -0700 (PDT)
Received: from ?IPv6:2406:e007:4457:1:28cc:dc4c:9703:6781? ([2406:e007:4457:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id e12sm6752042pat.19.2016.03.23.17.43.27 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 23 Mar 2016 17:43:30 -0700 (PDT)
To: =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , Laurent Ciavaglia , "Liubing (Leo)" , Carsten Bormann 
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com> <56F0E9CF.7070602@tzi.org> <56F1969D.1060804@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5E464@nkgeml514-mbx.china.huawei.com> <56F259F6.5000807@alcatel-lucent.com> <56F2EE26.9030107@gmail.com> 
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F33837.4020001@gmail.com>
Date: Thu, 24 Mar 2016 13:43:35 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: "anima@ietf.org" , Michael Behringer 
Subject: Re: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 00:43:34 -0000

On 24/03/2016 08:48, J=C3=A9ferson Campos Nobre wrote:
> Em qua, 23 de mar de 2016 =C3=A0s 16:27, Brian E Carpenter <
> brian.e.carpenter@gmail.com> escreveu:
>=20
>> On 23/03/2016 21:55, Laurent Ciavaglia wrote:
>> ...
>>>> b)  Sophisticated operation on Intent, e.g. partial update, deleting=
,
>> query etc.
>>
>> I'm pretty sure we will need partial update and deletion. I don't expe=
ct we
>> will need query.
>>
>=20
> Probably, query would be needed for (self?) monitoring and intent
> verification and consistency checks.

Actually the query in that case could just be a Request Synch for
the relevant subset of intent. The objective "Intent.PrefixManager" is th=
e
example used in draft-ietf-anima-prefix-management.

    Brian
>=20
>=20
>>
>> ...
>>> As for RESTful "operations", this is a possible approach but we shoul=
d
>> probably first discuss the requirements before choosing a type of
>> communication scheme (cf characteristics of autonomic networks).
>>
>> I agree. RESTful has become a sort of religion recently but it's reall=
y
>> just a
>> way of almost obtaining transactional integrity across proxies and cac=
hes.
>> It may not be what we need (for example, we might need semantics more =
like
>> DNCP).
>> It may be that simple flooding (already supported by GRASP) will be
>> sufficient,
>> if the update and deletion operations are atomic.
>>
>>
> + 1
>=20
>=20
>>    Brian
>>
>> _______________________________________________
>> Anima mailing list
>> Anima@ietf.org
>> https://www.ietf.org/mailman/listinfo/anima
>>
>=20


From nobody Wed Mar 23 19:41:40 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3E4CB12D13D for ; Wed, 23 Mar 2016 19:41:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XWUcYAKLlMsH for ; Wed, 23 Mar 2016 19:41:33 -0700 (PDT)
Received: from mail-lb0-x236.google.com (mail-lb0-x236.google.com [IPv6:2a00:1450:4010:c04::236]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 12CC212D0C4 for ; Wed, 23 Mar 2016 19:41:32 -0700 (PDT)
Received: by mail-lb0-x236.google.com with SMTP id bc4so21845698lbc.2 for ; Wed, 23 Mar 2016 19:41:31 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=Wo9TWfHBvwEqxtKW0W/4ezNbT3sR7E5pI+fZYU7Og1k=; b=WxXFUjSPoFDfXe7TcEq+56SYwCJ8a2NtDaStfmIEauidQ7OGtl3zvgBnKvc0eEKqyc nZmHBK1nDFxudOcP3FiKAgAcko0UjtfNchgwCkLpn/5fm8sLHYNibngDty/SxG3R/fn+ fpYfugOEA8VypnkVk5XCVrJmY8iZigcRbvkZmFngiebFOmRTY2TOI7nq8NdZcc5NWK9p QDxZBMTwptuDhfuWCC4yodE39ADiKA7WnB5J3aN6uyhrRkfSdfFVvkCVfRVqdKR+4vGg W+tZ5T7fQZUd7eL5MfzGoKUSYWuUgbOcFtuzt6DSZz7xPyPSXRj6Kmr5qD+vcmQZkLps Lfng==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=Wo9TWfHBvwEqxtKW0W/4ezNbT3sR7E5pI+fZYU7Og1k=; b=aMw7ZHTDPebKUKlpnfNjBYjcCXxY0wbt50o5DWd59pw5hqXiql1dTsKirtZN7n058x YSe+dlhaKmJe+cX0jC8TllShGK9bGBCwdLW61hlg7sMHU5kYrknyYeNT9agUKWR3prv0 E+eONUlWNGcLJISULMY11xqEEe7ENcFz4VI0n1BdGEWRX18nu72P+6GzZc7tgNM/wJCC nx7pqWTk+56pizBUqqeaBL1PoS82SB4NeXGmmGH4pt1Vz1IWxkYjz4C2ma+YJxhQ/ZR8 TUMtxoyGjrC3kiQ3q8maGeBp8fATviY87d+cyBo8CNiQyhf8OoG5k8LSy3kr6pCbd1oO SovQ==
X-Gm-Message-State: AD7BkJKNiBEdNb/+tUfqZn2dF/tGHUWMUectmLrzCpLqI36WA0pug4euUVZhT5yqHJzERh/O8JFwNBtzpWoPtQ==
MIME-Version: 1.0
X-Received: by 10.112.162.231 with SMTP id yd7mr2469267lbb.40.1458787290055; Wed, 23 Mar 2016 19:41:30 -0700 (PDT)
Received: by 10.25.32.9 with HTTP; Wed, 23 Mar 2016 19:41:29 -0700 (PDT)
In-Reply-To: <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>
Date: Wed, 23 Mar 2016 19:41:29 -0700
Message-ID: 
From: John Strassner 
To: "Michael Behringer (mbehring)" , John Strassner 
Content-Type: multipart/alternative; boundary=089e011608e03a0bd2052ec262bc
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , Sheng Jiang , "anima@ietf.org" , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 02:41:37 -0000

--089e011608e03a0bd2052ec262bc
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Michael,



thanks for your reply. I do think that we largely agree, and would enjoy
chatting during or (hopefully before) BA. Please see ... Also
trying to unclutter, so I am eliding points that we agree on.





While *CIEs don=E2=80=99t like magic, they do understand the concept of a d=
efault
behaviour which they can influence if needed. Any reasonable network
management approach tries to define =E2=80=9Cdefault=E2=80=9D behaviour and=
 =E2=80=9Cexceptions=E2=80=9D.
The AN story fills the =E2=80=9Cdefault=E2=80=9D part very nicely, and *CIE=
s get that.



Agreed!





The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussion:=
 My view is that we shouldn=E2=80=99t
try to define exactly **what** intent allows and doesn=E2=80=99t allow, sin=
ce we=E2=80=99ve
heard from many sides that the boundary is not clear, and it may be
impossible to have a really clear, usable definition. So my point was:
Let=E2=80=99s focus on the distribution model: Intent is flooded to all nod=
es.
Let=E2=80=99s NOT do selective flooding, unicast Intent to specific nodes, =
etc,
because that would blur the boundary even worse.

I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1 was
2006, almost 10 years ago), we used a message bus and pub-sub. In v2, we
flooded; in v3, we went back to pub-sub because we wanted to do some
optimization and auditing. But I would be happy to try some type of
controlled flooding again.


My suggestion: Intent is flooded to all nodes in a domain. The ASA owners
define the content of what they want to push in their bit of the Intent. If
we insist of full flooding, we encourage the right behaviour. And if
someone really puts into his Intent =E2=80=9Cnode A: do this; node B: do th=
e other;
node C: do the third=E2=80=9D it is awkward enough for people to take a ste=
p back
and think whether they=E2=80=99re doing the right thing J

I agree in principle with what you said. Plus, see above.
< /jcs2>


Do you mean that the intent writer will **know the difference**
between the fabric and each autonomic function? I hope not,
but perhaps I'm misunderstanding you...
< /jcs>
Not sure I understand your concern. Yes, in my model there is a section for
the fabric, and sections for the various functions. Seems a natural
structure?!? 
The intent **writer** would have to know implementation
details in order to place everything in a single file. I would
rather not require this, and instead, have the intent engine
(e.g., compiler and associated logic) do this.
< /jcs2>

=E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structure =
Intent around Autonomic
Functions. This is not optimal from a scientific point of view, but aligns
very well with the way how things are organised. Probably we need to
discuss that in BA. But refer to my YANG model sent a few days ago.

I would suggest a slight modification, to enable users of
intent (and even intent writers) to NOT have to have
detailed implementation information.

What if all of the Autonomic Functions, after they joined
the network, advertised both their capabilities as well as
their needs? We can put aside the latter (needs) as this
is more complex, but if the capabilities are advertised,
then the autonomic management system can assemble
those needs into a catalog. Then, the intent writer can
look at the catalog and write a declarative request to
use the functionality of the catalog.

I will look at your YANG model in a bit (sorry, I'm woefully
behind...).
< /jcs2>





regards,

John

On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <
mbehring@cisco.com> wrote:

> John,
>
>
>
> I think we=E2=80=99re on the same page. Couple of points, top posting her=
e,
> otherwise it gets too messy.
>
>
>
> While *CIEs don=E2=80=99t like magic, they do understand the concept of a=
 default
> behaviour which they can influence if needed. Any reasonable network
> management approach tries to define =E2=80=9Cdefault=E2=80=9D behaviour a=
nd =E2=80=9Cexceptions=E2=80=9D.
> The AN story fills the =E2=80=9Cdefault=E2=80=9D part very nicely, and *C=
IEs get that.
>
>
>
> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussio=
n: My view is that we shouldn=E2=80=99t
> try to define exactly **what** intent allows and doesn=E2=80=99t allow, s=
ince
> we=E2=80=99ve heard from many sides that the boundary is not clear, and i=
t may be
> impossible to have a really clear, usable definition. So my point was:
> Let=E2=80=99s focus on the distribution model: Intent is flooded to all n=
odes.
> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific nodes=
, etc,
> because that would blur the boundary even worse.
>
>
>
> My suggestion: Intent is flooded to all nodes in a domain. The ASA owners
> define the content of what they want to push in their bit of the Intent. =
If
> we insist of full flooding, we encourage the right behaviour. And if
> someone really puts into his Intent =E2=80=9Cnode A: do this; node B: do =
the other;
> node C: do the third=E2=80=9D it is awkward enough for people to take a s=
tep back
> and think whether they=E2=80=99re doing the right thing J
>
>
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
> Not sure I understand your concern. Yes, in my model there is a section
> for the fabric, and sections for the various functions. Seems a natural
> structure?!?
>
>
>
> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structur=
e Intent around Autonomic
> Functions. This is not optimal from a scientific point of view, but align=
s
> very well with the way how things are organised. Probably we need to
> discuss that in BA. But refer to my YANG model sent a few days ago.
>
>
>
> We should schedule some discussion time outside the ANIMA meeting window.=
..
>
>
>
> Michael
>
>
>
>
>
> *From:* John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* 22 March 2016 21:32
> *To:* Michael Behringer (mbehring) ; John Strassner <
> strazpdj@gmail.com>
> *Cc:* Duzongpeng ; anima@ietf.org; Laurent
> Ciavaglia ; J=C3=A9ferson Campos Nobre <
> jcnobre@inf.ufrgs.br>; Sheng Jiang 
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> In general, I agree with Michael. I'd like to emphasize and expand on a
> couple of key points:
>
>
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
>
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J
>
>
>
> 
>
> Agreed. Although to me, this means that we need to consider the types of
> actors that are using intent. For example, most of my CCIE and JCIE frien=
ds
> have no interest in using an intent language for pushing configuration. O=
n
> the other hand, when I talk to operators and application developers, they
> would love to be able to use a high-level intent and have the intent engi=
ne
> take care of the details for them.
>
> 
>
>
>
> It has become clear that in today=E2=80=99s networking, people will want =
to push
> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I th=
ink we=E2=80=99ll just have to acknowledge
> that.
>
>
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
>
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> 
>
> Agreed.
>
> 
>
>
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
>
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
>
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do.
>
>
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
>
>
>
>
> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=
=80=9Cnode 17: here is
> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CLI=
=E2=80=9D. But what
> I describe above COULD be abused in this way. RFC7575 says this is NOT
> Intent.
>
>
>
> 
>
> Absolutely agree!
>
> 
>
>
>
>  regards,
>
> John
>
>
>
>
>
>
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
> mbehring@cisco.com> wrote:
>
> Hi Zongpeng,
>
>
>
> Yes, this is a recurring question and topic. Let me give my view on two
> levels:
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J  It has=
 become clear
> that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writing =
Intent like =E2=80=9Cnode
> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here i=
s your
> CLI=E2=80=9D. But what I describe above COULD be abused in this way. RFC7=
575 says
> this is NOT Intent.
>
>
>
> In other words: Let=E2=80=99s not be too prescriptive, and hope that folk=
s will do
> more or less reasonable things.
>
>
>
> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly co=
ntribute,
> I=E2=80=99m just struggling with priorities at the moment (as you can see=
 by my
> relative silence over the past weeks).
>
>
>
> Michael
>
>
>
>
>
>
>
> *From:* Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Duzongpeng
> *Sent:* 16 March 2016 02:38
> *To:* anima@ietf.org
> *Cc:* Laurent Ciavaglia ; J=C3=A9ferson Camp=
os
> Nobre ; Sheng Jiang 
> *Subject:* [Anima] Does Autonomic network need some intervention beyond
> intent
>
>
>
> Hello, everyone
>
>
>
>          I understand that Autonomic network should be able to work with
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
>
>
>
>          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
>
>
>
>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
>
>
>
>          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
>
>
>
> Is =E2=80=9CASA parameters" ok here?
>
>
>
> I think some characteristics for this kind of low-level policies are
>
> 1.       Network level parameters, need to be distributed by GRASP and AC=
P
>
> 2.       Relatively static compared to intent, perhaps only needed to be
> configured once
>
> 3.       Unavoidable in current network environment, mostly for
> establishing the network infrastructure
>
> 4.       Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
>
>
>
> Welcome for comment. Your suggestions will be appreciated.
>
>
>
>
>
> Best Regards
>
> Zongpeng Du
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> --
>
> regards,
>
> John
>



--=20
regards,
John

--089e011608e03a0bd2052ec262bc
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi Michael,
=C2=A0
thanks for =
your reply. I do think that we largely agree, and would enjoy chatting duri=
ng or (hopefully before) BA. Please see <jcs2>..</jcs2>. Also t=
rying to unclutter, so I am eliding points that we agree on.<=
/p>
=C2=A0
=C2=A0
While *CIEs don=E2=80=99t like magic, they do understand the concept=
 of a default behaviour which they can influence if needed. Any reasonable =
network management approach tries to define =E2=80=9Cdefault=E2=80=9D behav=
iour and =E2=80=9Cexceptions=E2=80=9D.  The AN story fills the =E2=80=9Cdef=
ault=E2=80=9D part very nicely, and *CIEs get that.
<jcs2>
Agreed!
</jcs2>
=C2=A0
The =E2=80=9Cparameter=E2=80=9D versu=
s =E2=80=9Cpolicy=E2=80=9D discussion: My view is that we shouldn=E2=80=99t=
 try to define exactly *what* intent allows and doesn=E2=80=99t allo=
w, since we=E2=80=99ve heard from many sides that the boundary is not clear=
, and it may be impossible to have a really clear, usable definition. So my=
 point was: Let=E2=80=99s focus on the distribution model: Intent is floode=
d to all nodes. Let=E2=80=99s NOT do selective flooding, unicast Intent to =
specific nodes, etc, because that would blur the boundary even worse.
<jcs2>
 I like flooding intent to all nodes. In FOCALE (v1-3,=
 and remember, v1 was 2006, almost 10 years ago), we used a message bus and=
 pub-sub. In v2, we flooded; in v3, we went back to pub-sub because we want=
ed to do some optimization and auditing. But I would be happy to try some t=
ype of controlled flooding again.
</jcs2>

My suggestion:=
 Intent is flooded to all nodes in a domain. The ASA owners define the cont=
ent of what they want to push in their bit of the Intent. If we insist of f=
ull flooding, we encourage the right behaviour. And if someone really puts =
into his Intent =E2=80=9Cnode A: do this; node B: do the other; node C: do =
the third=E2=80=9D it is awkward enough for people to take a step back and =
think whether they=E2=80=99re doing the right thing J 
<jcs2>
 I agree in principle with what yo=
u said. Plus, see above.
< /jcs2>

<jcs>
 Do you=
 mean that the intent writer will **know the difference**
 between the f=
abric and each autonomic function? I hope not,
 but perhaps I'm misu=
nderstanding you...
< /jcs>
 Not sure I understand your concern=
. Yes, in my model there is a section for the fabric, and sections for the =
various functions. Seems a natural structure?!? <jcs2>
 The=
 intent **writer** would have to know implementation
 details in order t=
o place everything in a single file. I would
 rather not require this, a=
nd instead, have the intent engine
 (e.g., compiler and associated logic=
) do this.
< /jcs2>

=E2=80=9CFunction owner=E2=80=9D: =
Yes, I=E2=80=99m suggesting to structure Intent around Autonomic Functions.=
 This is not optimal from a scientific point of view, but aligns very well =
with the way how things are organised. Probably we need to discuss that in =
BA. But refer to my YANG model sent a few days ago. 
<jcs2>=

 I would suggest a slight modification, to enable users of
 intent (=
and even intent writers) to NOT have to have
 detailed implementation in=
formation.

 What if all of the Autonomic Functions, after they joine=
d
 the network, advertised both their capabilities as well as
 their =
needs? We can put aside the latter (needs) as this
 is more complex, but=
 if the capabilities are advertised,
 then the autonomic management syst=
em can assemble
 those needs into a catalog. Then, the intent writer can=

 look at the catalog and write a declarative request to
 use the fun=
ctionality of the catalog.

 I will look at your YANG model in a bit =
(sorry, I'm woefully
behind...).
< =
/jcs2>
=C2=A0
=C2=A0
rega=
rds,
John

On Wed, Mar 23, 2016 at 3:57 AM, =
Michael Behringer (mbehring) <mbehring@cisco.com> wrote:











=
John,



=
=C2=A0


=
I think we=E2=80=99re on the same page. Couple of point=
s, top posting here, otherwise it gets
 too messy.


=
=C2=A0


=
While *CIEs don=E2=80=99t like magic, they do understan=
d the concept of a default behaviour which
 they can influence if needed. Any reasonable network management approach t=
ries to define =E2=80=9Cdefault=E2=80=9D behaviour and =E2=80=9Cexceptions=
=E2=80=9D. The AN story fills the =E2=80=9Cdefault=E2=80=9D part very nicel=
y, and *CIEs get that.



=
=C2=A0


=
The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=
=E2=80=9D discussion: My view is that we shouldn=E2=80=99t try to define
 exactly *what* intent allow=
s and doesn=E2=80=99t allow, since we=E2=80=99ve heard from many sides that=
 the boundary is not clear, and it may be impossible to have a really clear=
, usable definition. So my point was: Let=E2=80=99s focus
 on the distribution model: Intent is flooded to all nodes. Let=E2=80=99s N=
OT do selective flooding, unicast Intent to specific nodes, etc, because th=
at would blur the boundary even worse.



=
=C2=A0


=
My suggestion: Intent is flooded to all nodes in a doma=
in. The ASA owners define the
 content of what they want to push in their bit of the Intent. If we insist=
 of full flooding, we encourage the right behaviour. And if someone really =
puts into his Intent =E2=80=9Cnode A: do this; node B: do the other; node C=
: do the third=E2=80=9D it is awkward enough for
 people to take a step back and think whether they=E2=80=99re doing the rig=
ht thing 
J=




=
=C2=A0


=
<jcs>


=
Do you mean that the intent writer will **know the diff=
erence**


=
between the fabric and each autonomic function? I hope =
not,


=
but perhaps I'm misunderstanding you...


=
</jcs>


=
=C2=A0


=
Not sure I understand your concern. Yes, in my model th=
ere is a section for the fabric,
 and sections for the various functions. Seems a natural structure?!? 


=
=C2=A0


=
=E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m sugg=
esting to structure Intent around Autonomic Functions.
 This is not optimal from a scientific point of view, but aligns very well =
with the way how things are organised. Probably we need to discuss that in =
BA. But refer to my YANG model sent a few days ago.



=
=C2=A0


=
We should schedule some discussion time outside the ANI=
MA meeting window...


=
=C2=A0


=
Michael


=
=C2=A0


=
=C2=A0








From:
 John Strassner [mailto:strazpdj@gmail.com] 

Sent: 22 March 2016 21:32
To: Michael Behringer (mbehr=
ing) <mbehring@c=
isco.com>; John Strassner <strazpdj@gmail.com>

Cc: Duzongpeng <duzongpeng@huawei.com>; anima@ietf.org<=
/a>; Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; J=C3=A9ferson Campos =
Nobre <jcnobre=
@inf.ufrgs.br>; Sheng Jiang <jiangsheng@huawei.com>

Subject: Re: [Anima] Does Au=
tonomic network need some intervention beyond intent






=C2=A0






In general, I agree with Michael. I'd like to emph=
asize and expand on a couple of key points:






=C2=A0






=C2=A0






=
First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfl=
uencing=E2=80=9D or =E2=80=9Csteering=E2=80=9D
 a network, at least not initially. There is still traditional config, plus=
 all the existing and emerging SDN models. My view has always been that ANI=
MA sets a =E2=80=9Cdefault=E2=80=9D policy in a network, and all other mean=
s are more specific and can override intent. This
 is described in RFC7575. Main point here: ANIMA Intent isn=E2=80=99t the o=
nly tool in a network.



=C2=A0






<jcs>






Absolutely agree. Note that even in the Congress case =
I mentioned






earlier, it is NOT pushing config, and it is NOT viewe=
d as the






"only" policy







</jcs>








=C2=A0






=C2=A0






=
Intent could perfectly well cover a high level policy
 such as =E2=80=9Call nodes of type x do this; type y does that=E2=80=9D. (=
Which I think is what you mention below.) When we started with Intent, the =
vision was to keep it very high level, more along the line =E2=80=9Cdo the =
right thing=E2=80=9D
J=
=C2=A0


=C2=A0








<jcs>






Agreed. Although to me, this means that we need to con=
sider the types of actors that are using intent. For example, most of my CC=
IE and JCIE friends have no interest in
 using an intent language for pushing configuration. On the other hand, whe=
n I talk to operators and application developers, they would love to be abl=
e to use a high-level intent and have the intent engine take care of the de=
tails for them.






</jcs>






=C2=A0










It has become c=
lear that in today=E2=80=99s networking, people will want to push more =E2=
=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=
=80=99ll just have to
 acknowledge that.






=C2=A0






<jcs>






Sorry Michael, I don't follow here. What do you me=
an






by "parameters"?=







</jcs>






=C2=A0








=C2=A0






=
Personally I think it=E2=80=99ll pan out like this: Int=
ent will
 have sections per autonomic function. An Intent parser will validate overa=
ll correctness (signature, etc), then segment the Intent into its bits: One=
 part will be the autonomic fabric itself, then there will be bits for each=
 autonomic function.



=C2=A0






<jcs>






Do you mean that the intent writer will **know the dif=
ference**






between the fabric and each autonomic function? I hope=
 not,






but perhaps I'm misunderstanding you...<=
/u>






</jcs>






=C2=A0






=C2=A0






=
So the Intent parser on a node will grab intent, see
 there is some Intent for autonomic function A, will see whether this is ru=
nning locally, and if so, it=E2=80=99ll just pass that piece of Intent to t=
hat function.









=C2=A0






<jcs>






Agreed.






</jcs>






=C2=A0








=C2=A0






=
In this way, a function owner defines both the intent
 itself, and how it is consumed by his/her function. <=
u>


=C2=A0






<jcs>






I'm not sure what you mean by "function owner=
". (Sorry to be






picky, I'm likely under-caffeinated.)






=C2=A0






The above seemed like an autonomous decision, made by =
the parser.






I don't think that the intent writer should be cog=
nizant of how many or






what types of autonomic functions are present.<=
u>






</jcs>






=C2=A0






=C2=A0








=
The downside is that there is no =E2=80=9Ccontrol=E2=80=
=9D and people
 can do what they want, in the worst case even push config, which is clearl=
y NOT what Intent is meant to do.



=C2=A0






<jcs>






And hence, the "HAL 2000" scenario. :-)






</jcs>






=C2=A0








=C2=A0






=
I=E2=80=99d hope that people don=E2=80=99t start writin=
g Intent like =E2=80=9Cnode 17: here is your CLI; node 23: here is your CLI=
=E2=80=9D; node
 37: here is your CLI=E2=80=9D. But what I describe above COULD be abused i=
n this way. RFC7575 says this is NOT Intent.







=C2=A0






<jcs>






Absolutely agree!






</jcs>






=C2=A0


















=
=C2=A0regards,


=
John


=C2=A0


=C2=A0








=C2=A0




On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mb=
ehring) <mbehrin=
g@cisco.com> wrote:








=
Hi Zongpeng,



=
=C2=A0


=
Yes, this is a recurring question and topic. Let me giv=
e
 my view on two levels: 


=
=C2=A0


=
First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfl=
uencing=E2=80=9D or =E2=80=9Csteering=E2=80=9D
 a network, at least not initially. There is still traditional config, plus=
 all the existing and emerging SDN models. My view has always been that ANI=
MA sets a =E2=80=9Cdefault=E2=80=9D policy in a network, and all other mean=
s are more specific and can override intent. This
 is described in RFC7575. Main point here: ANIMA Intent isn=E2=80=99t the o=
nly tool in a network.



=
=C2=A0


=
Intent could perfectly well cover a high level policy
 such as =E2=80=9Call nodes of type x do this; type y does that=E2=80=9D. (=
Which I think is what you mention below.) When we started with Intent, the =
vision was to keep it very high level, more along the line =E2=80=9Cdo the =
right thing=E2=80=9D
J=
=C2=A0 It has become clear that in today=E2=80=99s
 networking, people will want to push more =E2=80=9Cparameters=E2=80=9D tha=
n =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.



=
=C2=A0


=
Personally I think it=E2=80=99ll pan out like this: Int=
ent will
 have sections per autonomic function. An Intent parser will validate overa=
ll correctness (signature, etc), then segment the Intent into its bits: One=
 part will be the autonomic fabric itself, then there will be bits for each=
 autonomic function.



=
=C2=A0


=
So the Intent parser on a node will grab intent, see
 there is some Intent for autonomic function A, will see whether this is ru=
nning locally, and if so, it=E2=80=99ll just pass that piece of Intent to t=
hat function.



=
=C2=A0


=
In this way, a function owner defines both the intent
 itself, and how it is consumed by his/her function. <=
u>


=
=C2=A0


=
The downside is that there is no =E2=80=9Ccontrol=E2=80=
=9D and people
 can do what they want, in the worst case even push config, which is clearl=
y NOT what Intent is meant to do. I=E2=80=99d hope that people don=E2=80=99=
t start writing Intent like =E2=80=9Cnode 17: here is your CLI; node 23: he=
re is your CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D. But what I
 describe above COULD be abused in this way. RFC7575 says this is NOT Inten=
t. 



=
=C2=A0


=
In other words: Let=E2=80=99s not be too prescriptive, =
and hope
 that folks will do more or less reasonable things. 


=
=C2=A0


=
I=E2=80=99ve been meaning to comment on the Intent draf=
ts and
 possibly contribute, I=E2=80=99m just struggling with priorities at the mo=
ment (as you can see by my relative silence over the past weeks).



=
=C2=A0


=
Michael


=
=C2=A0


=
=C2=A0


=
=C2=A0








From:
 Anima [mailto:anima-bounces@ietf.org]
On Behalf Of Duzongpeng

Sent: 16 March 2016 02:38
To: anima@ietf.org

Cc: Laurent Ciavaglia <laurent.ci=
avaglia@nokia.com>;
 J=C3=A9ferson Campos Nobre <jcnobre@inf.ufrgs.br>; Sheng
 Jiang <jia=
ngsheng@huawei.com>

Subject: [Anima] Does Autono=
mic network need some intervention beyond intent






=C2=A0


He=
llo, everyone


=
=C2=A0


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 I understand that Autonomi=
c network should be able to work with
 as less configuration as possible. In the ideal case, an Autonomic network=
 should be able to work well while only intent is needed from human operato=
rs.


=
=C2=A0


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 However, I wonder whether =
Autonomic network may need some intervention
 from human operators through some low-level policies beyond intents.


=
=C2=A0


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 For example, in the
https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00=
,
 it is suggested that the prefix lengths for the CSG, ASG, RSG (different r=
oles in IP RAN) can be assigned as an "intent". These configurati=
on parameters need to be distributed in the autonomic domain to influence t=
he detail configurations on each autonomic
 node.


=
=C2=A0


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 As intent is described as =
an abstract, declarative, high-level
 policy used to operate an autonomic domain, such as an enterprise network.=
 Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix-=
management ID=E2=80=9D to some other words.


=
=C2=A0



Is =E2=80=9CASA parameter=
s" ok here?




=C2=A0



I think some characterist=
ics for this kind of low-level policies are


1.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Network level parameters, need to be dis=
tributed by GRASP and ACP


2.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Relatively static compared to intent, pe=
rhaps only needed to be configured once


3.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Unavoidable in current network environme=
nt, mostly for establishing the network infrastructure=



4.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Rarely need coordinations with others pa=
rameters, usually is closed related to one specific autonomic function


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0




Welcome for comment. Your=
 suggestions will be appreciated.



=C2=A0



=C2=A0


Be=
st Regards


Zo=
ngpeng Du










_______________________________________________

Anima mailing list

Anima@ietf.org

h=
ttps://www.ietf.org/mailman/listinfo/anima<=
/p>











-- 






regards,






John


















-- 
regards,
John




--089e011608e03a0bd2052ec262bc--


From nobody Wed Mar 23 20:03:52 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3668F12D5F6 for ; Wed, 23 Mar 2016 20:03:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rp9kMxTkEvbE for ; Wed, 23 Mar 2016 20:03:49 -0700 (PDT)
Received: from mail-pf0-x22e.google.com (mail-pf0-x22e.google.com [IPv6:2607:f8b0:400e:c00::22e]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 914ED12D0F1 for ; Wed, 23 Mar 2016 20:03:49 -0700 (PDT)
Received: by mail-pf0-x22e.google.com with SMTP id x3so45880599pfb.1 for ; Wed, 23 Mar 2016 20:03:49 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=HyU1YUbG6n7gaua7PvbfiIaK1+4Wz3XlSPWA9B8o56g=; b=v3LTTQrXrJo0D8bnpBhuA7x2cTuwK8VA+e3K37QSFgUgGhO9keI3yBRWJ9/Df8y1rN 7FeYG4aNOR+fqSm6VQXygJeME7I+AO7ugHQqJLYZBY2EwxX3ymiZ3WrG9DDn7fI+Dt7G KIjRMyUZ0KYf94uq0yAmGqs6M+yo2Y/Gp0r2N2It3ERgUNybjUgwrqJoNNiOBLq1h5yo t/7MKPqFFZIMVwJG2Te53+96i+ljs0UjS+nWihki1CIdPY8SX2pEDwjiUsBRx7sgDlPU 1X/CkUxCX2DMSWczoNPubTvMfGZLXsn63pBqz90pesYlAVSXcfAiqmDN4OAvhHaQbI3B ZSeA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=HyU1YUbG6n7gaua7PvbfiIaK1+4Wz3XlSPWA9B8o56g=; b=TsvvtblQA346/SwZFu4u3subVC4j/9dR41gCLtGCdsfRnlYTVExYlilq+DlLEPJg0S dkWXPS8tqR5w+rfhCvR8aRON3YJiMpSjiFF68FIvEdWT9U5jvI98+gCDS4/by4ubwcQO wDcOfGaF2v+Wu9ANN7mGo/iNmnFKaoJFpQvMKIbgIwvy89FGEuPmlUiSjUiQFBKZitUh HjffKxlLDhNCWmRhZaSLCIqDz9zOxIdBSE3tydAXVqBNoOIxYgsv+5B1v0gIczUuJISr Cg7zyJ3yHTwC7hO7vBlIi+ZqV0tlZmFEw3MDl2/Wkmqt0QOnz+pJhVUWhxJ91UHLsQhV v84g==
X-Gm-Message-State: AD7BkJLdPvTIOrkSIIoHDdiOUdLB4kcNxxf3q/OfGLqMLYcc0r4xR4dx48c2sc/eCXRO0w==
X-Received: by 10.98.87.5 with SMTP id l5mr9112175pfb.7.1458788628851; Wed, 23 Mar 2016 20:03:48 -0700 (PDT)
Received: from ?IPv6:2406:e007:4457:1:28cc:dc4c:9703:6781? ([2406:e007:4457:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id m84sm7121495pfi.75.2016.03.23.20.03.46 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 23 Mar 2016 20:03:47 -0700 (PDT)
To: Michael Richardson , "anima@ietf.org" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com> <485.1458564774@obiwan.sandelman.ca> <56F0BDB4.1040904@gmail.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F35918.4080701@gmail.com>
Date: Thu, 24 Mar 2016 16:03:52 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56F0BDB4.1040904@gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: [Anima] Who's a proxy? [was: need for some ASA definitions definitions to help bootstrap]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 03:03:51 -0000

>> These are things that *could* be determined by "power-on-above", i.e. a
>> controller in the NOC could push (netconf-style or otherweise) configuration
>> into a node to tell it, "You will be a bootstrap proxy on interfaces A,C and
>> Q, and the registrar is at address BAR"
> 
> That doesn't really seem right, because it assumes that the NOC has
> perfect knowledge of the topology. Potentially, any node with multiple
> link interfaces could be a proxy, and if the topology is non-redundant,
> they would all need to be proxies.

This has been niggling at me.

1. Does it matter if there is more than one proxy on a given LAN?

2. If it does matter, the algorithm seems to be something like:

   1. Do I have more than one LAN interface?
      If no, I cannot be a proxy; exit
   2. Is there an active proxy on any interface?
      If yes, exclude that interface for now.
   3. Can I discover the registrar via any interface?
      If no, I cannot be a proxy; exit
   4. Register myself as a proxy on the non-excluded interfaces.

I don't think we need any input from a higher power. We just need
the registrar to register itself with GRASP (so that it can
be discovered) and the proxy instance in each node to execute the
above algorithm once in a while.

Regards
   Brian


From nobody Wed Mar 23 20:22:43 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6A2A712D117 for ; Wed, 23 Mar 2016 20:22:42 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.718
X-Spam-Level: 
X-Spam-Status: No, score=-1.718 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_FONT_FACE_BAD=0.981, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iZclzG9aXR3S for ; Wed, 23 Mar 2016 20:22:37 -0700 (PDT)
Received: from mail-lb0-x232.google.com (mail-lb0-x232.google.com [IPv6:2a00:1450:4010:c04::232]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3299312D0ED for ; Wed, 23 Mar 2016 20:22:36 -0700 (PDT)
Received: by mail-lb0-x232.google.com with SMTP id bc4so22248807lbc.2 for ; Wed, 23 Mar 2016 20:22:36 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=GNSXdtjYgpBznTdAOolIR4G7Ln1sC8hZYI0ZaphxHFk=; b=aD+Tqid6fUz6Ax7JFLi0m8B8r83xMgs4oJpypShXclqQRVdncQ4g9wUCwcilKxRO1y ugjrsWOXnWEkLdijxkhT/VdIzjEq9l8QcP2s7H/CTyywkJrRKkaX5Z6mQ9MhrbLsaPES 9QrjjgIEHdtyKdJNahcZxuICbfF1E6jFHBJt0MB59DSrogh1HQwffvRnkTAlz4NRlPKY i8xf5k7f6KtRtLWTIeC89+OpcJb3S/FFgukSHyX6kz/K6+tmtpSNx8snnsdyDINDnWqt kYtfK0y7uv8AE2wE1DfACAGRobwKjDQaat/frQqFiTiUPwdF9vurykzPeNLGRDxHbgjV CJKg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=GNSXdtjYgpBznTdAOolIR4G7Ln1sC8hZYI0ZaphxHFk=; b=OYWdLoeINaijUrzXjAzSs4ar8Z5sWxHtE1Dgpia20CzceSsMbSpNuOjFuErbVZMIh8 7qzLk5jNw3V58+0ObD8xuWeMJ1TjETW5oSNd62InJi97+lkKIo6oFNJbxOYuu2TpN0az ZRyXYUCrCHE147wOOxHWrk4tDqfj1pRsgTQYjCQTmFGZUimOeLifFxP5x4EzheCdd8EC fJ+6PlkEzEa3CzWBC+20euUMoCXvuzM4qs9glb4V8G4odiITdHsjj+GtkQUfO0s90Ds+ UwHe0OalDTR4XG9eZSO29jWpoGZNPgPnPyJJTs4+NTVOtzQqqHnMA6o9JOJeHusSNPqQ aLuw==
X-Gm-Message-State: AD7BkJJa2uTVCinm2Qne+r5N81Xh7WjjJZBmRzP9jC2BqHMvlUoSg+Nxqd65kw2Xqr4sjQpNZVtMtKUO2+XyDg==
MIME-Version: 1.0
X-Received: by 10.112.52.230 with SMTP id w6mr2046012lbo.144.1458789754104; Wed, 23 Mar 2016 20:22:34 -0700 (PDT)
Received: by 10.25.16.160 with HTTP; Wed, 23 Mar 2016 20:22:33 -0700 (PDT)
In-Reply-To: <56F28403.6060401@alcatel-lucent.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <56F28403.6060401@alcatel-lucent.com>
Date: Wed, 23 Mar 2016 20:22:33 -0700
Message-ID: 
From: John Strassner 
To: Laurent Ciavaglia , John Strassner 
Content-Type: multipart/alternative; boundary=001a11c3eab418737b052ec2f5e8
Archived-At: 
Cc: Duzongpeng , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , "EXT Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 03:22:42 -0000

--001a11c3eab418737b052ec2f5e8
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Laurent,



Thanks for your note. I'd be happy to join in a virtual meeting. Please see
.. for my observations.





On 23/03/2016 11:57, EXT Michael Behringer (mbehring) wrote:

John,

*[stripped]*
The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussion:=
 My view is that we shouldn=E2=80=99t
try to define exactly **what** intent allows and doesn=E2=80=99t allow, sin=
ce we=E2=80=99ve
heard from many sides that the boundary is not clear, and it may be
impossible to have a really clear, usable definition. So my point was:
Let=E2=80=99s focus on the distribution model: Intent is flooded to all nod=
es.
Let=E2=80=99s NOT do selective flooding, unicast Intent to specific nodes, =
etc,
because that would blur the boundary even worse.
[LC]
Before stating that intent policies should be flooded to all nodes, can we
first compare/evaluate the different options?
-broadcast is simple...but how important is it compared to other
criteria...?
I see at least two issues:
1. flooding a "big" intent file (cf your current tentative example) may be
bandwidth hungry, if one considers how frequently such an intent file could
be distributed (ASA stop/start, device join/leave, policy change/update...)
compared to distributing only what is required / where it is required.


Hopefully, intent will not change frequently (or else it probably is too
low level in nature). I personally like using pub-sub, but that is because
it enables me to do additional pre- and post-processing of intent. However,
I think that this should be an option.
< /jcs>

2. security: flooding the entire file reveal information to target
(possibly undesirable/rogue) entities that do not need it (and thus may
mis-use it).

< jcs>
This is a good point! It definitely broadens the attack surface...
< /jcs>

More fundamentally, I think it is misleading to imagine that intent
policies will be contained in one "file" (it is possible to do so but
impractical/cumbersome to manage). cf also the next comment.
[/LC]

See my previous reply. I am more worried about the user model -
constructing a single file means that the constructor has to know
implementation details. I think that this is unreasonable for the intent
writer. However, I see no reason why the intent compiler could not produce
a single file, or sets of files that group similar intent (e.g., invoking
multiple autonomic functions in the same ASA) together.
< /jcs>



My suggestion: Intent is flooded to all nodes in a domain.

[LC]
It is the "resolution" of the intent clauses that determines the domain/set
of target "nodes".
So in that order, yes the intent will be flooded to all the relevant nodes
in the "intent policy domain". This domain may be all nodes in an ANIMA
network, or a sub-set, only one node, or zero. This is contextual.
There will be *many* intent policies for a given ANIMA network.
[/LC]



Let's use a concept of:



   +--------+     +----------+     +-----------+

   | Intent |---->|  Intent  |---->| Autonomic |

   | Writer |     | Compiler |     |  Network  |

   +--------+     +----------+     +-----------+



I **think** what Laurent is saying is that once the Intent Compiler
resolves the intent of the user/intent writer, the target of the intent
policy is then identified. I agree with this. Even for the broad case of
"intent fabric", I don't think that an intent writer need be conscious of
this (vs. an autonomic function). Why should he or she care?



Laurent, please correct me if I am wrong here.







The ASA owners define the content of what they want to push in their bit of
the Intent. If we insist of full flooding, we encourage the right
behaviour. And
if someone really puts into his Intent =E2=80=9Cnode A: do this; node B: do=
 the
other; node C: do the third=E2=80=9D it is awkward enough for people to tak=
e a step
back and think whether they=E2=80=99re doing the right thing J

[LC]
I think we have a disconnect on what an intent look like and how it can be
used.
In my "mental model", there is not one "big" intent with sub-parts for each
ASA, the ANI, etc...
Each ASA designer (owner) will define the content / specificity of its ASA
functionality or links to the existing elements of the ANIMA intent policy
model, but it does not mean that we must aggregate all parts into a single
"file".
What would be the reason to do that?

In addition, there will not only be intent policies per functions, or for
the ANI.
There will be intent policies conveying the Service level, the Business
level... these will not resolve/map 1-to-1 in a single ASA "policy".
[/LC]



I don't think that you and Michael are really disagreeing (other than one
implementation (the file), which could be one option of many).



Indeed, in FOCALE and other work I have done, once intent was compiled, we
did group things together for a number of reasons (e.g., dependencies). I
think that the Autonomic Functions should advertise their capabilities, or
else, how does the Intent Compiler have a clue as to what gets shipped to
what (not to mention, how does it translate intent into a form that is
consumable by ASAs). This is even more important for Services - one could
imagine that a single Service intent would need multiple Autonomic
Functions to be implemented.






jcs>
Do you mean that the intent writer will **know the difference**
between the fabric and each autonomic function? I hope not,
but perhaps I'm misunderstanding you...
< /jcs>

Not sure I understand your concern. Yes, in my model there is a section for
the fabric, and sections for the various functions. Seems a natural
structure?!?
[LC]
It is OK to have different intent-policies for the fabric and for the
functions, but why put them all into one file/policy?
[/LC]



See my previous reply. In short, I was worried about this requiring the
intent writer to be aware of which autonomic functions were present, and
the nature of the autonomic fabric. I don't care that much about the
structure of the input.






=E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structure =
Intent around Autonomic
Functions. This is not optimal from a scientific point of view, but aligns
very well with the way how things are organised. Probably we need to
discuss that in BA. But refer to my YANG model sent a few days ago.

[LC]
I fully agree on the focus per autonomic function, but it does not imply to
have one single intent policy with all parts inside.
[/LC]



My point is that the intent writer should not have to have detailed
knowledge about how the autonomic network works, or what its constituent
autonomic functions are. In my previous post, I talked about an "autonomic
catalog" that the intent writer could use. This would hide the fact that a
particular service in the autonomic function had a 1:1 or 1:n
correspondence to a set of autonomic functions.






We should schedule some discussion time outside the ANIMA meeting window...

[LC]
Yes. And a white board too!
Unfortunately, I cannot attend physically this time.
Can we schedule a face-to-face ad-hoc meeting?
[/LC]



Count me in.





regards,

John

On Wed, Mar 23, 2016 at 4:54 AM, Laurent Ciavaglia <
laurent.ciavaglia@nokia.com> wrote:

> Michael,
>
> There are some aspects I am not sure to fully understand / agree with
> (marked below with [LC]).
>
> Best regards, Laurent.
>
>
> On 23/03/2016 11:57, EXT Michael Behringer (mbehring) wrote:
>
> John,
> *[stripped]*
>
> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussio=
n: My view is that we shouldn=E2=80=99t
> try to define exactly **what** intent allows and doesn=E2=80=99t allow, s=
ince
> we=E2=80=99ve heard from many sides that the boundary is not clear, and i=
t may be
> impossible to have a really clear, usable definition. So my point was:
> Let=E2=80=99s focus on the distribution model: Intent is flooded to all n=
odes.
> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific nodes=
, etc,
> because that would blur the boundary even worse.
>
>
> [LC]
> Before stating that intent policies should be flooded to all nodes, can w=
e
> first compare/evaluate the different options?
> -broadcast is simple...but how important is it compared to other
> criteria...?
> I see at least two issues:
> 1. flooding a "big" intent file (cf your current tentative example) may b=
e
> bandwidth hungry, if one considers how frequently such an intent file cou=
ld
> be distributed (ASA stop/start, device join/leave, policy change/update..=
.)
> compared to distributing only what is required / where it is required.
> 2. security: flooding the entire file reveal information to target
> (possibly undesirable/rogue) entities that do not need it (and thus may
> mis-use it).
>
> More fundamentally, I think it is misleading to imagine that intent
> policies will be contained in one "file" (it is possible to do so but
> impractical/cumbersome to manage). cf also the next comment.
> [/LC]
>
> My suggestion: Intent is flooded to all nodes in a domain.
>
> [LC]
> It is the "resolution" of the intent clauses that determines the
> domain/set of target "nodes".
> So in that order, yes the intent will be flooded to all the relevant node=
s
> in the "intent policy domain". This domain may be all nodes in an ANIMA
> network, or a sub-set, only one node, or zero. This is contextual.
> There will be *many* intent policies for a given ANIMA network.
> [/LC]
>
> The ASA owners define the content of what they want to push in their bit
> of the Intent. If we insist of full flooding, we encourage the right
> behaviour. And if someone really puts into his Intent =E2=80=9Cnode A: do=
 this;
> node B: do the other; node C: do the third=E2=80=9D it is awkward enough =
for people
> to take a step back and think whether they=E2=80=99re doing the right thi=
ng J
>
> [LC]
> I think we have a disconnect on what an intent look like and how it can b=
e
> used.
> In my "mental model", there is not one "big" intent with sub-parts for
> each ASA, the ANI, etc...
> Each ASA designer (owner) will define the content / specificity of its AS=
A
> functionality or links to the existing elements of the ANIMA intent polic=
y
> model, but it does not mean that we must aggregate all parts into a singl=
e
> "file".
> What would be the reason to do that?
>
> In addition, there will not only be intent policies per functions, or for
> the ANI.
> There will be intent policies conveying the Service level, the Business
> level... these will not resolve/map 1-to-1 in a single ASA "policy".
> [/LC]
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
> Not sure I understand your concern. Yes, in my model there is a section
> for the fabric, and sections for the various functions. Seems a natural
> structure?!?
>
> [LC]
> It is OK to have different intent-policies for the fabric and for the
> functions, but why put them all into one file/policy?
> [/LC]
>
>
>
> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structur=
e Intent around Autonomic
> Functions. This is not optimal from a scientific point of view, but align=
s
> very well with the way how things are organised. Probably we need to
> discuss that in BA. But refer to my YANG model sent a few days ago.
>
> [LC]
> I fully agree on the focus per autonomic function, but it does not imply
> to have one single intent policy with all parts inside.
> [/LC]
>
>
>
> We should schedule some discussion time outside the ANIMA meeting window.=
..
>
>
>
> [LC]
> Yes. And a white board too!
> Unfortunately, I cannot attend physically this time.
> Can we schedule a face-to-face ad-hoc meeting?
> [/LC]
>
>
> Michael
>
>
>
>
>
> *From:* John Strassner [mailto:strazpdj@gmail.com ]
> *Sent:* 22 March 2016 21:32
> *To:* Michael Behringer (mbehring) 
> ; John Strassner 
> 
> *Cc:* Duzongpeng  ;
> anima@ietf.org; Laurent Ciavaglia 
> ; J=C3=A9ferson Campos Nobre
>  ; Sheng Jiang
>  
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> In general, I agree with Michael. I'd like to emphasize and expand on a
> couple of key points:
>
>
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
>
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J
>
>
>
> 
>
> Agreed. Although to me, this means that we need to consider the types of
> actors that are using intent. For example, most of my CCIE and JCIE frien=
ds
> have no interest in using an intent language for pushing configuration. O=
n
> the other hand, when I talk to operators and application developers, they
> would love to be able to use a high-level intent and have the intent engi=
ne
> take care of the details for them.
>
> 
>
>
>
> It has become clear that in today=E2=80=99s networking, people will want =
to push
> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I th=
ink we=E2=80=99ll just have to acknowledge
> that.
>
>
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
>
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> 
>
> Agreed.
>
> 
>
>
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
>
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
>
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do.
>
>
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
>
>
>
>
> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=
=80=9Cnode 17: here is
> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CLI=
=E2=80=9D. But what
> I describe above COULD be abused in this way. RFC7575 says this is NOT
> Intent.
>
>
>
> 
>
> Absolutely agree!
>
> 
>
>
>
>  regards,
>
> John
>
>
>
>
>
>
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
> mbehring@cisco.com> wrote:
>
> Hi Zongpeng,
>
>
>
> Yes, this is a recurring question and topic. Let me give my view on two
> levels:
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J  It has=
 become clear
> that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writing =
Intent like =E2=80=9Cnode
> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here i=
s your
> CLI=E2=80=9D. But what I describe above COULD be abused in this way. RFC7=
575 says
> this is NOT Intent.
>
>
>
> In other words: Let=E2=80=99s not be too prescriptive, and hope that folk=
s will do
> more or less reasonable things.
>
>
>
> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly co=
ntribute,
> I=E2=80=99m just struggling with priorities at the moment (as you can see=
 by my
> relative silence over the past weeks).
>
>
>
> Michael
>
>
>
>
>
>
>
> *From:* Anima [mailto: anima-bounces@ietf.org] *O=
n
> Behalf Of *Duzongpeng
> *Sent:* 16 March 2016 02:38
> *To:* anima@ietf.org
> *Cc:* Laurent Ciavaglia < 
> laurent.ciavaglia@nokia.com>; J=C3=A9ferson Campos Nobre <
> jcnobre@inf.ufrgs.br>; Sheng Jiang <
> jiangsheng@huawei.com>
> *Subject:* [Anima] Does Autonomic network need some intervention beyond
> intent
>
>
>
> Hello, everyone
>
>
>
>          I understand that Autonomic network should be able to work with
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
>
>
>
>          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
>
>
>
>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
>
>
>
>          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
>
>
>
> Is =E2=80=9CASA parameters" ok here?
>
>
>
> I think some characteristics for this kind of low-level policies are
>
> 1.       Network level parameters, need to be distributed by GRASP and AC=
P
>
> 2.       Relatively static compared to intent, perhaps only needed to be
> configured once
>
> 3.       Unavoidable in current network environment, mostly for
> establishing the network infrastructure
>
> 4.       Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
>
>
>
> Welcome for comment. Your suggestions will be appreciated.
>
>
>
>
>
> Best Regards
>
> Zongpeng Du
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> --
>
> regards,
>
> John
>
>
> --
>
> Laurent Ciavaglia
>
> Senior Research Manager
>
> Bell Labs, Nokia
>
>
>
> +33 160 402 636
>
> Route de Villejust | 91620 Nozay | France
>
> LinkedIn: laurentciavaglia 
>
>
>



--=20
regards,
John

--001a11c3eab418737b052ec2f5e8
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi Laurent,
=C2=A0
Thanks for =
your note. I'd be happy to join in a virtual meeting. Please see <jc=
s>..</jcs> for my observations.
=C2=A0
=C2=A0
On 23/03/2016 11=
:57, EXT Michael Behringer (mbehring) wrote:
John, 
[stripped]=

The =E2=80=9Cparam=
eter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussion: My view is that w=
e shouldn=E2=80=99t try to define exactly *what* intent allows and d=
oesn=E2=80=99t allow, since we=E2=80=99ve heard from many sides that the bo=
undary is not clear, and it may be impossible to have a really clear, usabl=
e definition. So my point was: Let=E2=80=99s focus on the distribution mode=
l: Intent is flooded to all nodes. Let=E2=80=99s NOT do selective flooding,=
 unicast Intent to specific nodes, etc, because that would blur the boundar=
y even worse.  
 [LC]
 Befor=
e stating that intent policies should be flooded to all nodes, can we first=
 compare/evaluate the different options?
 -broadcast is simple...but how=
 important is it compared to other criteria...?
 I see at least two issu=
es:
 1. flooding a "big" intent file (cf your current tentativ=
e example) may be bandwidth hungry, if one considers how frequently such an=
 intent file could be distributed (ASA stop/start, device join/leave, polic=
y change/update...) compared to distributing only what is required / where =
it is required.

<jcs>
 Hopefully, intent will not change =
frequently (or else it probably is too low level in nature). I personally l=
ike using pub-sub, but that is because it enables me to do additional pre- =
and post-processing of intent. However, I think that this should be an opti=
on.
< /jcs>

 2. security: flooding the entire file rev=
eal information to target (possibly undesirable/rogue) entities that do not=
 need it (and thus may mis-use it).

< jcs>
 This is a good point! =
It definitely broadens the attack surface...
< /jcs>

 More =
fundamentally, I think it is misleading to imagine that intent policies wil=
l be contained in one "file" (it is possible to do so but impract=
ical/cumbersome to manage). cf also the next comment.
 [/LC]
&=
lt;jcs>
 See my previous reply. I am more worried about the user mode=
l - constructing a single file means that the constructor has to know imple=
mentation details. I think that this is unreasonable for the intent writer.=
 However, I see no reason why the intent compiler could not produce a singl=
e file, or sets of files that group similar intent (e.g., invoking multiple=
 autonomic functions in the same ASA) together.
< /jcs><=
/u>
=C2=A0
My suggestion: Intent is flooded t=
o all nodes in a domain.=

[LC=
]
 It is the "resolution" of the intent clauses that determine=
s the domain/set of target "nodes".
 So in that order, yes the=
 intent will be flooded to all the relevant nodes in the "intent polic=
y domain". This domain may be all nodes in an ANIMA network, or a sub-=
set, only one node, or zero. This is contextual.
 There will be ma=
ny intent policies for a given ANIMA network.
 [/LC]
<jcs>
Let's use a concept of:
=C2=A0
=C2=A0=C2=A0 +--------+=C2=A0=C2=A0=C2=
=A0=C2=A0 +----------+=C2=A0=C2=A0=C2=A0=C2=A0 +-----------+<=
/span>
=C2=A0=C2=A0 | Intent |---->| =C2=A0Intent=C2=A0 |---->|=
 Autonomic |
=C2=A0=C2=A0 | Writer |=C2=A0=C2=A0=
=C2=A0=C2=A0 | Compiler |=C2=A0=C2=A0=C2=A0=C2=A0 |=C2=A0 Network=C2=A0 |
=C2=A0=C2=A0 +--------+=C2=A0=C2=A0=C2=A0=C2=A0 +=
----------+=C2=A0=C2=A0=C2=A0=C2=A0 +-----------+
<=
p class=3D"MsoNormal">I **think** what La=
urent is saying is that once the Intent Compiler resolves the intent of the=
 user/intent writer, the target of the intent policy is then identified. I =
agree with this. Even for the broad case of "intent fabric", I do=
n't think that an intent writer need be conscious of this (vs. an auton=
omic function). Why should he or she care?
=C2=A0
Laurent, please corr=
ect me if I am wrong here.
</jcs=
>
=C2=A0
=C2=A0
The ASA owners define the content of what they wa=
nt to push in their bit of
 the Intent. If we insist of full flooding, w=
e encourage the right behaviour. And
 if someone really puts into his In=
tent =E2=80=9Cnode A: do this; node B: do the other; node C: do the third=
=E2=80=9D it is awkward enough for people to take a step back and think whe=
ther they=E2=80=99re doing the right thing J
[LC]
 I think we have a disconnect on what an intent look like a=
nd how it can be used.
 In my "mental model", there is not one=
 "big" intent with sub-parts for each ASA, the ANI, etc... 
 E=
ach ASA designer (owner) will define the content / specificity of its ASA f=
unctionality or links to the existing elements of the ANIMA intent policy m=
odel, but it does not mean that we must aggregate all parts into a single &=
quot;file".
 What would be the reason to do that?

 In additi=
on, there will not only be intent policies per functions, or for the ANI. There will be intent policies conveying the Service level, the Business =
level... these will not resolve/map 1-to-1 in a single ASA "policy&quo=
t;.
 [/LC]
<jcs>=

I don't think that you and Michae=
l are really disagreeing (other than one implementation (the file), which c=
ould be one option of many).
=C2=A0
Indeed, in FOCALE and other work I=
 have done, once intent was compiled, we did group things together for a nu=
mber of reasons (e.g., dependencies). I think that the Autonomic Functions =
should advertise their capabilities, or else, how does the Intent Compiler =
have a clue as to what gets shipped to what (not to mention, how does it tr=
anslate intent into a form that is consumable by ASAs). This is even more i=
mportant for Services - one could imagine that a single Service intent woul=
d need multiple Autonomic Functions to be implemented.
</jcs>
=C2=A0

jcs>
 Do you mean that the intent writer will **know the d=
ifference**
 between the fabric and each autonomic function? I hope not,=

 but perhaps I'm misunderstanding you...
< /jcs>=C2=A0
Not sure I understand your c=
oncern. Yes, in my model there is a section for the fabric, and sections fo=
r the various functions. Seems a natural structure?!? 
[LC]
 It is OK to have different intent-pol=
icies for the fabric and for the functions, but why put them all into one f=
ile/policy?
 [/LC]

<=
;jcs>
See my previous reply. In =
short, I was worried about this requiring the intent writer to be aware of =
which autonomic functions were present, and the nature of the autonomic fab=
ric. I don't care that much about the structure of the input. 
</jcs>
=C2=A0

=E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m s=
uggesting to structure Intent around Autonomic Functions. This is not optim=
al from a scientific point of view, but aligns very well with the way how t=
hings are organised. Probably we need to discuss that in BA. But refer to m=
y YANG model sent a few days ago. =

[LC]
 I fully agree on the focus per autonomic function, but it d=
oes not imply to have one single intent policy with all parts inside.
 [=
/LC]
<jcs>
My point is that the intent writer should not =
have to have detailed knowledge about how the autonomic network works, or w=
hat its constituent autonomic functions are. In my previous post, I talked =
about an "autonomic catalog" that the intent writer could use. Th=
is would hide the fact that a particular service in the autonomic function =
had a 1:1 or 1:n correspondence to a set of autonomic functions.<=
/u>
</jcs>
=C2=A0

We should schedule some discussion time outside the AN=
IMA meeting window...
[LC] Yes. And a white board too!
 Unfortunately, I cannot attend physicall=
y this time.
 Can we schedule a face-to-face ad-hoc meeting?
 [/LC]
<jcs>
=

Count me in.
</jcs>
=C2=A0<=
/p>
regards,
=
John

On =
Wed, Mar 23, 2016 at 4:54 AM, Laurent Ciavaglia <laurent.ciavagl=
ia@nokia.com> wrote:

 =20
   =20
 =20
  

    Michael,

      

      There are some aspects I am not sure to fully understand / agree
      with (marked below with [LC]).

      

      Best regards, Laurent.

      

    

    
On 23/03/2016 11:57, EXT Michael
      Behringer (mbehring) wrote:

    

    

     =20
     =20
     =20
      

        
John,
              

        [stripped]

        

        
The
              =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D d=
iscussion: My view is that we
              shouldn=E2=80=99t try to define exactly *what* intent
              allows and doesn=E2=80=99t allow, since we=E2=80=99ve heard f=
rom many
              sides that the boundary is not clear, and it may be
              impossible to have a really clear, usable definition. So
              my point was: Let=E2=80=99s focus on the distribution model:
              Intent is flooded to all nodes. Let=E2=80=99s NOT do selectiv=
e
              flooding, unicast Intent to specific nodes, etc, because
              that would blur the boundary even worse.
            

      

    

    

    [LC]

    Before stating that intent policies should be flooded to all nodes,
    can we first compare/evaluate the different options?

    -broadcast is simple...but how important is it compared to other
    criteria...?

    I see at least two issues:

    1. flooding a "big" intent file (cf your current tentative ex=
ample)
    may be bandwidth hungry, if one considers how frequently such an
    intent file could be distributed (ASA stop/start, device join/leave,
    policy change/update...) compared to distributing only what is
    required / where it is required.

    2. security: flooding the entire file reveal information to target
    (possibly undesirable/rogue) entities that do not need it (and thus
    may mis-use it).

    

    More fundamentally, I think it is misleading to imagine that intent
    policies will be contained in one "file" (it is possible to d=
o so
    but impractical/cumbersome to manage). cf also the next comment.

    [/LC]

    

    

      

        

        
My
              suggestion: Intent is flooded to all nodes in a domain.

      

    

    =
[LC]

    It is the "resolution" of the intent clauses that determines =
the
    domain/set of target "nodes".

    So in that order, yes the intent will be flooded to all the relevant
    nodes in the "intent policy domain". This domain may be all n=
odes in
    an ANIMA network, or a sub-set, only one node, or zero. This is
    contextual.

    There will be many intent policies for a given ANIMA
    network.

    [/LC]

    

    

      

        

              The ASA owners define the content of what they want to
              push in their bit of the Intent. If we insist of full
              flooding, we encourage the right behaviour. And if someone
              really puts into his Intent =E2=80=9Cnode A: do this; node B:=
 do
              the other; node C: do the third=E2=80=9D it is awkward enough=
 for
              people to take a step back and think whether they=E2=80=99re =
doing
              the right thing 
          J

      

    

    [LC]

    I think we have a disconnect on what an intent look like and how it
    can be used.

    In my "mental model", there is not one "big" intent=
 with sub-parts
    for each ASA, the ANI, etc... 

    Each ASA designer (owner) will define the content / specificity of
    its ASA functionality or links to the existing elements of the ANIMA
    intent policy model, but it does not mean that we must aggregate all
    parts into a single "file".

    What would be the reason to do that?

    

    In addition, there will not only be intent policies per functions,
    or for the ANI.

    There will be intent policies conveying the Service level, the
    Business level... these will not resolve/map 1-to-1 in a single ASA
    "policy".

    [/LC]

    

    

      

        
<jcs>

        
Do
              you mean that the intent writer will **know the
              difference**

        
between
              the fabric and each autonomic function? I hope not,=


        
but
              perhaps I'm misunderstanding you...<=
/font>

        
</jcs>

        
=C2=A0

        
Not
              sure I understand your concern. Yes, in my model there is
              a section for the fabric, and sections for the various
              functions. Seems a natural structure?!? 

      

    

    [LC]

    It is OK to have different intent-policies for the fabric and for
    the functions, but why put them all into one file/policy?

    [/LC]

    

      

        

        
=C2=A0

        
=E2=80=9CFunction
              owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structure Inte=
nt around
              Autonomic Functions. This is not optimal from a scientific
              point of view, but aligns very well with the way how
              things are organised. Probably we need to discuss that in
              BA. But refer to my YANG model sent a few days ago.
            

      

    

    [LC]

    I fully agree on the focus per autonomic function, but it does not
    imply to have one single intent policy with all parts inside.

    [/LC]

    =


    

      

        

        
=C2=A0

        
We
              should schedule some discussion time outside the ANIMA
              meeting window...

        
=C2=A0

      

    

    [LC]

    Yes. And a white board too!

    Unfortunately, I cannot attend physically this time.

    Can we schedule a face-to-face ad-hoc meeting?

    [/LC]

    

    

    

      

        
Michael

        
=C2=A0

        
=C2=A0

        

          

            

              
<=
span lang=3D"EN-US" style=3D"font-family:"Calibri",sans-serif;fon=
t-size:11pt;font-weight:bold">From: John Strassner
                    [mailto:strazpdj@gmail.com] 

                    Sent:
                    22 March 2016 21:32

                    To:
                    Michael Behringer (mbehring)
                    <mbehring@cisco.com>; John Strassner
                    <strazpdj@gmail.com>

                    Cc:
                    Duzongpeng <duzongpeng@huawei.com>;
                    ani=
ma@ietf.org; Laurent Ciavaglia
                    <laurent.ciavaglia@nokia.com>; J=C3=A9ferson Campos
                    Nobre <jcnobre@inf.ufrgs.br>; Sheng Jiang
                    <jiangsheng@huawei.com>

                    Subject:
                    Re: [Anima] Does Autonomic network need some
                    intervention beyond intent<=
/p>
            

          

          
=
=C2=A0

          

            

              
In general, I agree with
                    Michael. I'd like to emphasize and expand on a
                    couple of key points:

            

            

              
=C2=A0

            

            

              
=C2=A0

            

            

              
First,
                    ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=
=E2=80=9D or
                    =E2=80=9Csteering=E2=80=9D a network, at least not init=
ially. There
                    is still traditional config, plus all the existing
                    and emerging SDN models. My view has always been
                    that ANIMA sets a =E2=80=9Cdefault=E2=80=9D policy in a=
 network, and
                    all other means are more specific and can override
                    intent. This is described in RFC7575. Main point
                    here: ANIMA Intent isn=E2=80=99t the only tool in a net=
work.
                  

              
=C2=A0

            

            

              
<jcs><=
/p>
            

            

              
Absolutely agree. Note that
                    even in the Congress case I mentioned

            

            

              
earlier, it is NOT pushing
                    config, and it is NOT viewed as the

            

            

              
"only" policy

            

            

              

                
</j=
cs>

              

            

            

              
=C2=A0

            

            

              
=C2=A0

            

            

              
Intent
                    could perfectly well cover a high level policy such
                    as =E2=80=9Call nodes of type x do this; type y does th=
at=E2=80=9D.
                    (Which I think is what you mention below.) When we
                    started with Intent, the vision was to keep it very
                    high level, more along the line =E2=80=9Cdo the right t=
hing=E2=80=9D
                  J=C2=A0

              
=C2=A0

            

            

              

                
<jc=
s>

              

              

                
Agreed=
.
                      Although to me, this means that we need to
                      consider the types of actors that are using
                      intent. For example, most of my CCIE and JCIE
                      friends have no interest in using an intent
                      language for pushing configuration. On the other
                      hand, when I talk to operators and application
                      developers, they would love to be able to use a
                      high-level intent and have the intent engine take
                      care of the details for them.

              

              

                
</j=
cs>

              

              

                
=C2=A0

              

            

            

              

                
It has
                      become clear that in today=E2=80=99s networking, peop=
le
                      will want to push more =E2=80=9Cparameters=E2=80=9D t=
han =E2=80=9Cpolicy=E2=80=9D,
                      and I think we=E2=80=99ll just have to acknowledge th=
at.

              

              

                
=C2=A0

              

              

                
<jc=
s>

              

              

                
Sorry =
Michael,
                      I don't follow here. What do you mean

              

              

                
by
                      "parameters"?
              

              

                
</j=
cs>

              

              

                
=C2=A0

              

              

                

                  
<=
/u>=C2=A0

                

                

                  
Personally
                        I think it=E2=80=99ll pan out like this: Intent wil=
l
                        have sections per autonomic function. An Intent
                        parser will validate overall correctness
                        (signature, etc), then segment the Intent into
                        its bits: One part will be the autonomic fabric
                        itself, then there will be bits for each
                        autonomic function.
                      

                  
<=
/u>=C2=A0

                

                

                  
<=
jcs>

                

                

                  
Do y=
ou mean
                        that the intent writer will **know the
                        difference**

                

                

                  
betw=
een the
                        fabric and each autonomic function? I hope not,<=
/u>

                

                

                  
but =
perhaps
                        I'm misunderstanding you...

                

                

                  
<=
/jcs>

                

                

                  
<=
/u>=C2=A0

                

                

                  
<=
/u>=C2=A0

                

                

                  
So
                        the Intent parser on a node will grab intent,
                        see there is some Intent for autonomic function
                        A, will see whether this is running locally, and
                        if so, it=E2=80=99ll just pass that piece of Intent=
 to
                        that function.
                      

                

                

                  

                    
=C2=A0

                  

                  

                    
&l=
t;jcs>

                  

                  

                    
Ag=
reed.

                  

                  

                    
&l=
t;/jcs>

                  

                  

                    
=C2=A0

                  

                  

                    

                      
=C2=A0

                    

                    

                      
In
                            this way, a function owner defines both the
                            intent itself, and how it is consumed by
                            his/her function. <=
/p>
                      
=C2=A0

                    

                    

                      
<jcs>

                    

                    

                      
I'm
                            not sure what you mean by "function owner&=
quot;.
                            (Sorry to be

                    

                    

                      
picky,
                            I'm likely under-caffeinated.)

                    

                    

                      
=C2=A0

                    

                    

                      
The
                            above seemed like an autonomous decision,
                            made by the parser.=


                    

                    

                      
I
                            don't think that the intent writer should b=
e
                            cognizant of how many or

                    

                    

                      
what
                            types of autonomic functions are present.

                    

                    

                      
</jcs>

                    

                    

                      
=C2=A0

                    

                    

                      
=C2=A0

                    

                    

                      

                        
The
                              downside is that there is no =E2=80=9Ccontrol=
=E2=80=9D and
                              people can do what they want, in the worst
                              case even push config, which is clearly
                              NOT what Intent is meant to do.
                            

                        
=C2=A0

                      

                      

                        
<jcs>

                      

                      

                        
And
                              hence, the "HAL 2000" scenario. :-)=


                      

                      

                        
</jcs>

                      

                      

                        
=C2=A0

                      

                      

                        

                          
=C2=A0<=
/font>

                        

                        

                          
I=E2=80=99d
                                hope that people don=E2=80=99t start writin=
g
                                Intent like =E2=80=9Cnode 17: here is your =
CLI;
                                node 23: here is your CLI=E2=80=9D; node 37=
:
                                here is your CLI=E2=80=9D. But what I descr=
ibe
                                above COULD be abused in this way.
                                RFC7575 says this is NOT Intent.
                              

                        

                        

                          
=C2=A0<=
/font>

                        

                        

                          
<jcs>

                        

                        

                          
Absolutely
                                agree!

                        

                        

                          
</jcs>

                        

                        

                          
=C2=A0<=
/font>

                        

                      

                    

                  

                

              

            

            

              
=C2=A0regards,

              
John

              
=
=C2=A0

              
=C2=A0

            

          

          

            
=C2=A0

            

              
On Wed, Mar 16, 2016 at
                    1:16 AM, Michael Behringer (mbehring) <mbehring@cisco.com>
                    wrote:

              

                

                  

                    
Hi
                          Zongpeng,
                        

                    
=C2=A0<=
/p>
                    
Yes,
                          this is a recurring question and topic. Let me
                          give my view on two levels: =


                    
=C2=A0<=
/p>
                    
First,
                          ANIMA isn=E2=80=99t the only thing =E2=80=9Cinflu=
encing=E2=80=9D or
                          =E2=80=9Csteering=E2=80=9D a network, at least no=
t initially.
                          There is still traditional config, plus all
                          the existing and emerging SDN models. My view
                          has always been that ANIMA sets a =E2=80=9Cdefaul=
t=E2=80=9D
                          policy in a network, and all other means are
                          more specific and can override intent. This is
                          described in RFC7575. Main point here: ANIMA
                          Intent isn=E2=80=99t the only tool in a network.
                        

                    
=C2=A0<=
/p>
                    
Intent
                          could perfectly well cover a high level policy
                          such as =E2=80=9Call nodes of type x do this; typ=
e y
                          does that=E2=80=9D. (Which I think is what you me=
ntion
                          below.) When we started with Intent, the
                          vision was to keep it very high level, more
                          along the line =E2=80=9Cdo the right thing=E2=80=
=9D
                        J=C2=A0 It
                          has become clear that in today=E2=80=99s networki=
ng,
                          people will want to push more =E2=80=9Cparameters=
=E2=80=9D
                          than =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=
=80=99ll just have to
                          acknowledge that.
                        

                    
=C2=A0<=
/p>
                    
Personally
                          I think it=E2=80=99ll pan out like this: Intent w=
ill
                          have sections per autonomic function. An
                          Intent parser will validate overall
                          correctness (signature, etc), then segment the
                          Intent into its bits: One part will be the
                          autonomic fabric itself, then there will be
                          bits for each autonomic function.
                        

                    
=C2=A0<=
/p>
                    
So
                          the Intent parser on a node will grab intent,
                          see there is some Intent for autonomic
                          function A, will see whether this is running
                          locally, and if so, it=E2=80=99ll just pass that =
piece
                          of Intent to that function.
                        

                    
=C2=A0<=
/p>
                    
In
                          this way, a function owner defines both the
                          intent itself, and how it is consumed by
                          his/her function. 
                    
=C2=A0<=
/p>
                    
The
                          downside is that there is no =E2=80=9Ccontrol=E2=
=80=9D and
                          people can do what they want, in the worst
                          case even push config, which is clearly NOT
                          what Intent is meant to do. I=E2=80=99d hope that
                          people don=E2=80=99t start writing Intent like =
=E2=80=9Cnode
                          17: here is your CLI; node 23: here is your
                          CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D.=
 But what I
                          describe above COULD be abused in this way.
                          RFC7575 says this is NOT Intent. 
                      

                    
=C2=A0<=
/p>
                    
In
                          other words: Let=E2=80=99s not be too prescriptiv=
e,
                          and hope that folks will do more or less
                          reasonable things. 
                    
=C2=A0<=
/p>
                    
I=E2=80=99ve
                          been meaning to comment on the Intent drafts
                          and possibly contribute, I=E2=80=99m just struggl=
ing
                          with priorities at the moment (as you can see
                          by my relative silence over the past weeks).
                        

                    
=C2=A0<=
/p>
                    
Michael=


                    
=C2=A0<=
/p>
                    
=C2=A0<=
/p>
                    
=C2=A0<=
/p>
                    

                      

                        

                          
From: Anima [mailto:=
=
anima-bounces@ietf.org]
                                On
                                    Behalf Of Duzongpeng

                                Sent:
                                16 March 2016 02:38

                                To:
                              anima@ietf.org

                                Cc:
                                Laurent Ciavaglia <=
lau=
rent.ciavaglia@nokia.com>; J=C3=A9ferson Campos Nobre
                                <jcnobre@inf.ufrgs.br>; Sheng Jiang=
 <jiangsheng@huawei.com>

                                Subject=
:
                                [Anima] Does Autonomic network need some
                                intervention beyond intent=


                        

                      

                      
=C2=A0

                      
Hello, everyone

                      
=C2=A0

                      
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 I unde=
rstand that
                            Autonomic network should be able to work
                            with as less configuration as possible. In
                            the ideal case, an Autonomic network should
                            be able to work well while only intent is
                            needed from human operators.

                      
=C2=A0

                      
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Howeve=
r, I wonder
                            whether Autonomic network may need some
                            intervention from human operators through
                            some low-level policies beyond intents.<=
/font>

                      
=C2=A0

                      
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 For ex=
ample, in the
                          https://tools.ietf.org/html/draft-ietf-=
anima-prefix-management-00, it is suggested that the
                            prefix lengths for the CSG, ASG, RSG
                            (different roles in IP RAN) can be assigned
                            as an "intent". These configuration
                            parameters need to be distributed in the
                            autonomic domain to influence the detail
                            configurations on each autonomic node.

                      
=C2=A0

                      
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 As int=
ent is described
                            as an abstract, declarative, high-level
                            policy used to operate an autonomic domain,
                            such as an enterprise network. Perhaps we
                            can rename the =E2=80=9Cintent=E2=80=9D in the
                            =E2=80=9Cprefix-management ID=E2=80=9D to some =
other words.

                      
=C2=A0

                      
Is =E2=80=9CASA parameters&q=
uot; ok here?
                          

                      
=C2=A0<=
u>

                      
I think some characteristics
                            for this kind of low-level policies are<=
/font>

                      
1.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
                          Network level
                          parameters, need to be distributed by GRASP
                          and ACP

                      
2.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
                          Relatively
                          static compared to intent, perhaps only needed
                          to be configured once

                      
3.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
                          Unavoidable
                          in current network environment, mostly for
                          establishing the network infrastructure=


                      
4.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
                          Rarely need
                          coordinations with others parameters, usually
                          is closed related to one specific autonomic
                          function

                      
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
                          

                      
Welcome for comment. Your
                            suggestions will be appreciated.<=
u>

                      
=C2=A0<=
u>

                      
=C2=A0<=
u>

                      
Best Regards

                      
Zongpeng Du

                    

                  

                

                


                      _______________________________________________

                      Anima mailing list

                      A=
nima@ietf.org

                      https://www.ietf.org/mailman/listinfo/anima=


              

            

            


                  

                  

                  -- 

            

              

                
regard=
s,

              

              

                
John

              

            

          

        

      

    

    

    
-- 

     =20
     =20
     =20
     =20
     =20
     =20
     =20
     =20
     =20
     =20
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
=C2=A0

        
+33=C2=A0160 402=C2=A0636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia=


        
=C2=
=A0

      

    

  






-- 
regards,
John




--001a11c3eab418737b052ec2f5e8--


From nobody Wed Mar 23 20:40:46 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5A07B12D167 for ; Wed, 23 Mar 2016 20:40:44 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ksaa9C6C8eKA for ; Wed, 23 Mar 2016 20:40:40 -0700 (PDT)
Received: from mail-lf0-x235.google.com (mail-lf0-x235.google.com [IPv6:2a00:1450:4010:c07::235]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2FF9612D0BC for ; Wed, 23 Mar 2016 20:40:39 -0700 (PDT)
Received: by mail-lf0-x235.google.com with SMTP id q73so22893567lfe.2 for ; Wed, 23 Mar 2016 20:40:39 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=OI5AWI9GScuD57Ww/yH+nP8uy+jpUUG4wPpKoZ+u41o=; b=hAfX+w3kpjnVUXq15K7BBYWaCQ2E2If4sUELoo5XO98wPhl17CQJeQhxhCHDnAg/v4 6+I2xKo2R9ToQgg9dhMR3QpZexobhcg0fXc+zMrafh9bfqo+bzCR8jeeIKm/3dvBuSt3 9/C/EC1aGicam9d+SbZ8Zx+w7SI45dSwOE3sD0JdHRgWgDsmltt9dPxPcxuGrGwq09fH nHIC3RiQa4QrrUley1VJtce6rE42SCOtHsILVaonXLpQDGav044BnrtJUN/uLmmSVYkw uNf/ODHY33GReK3gtqlMc1pmgxBMUMGR5OhD/LLa4KBdt4qYUVyhLEs6p8XkE9FlM/d/ q8VQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=OI5AWI9GScuD57Ww/yH+nP8uy+jpUUG4wPpKoZ+u41o=; b=T5hkwRuWh71BjPzJ+aXicPtRhsAr5V+kEmEsXuHf+J/DNNAu7AkTMmHxtRdWEUl8hA N7l5kxB1y4z0T2SrWlTz4nSKSfnkOMUj04knfgfyHxlUQ12K+DGWBT+nyot8LbHX9Cfh 9AOcubTKWTk/li+zmrHEdI/FYK3aFxdDz/Pa1aqNH+ZGpNuBt6WdjLQ6babDo65PBmYv ICrMK5VTh4QYpgBPeoHP1sXjftlseXjSdNYIl6TunauWg8UdLV5j2v8YMCOELtr43bT7 t7B2W6/9hftXxLekED38X2PSFdpJ9+SUCGw6wYE0awKCqIRs8UiYy64s6MCBvEdmEENW sJoA==
X-Gm-Message-State: AD7BkJLPYh0XVXzzXPjvUv418PhBdE7TcjTq8BwJfLjFYp2b1qos/+EyzR0GulaKBg8vj3uYS0JZQbUAaShqoA==
MIME-Version: 1.0
X-Received: by 10.25.143.141 with SMTP id r135mr2725659lfd.100.1458790837270;  Wed, 23 Mar 2016 20:40:37 -0700 (PDT)
Received: by 10.25.16.160 with HTTP; Wed, 23 Mar 2016 20:40:37 -0700 (PDT)
In-Reply-To: <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>
Date: Wed, 23 Mar 2016 20:40:37 -0700
Message-ID: 
From: John Strassner 
To: "Alexander Clemm (alex)" , John Strassner 
Content-Type: multipart/alternative; boundary=001a114016e4a83fc6052ec33530
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 03:40:44 -0000

--001a114016e4a83fc6052ec33530
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Alex,



I'm not sure I follow a couple of your points, please see inline...though
in retrospect, perhaps this is merely a case of who the actor is (in my
case, I am worried about the writer and user of intent, not how we build
it).





I really like your suggestion of flooding intent to all nodes (even if
there may be optimizations later).  It provides a level of conceptual
clarity.  Every node should be aware of overall intent (and need to decide
for itself if there are actions for it to take).



Why should **every** node be aware of all intent? Many times, this won't be
applicable.


The only reason I can see for doing this is to enable collaboration and/or
emergent behavior (in the AI sense). However, if you want emergent
behavior from autonomic nodes, then each node that is part of that
collaboration must interact. I don't think we know enough about this to
support or prohibit such behavior, other than this places an extra
requirement on the computational and reasoning abilities of the node.

If you have a different purpose, please do explain.







If there is some intent that applies only to some parts of the network and
not others, that can be addressed by using different Autonomic Domains.
And, of course, Autonomic Domains may be associated with certain Autonomic
Functions.  Intervention beyond intent should NOT be allowed for the
Autonomic Function(s) being served an Autonomic Domain, but of course other
interventions (for different functions, hence outside the scope) are still
possible.



I think that Autonomic Domains should (must?) be hidden from the
user/intent writer. They are clearly of huge internal benefit, but why
should the user of intent know about them? In addition, I think that we
should use roles as an indirection mechanism for scalability when working
with Autonomic Domains.







regards,

John

On Wed, Mar 23, 2016 at 10:56 AM, Alexander Clemm (alex) 
wrote:

> I really like your suggestion of flooding intent to all nodes (even if
> there may be optimizations later).  It provides a level of conceptual
> clarity.  Every node should be aware of overall intent (and need to decid=
e
> for itself if there are actions for it to take).
>
>
>
> If there is some intent that applies only to some parts of the network an=
d
> not others, that can be addressed by using different Autonomic Domains.
> And, of course, Autonomic Domains may be associated with certain Autonomi=
c
> Functions.  Intervention beyond intent should NOT be allowed for the
> Autonomic Function(s) being served an Autonomic Domain, but of course oth=
er
> interventions (for different functions, hence outside the scope) are stil=
l
> possible.
>
>
>
> --- Alex
>
>
>
> *From:* Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Michael
> Behringer (mbehring)
> *Sent:* Wednesday, March 23, 2016 3:57 AM
> *To:* John Strassner 
> *Cc:* Duzongpeng ; Laurent Ciavaglia <
> laurent.ciavaglia@nokia.com>; Sheng Jiang ;
> anima@ietf.org; J=C3=A9ferson Campos Nobre 
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> John,
>
>
>
> I think we=E2=80=99re on the same page. Couple of points, top posting her=
e,
> otherwise it gets too messy.
>
>
>
> While *CIEs don=E2=80=99t like magic, they do understand the concept of a=
 default
> behaviour which they can influence if needed. Any reasonable network
> management approach tries to define =E2=80=9Cdefault=E2=80=9D behaviour a=
nd =E2=80=9Cexceptions=E2=80=9D.
> The AN story fills the =E2=80=9Cdefault=E2=80=9D part very nicely, and *C=
IEs get that.
>
>
>
> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussio=
n: My view is that we shouldn=E2=80=99t
> try to define exactly **what** intent allows and doesn=E2=80=99t allow, s=
ince
> we=E2=80=99ve heard from many sides that the boundary is not clear, and i=
t may be
> impossible to have a really clear, usable definition. So my point was:
> Let=E2=80=99s focus on the distribution model: Intent is flooded to all n=
odes.
> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific nodes=
, etc,
> because that would blur the boundary even worse.
>
>
>
> My suggestion: Intent is flooded to all nodes in a domain. The ASA owners
> define the content of what they want to push in their bit of the Intent. =
If
> we insist of full flooding, we encourage the right behaviour. And if
> someone really puts into his Intent =E2=80=9Cnode A: do this; node B: do =
the other;
> node C: do the third=E2=80=9D it is awkward enough for people to take a s=
tep back
> and think whether they=E2=80=99re doing the right thing J
>
>
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
> Not sure I understand your concern. Yes, in my model there is a section
> for the fabric, and sections for the various functions. Seems a natural
> structure?!?
>
>
>
> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structur=
e Intent around Autonomic
> Functions. This is not optimal from a scientific point of view, but align=
s
> very well with the way how things are organised. Probably we need to
> discuss that in BA. But refer to my YANG model sent a few days ago.
>
>
>
> We should schedule some discussion time outside the ANIMA meeting window.=
..
>
>
>
> Michael
>
>
>
>
>
> *From:* John Strassner [mailto:strazpdj@gmail.com ]
> *Sent:* 22 March 2016 21:32
> *To:* Michael Behringer (mbehring) ; John Strassner <
> strazpdj@gmail.com>
> *Cc:* Duzongpeng ; anima@ietf.org; Laurent
> Ciavaglia ; J=C3=A9ferson Campos Nobre <
> jcnobre@inf.ufrgs.br>; Sheng Jiang 
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> In general, I agree with Michael. I'd like to emphasize and expand on a
> couple of key points:
>
>
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
>
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J
>
>
>
> 
>
> Agreed. Although to me, this means that we need to consider the types of
> actors that are using intent. For example, most of my CCIE and JCIE frien=
ds
> have no interest in using an intent language for pushing configuration. O=
n
> the other hand, when I talk to operators and application developers, they
> would love to be able to use a high-level intent and have the intent engi=
ne
> take care of the details for them.
>
> 
>
>
>
> It has become clear that in today=E2=80=99s networking, people will want =
to push
> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I th=
ink we=E2=80=99ll just have to acknowledge
> that.
>
>
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
>
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> 
>
> Agreed.
>
> 
>
>
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
>
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
>
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do.
>
>
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
>
>
>
>
> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=
=80=9Cnode 17: here is
> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CLI=
=E2=80=9D. But what
> I describe above COULD be abused in this way. RFC7575 says this is NOT
> Intent.
>
>
>
> 
>
> Absolutely agree!
>
> 
>
>
>
>  regards,
>
> John
>
>
>
>
>
>
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
> mbehring@cisco.com> wrote:
>
> Hi Zongpeng,
>
>
>
> Yes, this is a recurring question and topic. Let me give my view on two
> levels:
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J  It has=
 become clear
> that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writing =
Intent like =E2=80=9Cnode
> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here i=
s your
> CLI=E2=80=9D. But what I describe above COULD be abused in this way. RFC7=
575 says
> this is NOT Intent.
>
>
>
> In other words: Let=E2=80=99s not be too prescriptive, and hope that folk=
s will do
> more or less reasonable things.
>
>
>
> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly co=
ntribute,
> I=E2=80=99m just struggling with priorities at the moment (as you can see=
 by my
> relative silence over the past weeks).
>
>
>
> Michael
>
>
>
>
>
>
>
> *From:* Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Duzongpeng
> *Sent:* 16 March 2016 02:38
> *To:* anima@ietf.org
> *Cc:* Laurent Ciavaglia ; J=C3=A9ferson Camp=
os
> Nobre ; Sheng Jiang 
> *Subject:* [Anima] Does Autonomic network need some intervention beyond
> intent
>
>
>
> Hello, everyone
>
>
>
>          I understand that Autonomic network should be able to work with
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
>
>
>
>          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
>
>
>
>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
>
>
>
>          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
>
>
>
> Is =E2=80=9CASA parameters" ok here?
>
>
>
> I think some characteristics for this kind of low-level policies are
>
> 1.       Network level parameters, need to be distributed by GRASP and AC=
P
>
> 2.       Relatively static compared to intent, perhaps only needed to be
> configured once
>
> 3.       Unavoidable in current network environment, mostly for
> establishing the network infrastructure
>
> 4.       Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
>
>
>
> Welcome for comment. Your suggestions will be appreciated.
>
>
>
>
>
> Best Regards
>
> Zongpeng Du
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> --
>
> regards,
>
> John
>



--=20
regards,
John

--001a114016e4a83fc6052ec33530
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi Alex,
=C2=A0
I'm not s=
ure I follow a couple of your points, please see inline...though in retrosp=
ect, perhaps this is merely a case of who the actor is (in my case, I am wo=
rried about the writer and user of intent, not how we build it).<=
/u>
=C2=A0
=C2=A0
I really like your suggestion of flooding intent to all nodes (e=
ven if there may be optimizations later).=C2=A0 It provides a level of conc=
eptual clarity.=C2=A0 Every node should be aware of overall intent (and nee=
d to decide for itself if there are actions for it to take).
<jcs>=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 
Why should **every** node be aware of all intent? Many times, this wo=
n't be applicable.
=C2=
=A0
The only reason I can see for doing =
this is to enable collaboration and/or emergent behavior (in the AI sense).=
 However, if you want emergent behavior=C2=A0from autonomic nodes, then eac=
h node that is part of that collaboration must interact. I don't think =
we know enough about this to support or prohibit such behavior, other than =
this places an extra requirement on the computational and reasoning abiliti=
es of the node.

If you have a different purpose, please do explain.
</jcs>
=C2=A0
=C2=A0
<=
p class=3D"MsoNormal">If there is some=
 intent that applies only to some parts of the network and not others, that=
 can be addressed by using different Autonomic Domains.=C2=A0 And, of cours=
e, Autonomic Domains may be associated with certain Autonomic Functions.=C2=
=A0 Intervention beyond intent should NOT be allowed for the Autonomic Func=
tion(s) being served an Autonomic Domain, but of course other interventions=
 (for different functions, hence outside the scope) are still possible.=C2=
=A0
<=
;jcs>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 
=

I think that Autonomic Domains should (must?) be hid=
den from the user/intent writer. They are clearly of huge internal benefit,=
 but why should the user of intent know about them? In addition, I think th=
at we should use roles as an indirection mechanism for scalability when wor=
king with Autonomic Domains.
</j=
cs>
=C2=A0
=C2=A0
regards,<=
u>
John 

On Wed, Mar 23, 2016 at 10:56 AM, Ale=
xander Clemm (alex) <alex@cisco.com> wrote:











I really like your suggestion of=
 flooding intent to all nodes (even if there may be optimizations later).=
=C2=A0 It provides a level of conceptual clarity.=C2=A0 Every node
 should be aware of overall intent (and need to decide for itself if there =
are actions for it to take).=C2=A0



=C2=A0


If there is some intent that app=
lies only to some parts of the network and not others, that can be addresse=
d by using different Autonomic Domains.=C2=A0 And, of course,
 Autonomic Domains may be associated with certain Autonomic Functions.=C2=
=A0 Intervention beyond intent should NOT be allowed for the Autonomic Func=
tion(s) being served an Autonomic Domain, but of course other interventions=
 (for different functions, hence outside
 the scope) are still possible.=C2=A0 


=C2=A0


--- Alex

=C2=A0






From: Anima [mailto:anima-bounces@ietf.org]
On Behalf Of Michael Behringer (mbehring)

Sent: Wednesday, March 23, 2016 3:57 AM

To: John Strassner <strazpdj@gmail.com>

Cc: Duzongpeng <duzongpeng@huawei.com>; Laurent Ciavaglia <laurent.ciavaglia@no=
kia.com>; Sheng Jiang <jiangsheng@huawei.com>; anima@ietf.org; J=C3=A9ferson Campos Nobre <=
;jcnobre@inf.ufrg=
s.br>

Subject: Re: [Anima] Does Autonomic network need some intervention b=
eyond intent






=C2=A0


John,



=C2=A0<=
/u>


I think we=E2=80=
=99re on the same page. Couple of points, top posting here, otherwise it ge=
ts too messy.


=C2=A0<=
/u>


While *CIEs don=
=E2=80=99t like magic, they do understand the concept of a default behaviou=
r which they can influence if needed. Any reasonable network management
 approach tries to define =E2=80=9Cdefault=E2=80=9D behaviour and =E2=80=9C=
exceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=E2=80=9D part =
very nicely, and *CIEs get that.



=C2=A0<=
/u>


The =E2=80=9Cpara=
meter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussion: My view is that =
we shouldn=E2=80=99t try to define exactly *what* intent allows and =
doesn=E2=80=99t allow, since
 we=E2=80=99ve heard from many sides that the boundary is not clear, and it=
 may be impossible to have a really clear, usable definition. So my point w=
as: Let=E2=80=99s focus on the distribution model: Intent is flooded to all=
 nodes. Let=E2=80=99s NOT do selective flooding, unicast
 Intent to specific nodes, etc, because that would blur the boundary even w=
orse. 



=C2=A0<=
/u>


My suggestion: In=
tent is flooded to all nodes in a domain. The ASA owners define the content=
 of what they want to push in their bit of the Intent.
 If we insist of full flooding, we encourage the right behaviour. And if so=
meone really puts into his Intent =E2=80=9Cnode A: do this; node B: do the =
other; node C: do the third=E2=80=9D it is awkward enough for people to tak=
e a step back and think whether they=E2=80=99re doing the
 right thing J=C2=A0<=
/u>


<jcs>


Do you mean that =
the intent writer will **know the difference**


between the fabri=
c and each autonomic function? I hope not,


but perhaps I'=
;m misunderstanding you...


</jcs>


=C2=A0<=
/u>


Not sure I unders=
tand your concern. Yes, in my model there is a section for the fabric, and =
sections for the various functions. Seems a natural structure?!?



=C2=A0<=
/u>


=E2=80=9CFunction=
 owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structure Intent around Aut=
onomic Functions. This is not optimal from a scientific point of view, but =
aligns
 very well with the way how things are organised. Probably we need to discu=
ss that in BA. But refer to my YANG model sent a few days ago.



=C2=A0<=
/u>


We should schedul=
e some discussion time outside the ANIMA meeting window...


=C2=A0<=
/u>


Michael=



=C2=A0<=
/u>


=C2=A0<=
/u>








From: John Strassner [mailto:strazpdj@gmail.com]


Sent: 22 March 2016 21:32

To: Michael Behringer (mbehring) <mbehring@cisco.com>; John Strassner <strazpdj@gmail.com=
>

Cc: Duzongpeng <duzongpeng@huawei.com>;
anima@ietf.org; Lau=
rent Ciavaglia <laurent.ciavaglia@nokia.com>; J=C3=A9ferson Campos Nobre &l=
t;jcnobre@inf.ufr=
gs.br>; Sheng Jiang <jiangsheng@huawei.com>

Subject: Re: [Anima] Does Autonomic network need some intervention b=
eyond intent






=C2=A0






In general, I agree with Michae=
l. I'd like to emphasize and expand on a couple of key points:






=C2=A0






=C2=A0






First, ANIMA isn=
=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =E2=80=9Csteerin=
g=E2=80=9D a network, at least not
 initially. There is still traditional config, plus all the existing and em=
erging SDN models. My view has always been that ANIMA sets a =E2=80=9Cdefau=
lt=E2=80=9D policy in a network, and all other means are more specific and =
can override intent. This is described in RFC7575.
 Main point here: ANIMA Intent isn=E2=80=99t the only tool in a network. 


=C2=A0






<jcs>






Absolutely agree. Note that eve=
n in the Congress case I mentioned






earlier, it is NOT pushing conf=
ig, and it is NOT viewed as the






"only" policy<=
u>








</jcs>








=C2=A0






=C2=A0






Intent could perf=
ectly well cover a high level policy such as =E2=80=9Call nodes of type x d=
o
 this; type y does that=E2=80=9D. (Which I think is what you mention below.=
) When we started with Intent, the vision was to keep it very high level, m=
ore along the line =E2=80=9Cdo the right thing=E2=80=9D
J=C2=A0


=C2=A0








<jcs>






Agreed. Although to me, this me=
ans that we need to consider the types of actors that are using intent. For=
 example, most of my CCIE and JCIE friends have no interest in using an int=
ent language for pushing configuration.
 On the other hand, when I talk to operators and application developers, th=
ey would love to be able to use a high-level intent and have the intent eng=
ine take care of the details for them.






</jcs>






=C2=A0










=
It has become clear that in today=E2=80=99s networking, people will want to=
 push more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and =
I think we=E2=80=99ll just have to acknowledge that.






=C2=A0






<jcs>






Sorry Michael, I don't foll=
ow here. What do you mean






by "parameters"?






</jcs>






=C2=A0








=C2=A0






Personally I thin=
k it=E2=80=99ll pan out like this: Intent will have sections per autonomic =
function.
 An Intent parser will validate overall correctness (signature, etc), then =
segment the Intent into its bits: One part will be the autonomic fabric its=
elf, then there will be bits for each autonomic function.



=C2=A0






<jcs>






Do you mean that the intent wri=
ter will **know the difference**






between the fabric and each aut=
onomic function? I hope not,






but perhaps I'm misundersta=
nding you...






</jcs>






=C2=A0






=C2=A0






So the Intent par=
ser on a node will grab intent, see there is some Intent for autonomic
 function A, will see whether this is running locally, and if so, it=E2=80=
=99ll just pass that piece of Intent to that function.









=C2=A0






<jcs>






Agreed.





</jcs>






=C2=A0








=C2=A0






In this way, a fu=
nction owner defines both the intent itself, and how it is consumed by
 his/her function. 


=C2=A0






<jcs>






I'm not sure what you mean =
by "function owner". (Sorry to be






picky, I'm likely under-caf=
feinated.)






=C2=A0






The above seemed like an autono=
mous decision, made by the parser.






I don't think that the inte=
nt writer should be cognizant of how many or






what types of autonomic functio=
ns are present.






</jcs>






=C2=A0






=C2=A0








The downside is that there is no=
 =E2=80=9Ccontrol=E2=80=9D and people can do what they want, in the worst c=
ase even push
 config, which is clearly NOT what Intent is meant to do. 


=C2=A0






<jcs>






And hence, the "HAL 2000&q=
uot; scenario. :-)






</jcs>






=C2=A0








=C2=A0






I=E2=80=99d hope that people don=
=E2=80=99t start writing Intent like =E2=80=9Cnode 17: here is your CLI; no=
de 23: here is your CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D. But w=
hat I describe above
 COULD be abused in this way. RFC7575 says this is NOT Intent. 






=C2=A0






<jcs>






Absolutely agree!=







</jcs>






=C2=A0


















=C2=A0regards,


John


=
=C2=A0


=C2=A0








=C2=A0




On Wed, Mar 16, 2016 at 1:16 AM=
, Michael Behringer (mbehring) <mbehring@cisco.com> wrote:








Hi Zongpeng,



=C2=A0


Yes, this is a re=
curring question and topic. Let me give my view on two levels:



=C2=A0


First, ANIMA isn=
=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =E2=80=9Csteerin=
g=E2=80=9D a network, at least not
 initially. There is still traditional config, plus all the existing and em=
erging SDN models. My view has always been that ANIMA sets a =E2=80=9Cdefau=
lt=E2=80=9D policy in a network, and all other means are more specific and =
can override intent. This is described in RFC7575.
 Main point here: ANIMA Intent isn=E2=80=99t the only tool in a network. 


=C2=A0


Intent could perf=
ectly well cover a high level policy such as =E2=80=9Call nodes of type x d=
o
 this; type y does that=E2=80=9D. (Which I think is what you mention below.=
) When we started with Intent, the vision was to keep it very high level, m=
ore along the line =E2=80=9Cdo the right thing=E2=80=9D
J=C2=A0 It has become clear that in today=E2=80=99s net=
working, people will want to push more =E2=80=9Cparameters=E2=80=9D than =
=E2=80=9Cpolicy=E2=80=9D,
 and I think we=E2=80=99ll just have to acknowledge that. 


=C2=A0


Personally I thin=
k it=E2=80=99ll pan out like this: Intent will have sections per autonomic =
function.
 An Intent parser will validate overall correctness (signature, etc), then =
segment the Intent into its bits: One part will be the autonomic fabric its=
elf, then there will be bits for each autonomic function.



=C2=A0


So the Intent par=
ser on a node will grab intent, see there is some Intent for autonomic
 function A, will see whether this is running locally, and if so, it=E2=80=
=99ll just pass that piece of Intent to that function.



=C2=A0


In this way, a fu=
nction owner defines both the intent itself, and how it is consumed by
 his/her function. 


=C2=A0


The downside is t=
hat there is no =E2=80=9Ccontrol=E2=80=9D and people can do what they want,=
 in the worst
 case even push config, which is clearly NOT what Intent is meant to do. I=
=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=80=
=9Cnode 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: =
here is your CLI=E2=80=9D. But what I describe above COULD be abused in thi=
s
 way. RFC7575 says this is NOT Intent. <=
u>


=C2=A0


In other words: L=
et=E2=80=99s not be too prescriptive, and hope that folks will do more or l=
ess
 reasonable things. 


=C2=A0


I=E2=80=99ve been=
 meaning to comment on the Intent drafts and possibly contribute, I=E2=80=
=99m just struggling
 with priorities at the moment (as you can see by my relative silence over =
the past weeks).



=C2=A0


Michael


=C2=A0


=C2=A0


=C2=A0








From: An=
ima [mailto:anim=
a-bounces@ietf.org]
On Behalf Of Duzongpeng

Sent: 16 March 2016 02:38

To: anima@ietf.or=
g

Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; J=C3=A9ferson Campos Nobre
 <jcnobre@inf.u=
frgs.br>; Sheng Jiang &=
lt;jiangsheng@hua=
wei.com>

Subject: [Anima] Does Autonomic network need some intervention beyon=
d intent






=C2=A0


Hello, everyone<=
u>


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 I =
understand that Autonomic network should be able to work with as less confi=
guration as possible. In the
 ideal case, an Autonomic network should be able to work well while only in=
tent is needed from human operators.<=
/u>


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Ho=
wever, I wonder whether Autonomic network may need some intervention from h=
uman operators through some
 low-level policies beyond intents.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Fo=
r example, in the
https://to=
ols.ietf.org/html/draft-ietf-anima-prefix-management-00,
 it is suggested that the prefix lengths for the CSG, ASG, RSG (different r=
oles in IP RAN) can be assigned as an "intent". These configurati=
on parameters need to be distributed in the autonomic domain to influence t=
he detail configurations on each autonomic
 node.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 As=
 intent is described as an abstract, declarative, high-level policy used to=
 operate an autonomic domain,
 such as an enterprise network. Perhaps we can rename the =E2=80=9Cintent=
=E2=80=9D in the =E2=80=9Cprefix-management ID=E2=80=9D to some other words=
.


=C2=A0



Is =E2=80=9CASA parameters" ok here?




=C2=A0



I think some characteristics for this kind of low-level policies are


1.=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Network level parameters, need to be distributed by GRASP and ACP


2.=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Relatively static compared to intent, perhaps only needed to be conf=
igured once


3.=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Unavoidable in current network environment, mostly for establishing =
the network infrastructure


4.=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Rarely need coordinations with others parameters, usually is closed =
related to one specific autonomic function


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0




Welcome for comment. Your suggestions will be appreciated.



=C2=A0



=C2=A0


Best Regards<=
/u>


Zongpeng Du









_______________________________________________

Anima mailing list

Anima@ietf.org

h=
ttps://www.ietf.org/mailman/listinfo/anima












-- 






regards,





John


















-- 
regards,
John




--001a114016e4a83fc6052ec33530--


From nobody Wed Mar 23 20:49:30 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E555D12D187 for ; Wed, 23 Mar 2016 20:49:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id o2qHD2lD-V87 for ; Wed, 23 Mar 2016 20:49:25 -0700 (PDT)
Received: from mail-lf0-x22a.google.com (mail-lf0-x22a.google.com [IPv6:2a00:1450:4010:c07::22a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id DBCAE12D10A for ; Wed, 23 Mar 2016 20:49:24 -0700 (PDT)
Received: by mail-lf0-x22a.google.com with SMTP id o73so25462172lfe.0 for ; Wed, 23 Mar 2016 20:49:24 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=J8vLm6KeFyfJJf1CTYDvSVoDEOfmdo8O7XUf+v7y2lA=; b=LOxwDeXYP7iRgvdC78Gfd8auWDdGqYcejLsNrlcZ2+vCQAKFw6DCONOzNVh66Y2uvk 6u7uRXX7ytujFs+PM8zeS6wzGkKJSgt+5nP/oamc30BctPkOap0yGuUT+aC+WLW10sfK NDOwkPuINLAmRD52qy/OFX+EHeDrMOx733yZyVYK1Xn9XM0SLOYx7VuPGspAkFCdD4X3 R656qp1gyW6iCkyDUkdfT2c8ydeMq21r3HQosbCN7R0XTKuM8NtP9nMqL4NW5m6I0T3C sU2lT0OiB8Un3tc+5j6Tcj4PWNPdz+SKT0zBmecnL113mBetGg2PwPr+DKPScPSpDaOu 8vNg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=J8vLm6KeFyfJJf1CTYDvSVoDEOfmdo8O7XUf+v7y2lA=; b=kwDDxU1kSsIlw4AnpfTfu/LPDJGhHJBO0O7Pc+XT7415GjcE/2epta/FWYFuO3AKQz UU22kevPGBIF2WXoW2OqU38pvAA0AROlzT+EA2hxLCKn3JS9fmIusBgqYgNAmho8msG6 Xir9EHhGV7AOBQSeqNomaDtH9xWXRULmC8iEpU0TfDaQRitR8WvjtczpPIP0OY8c62bT FvA7mdd/0h+AWtdAYeIaYjiZioTFyaqQ9VA81Sg/L3I6EHQqyFEJXNBPuGcgiQk9DMXJ 4VNY4olpYNbryP5G4lySWyu8geu8jO0VpTbQe9eXy+9fAt4N7KZS9V6mEJtsyyCtnbzz elcw==
X-Gm-Message-State: AD7BkJIVrzIdZgm0BnJAkb4VzZBUGfKdGEOt+qDZK69Pd4lJZBij+c05My3Td+LaiJdjKb6NRi8bvDzDHLqz2w==
MIME-Version: 1.0
X-Received: by 10.25.21.151 with SMTP id 23mr2482143lfv.89.1458791363095; Wed, 23 Mar 2016 20:49:23 -0700 (PDT)
Received: by 10.25.16.160 with HTTP; Wed, 23 Mar 2016 20:49:22 -0700 (PDT)
In-Reply-To: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> 
Date: Wed, 23 Mar 2016 20:49:22 -0700
Message-ID: 
From: John Strassner 
To: Duzongpeng , John Strassner 
Content-Type: multipart/alternative; boundary=001a11406aaeffb43a052ec35406
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , "Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 03:49:29 -0000

--001a11406aaeffb43a052ec35406
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

I am confused as to why we would consider prefix-management as an autonomic
intent. To me, this breaks a fundamental rule of what is intent by
requiring the intent writer to understand too many details about the target
of the intent. The analogy (at least in the ONF and TMF) was specifying to
a taxi company that you want a taxi, but you need to take a certain route,
because you (the customer) know traffic patterns better than the taxi
driver.

Surely we could pick a more appropriate example?


regards,
John

On Wed, Mar 16, 2016 at 5:41 AM, Duzongpeng  wrote:

> Hi, Michael,
>
>
>
>          Thanks for the reply. I can understand that intent is not in the
> current milestones, and you are working on several milestones drafts now.
>
>
>
>          You are welcome to contribute to the intent draft at any time,
> and thanks in advance.
>
>
>
>          According to your statements, in my understanding, we can keep
> regarding the =E2=80=9Cintent=E2=80=9D example in =E2=80=9Cprefix-managem=
ent ID=E2=80=9D as an Autonomic
> intent.
>
>          It=E2=80=99s semantics are just like =E2=80=9Call nodes of type =
x do this; type
> y does that=E2=80=9D, so it is a high level policy.
>
>
>
>          It really helps a lot. Thanks.
>
>
>
>          I generally agree with your statement, while some other personal
> understandings exist (shown below). But perhaps they can be discussed lat=
er.
>
>
>
> Best
> Regards
>
> Zongpeng Du
>
>
>
> *From:* Michael Behringer (mbehring) [mailto:mbehring@cisco.com]
> *Sent:* Wednesday, March 16, 2016 4:17 PM
> *To:* Duzongpeng; anima@ietf.org
> *Cc:* Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang
> *Subject:* RE: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> Hi Zongpeng,
>
>
>
> Yes, this is a recurring question and topic. Let me give my view on two
> levels:
>
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> [duzongpeng] I agree that intent can coexist with traditional config.
>
> I generally agree with the policy priorities, but I have some other
> concerns.
>
> Some default policies/parameters set through =E2=80=9Cintent=E2=80=9D per=
haps are
> fundamental for the uplayer services.
>
> It should be prevented that human operator uses some CLIs to change them
> by mistake, and the uplayer services are interrupted.
>
> At least, some warning should be reported to the human operator.
> [duzongpeng]
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J  It has=
 become clear
> that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> [duzongpeng]Agree.[duzongpeng]
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D an//d people c=
an do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writing =
Intent like =E2=80=9Cnode
> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here i=
s your
> CLI=E2=80=9D. But what I describe above COULD be abused in this way. RFC7=
575 says
> this is NOT Intent.
>
>
>
> [duzongpeng]Agree. It is hard to give an exact boundary.
>
> However, I think we have two choices here.
>
> The first one is to define a policy first in every ASA, but with some ope=
n
> parameters that can be configured. What we need to do is just to distribu=
te
> the network-level parameters for that function. This way is suitable to
> some relatively static functions.
>
> The second is more advanced, in the ideal case, the human operator can
> communicate with network using natural language or something similar. Thi=
s
> way is more flexible, but harder to realize.
>
> [duzongpeng]
>
>
>
> In other words: Let=E2=80=99s not be too prescriptive, and hope that folk=
s will do
> more or less reasonable things.
>
>
>
> [duzongpeng]Agree.[duzongpeng]
>
>
>
> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly co=
ntribute,
> I=E2=80=99m just struggling with priorities at the moment (as you can see=
 by my
> relative silence over the past weeks).
>
>
>
> Michael
>
>
>
>
>
>
>
> *From:* Anima [mailto:anima-bounces@ietf.org ] *O=
n
> Behalf Of *Duzongpeng
> *Sent:* 16 March 2016 02:38
> *To:* anima@ietf.org
> *Cc:* Laurent Ciavaglia ; J=C3=A9ferson Camp=
os
> Nobre ; Sheng Jiang 
> *Subject:* [Anima] Does Autonomic network need some intervention beyond
> intent
>
>
>
> Hello, everyone
>
>
>
>          I understand that Autonomic network should be able to work with
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
>
>
>
>          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
>
>
>
>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
>
>
>
>          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
>
>
>
> Is =E2=80=9CASA parameters" ok here?
>
>
>
> I think some characteristics for this kind of low-level policies are
>
> 1.       Network level parameters, need to be distributed by GRASP and AC=
P
>
> 2.       Relatively static compared to intent, perhaps only needed to be
> configured once
>
> 3.       Unavoidable in current network environment, mostly for
> establishing the network infrastructure
>
> 4.       Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
>
>
>
> Welcome for comment. Your suggestions will be appreciated.
>
>
>
>
>
> Best Regards
>
> Zongpeng Du
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>


--=20
regards,
John

--001a11406aaeffb43a052ec35406
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


I am confused as to why we would consider prefix-mana=
gement as an autonomic intent. To me, this breaks a fundamental rule of wha=
t is intent by requiring the intent writer to understand too many details a=
bout the target of the intent. The analogy (at least in the ONF and TMF) wa=
s specifying to a taxi company that you want a taxi, but you need to take a=
 certain route, because you (the customer) know traffic patterns better tha=
n the taxi driver.

Surely we could pick a more app=
ropriate example?


regards,
John

On Wed, Mar 16, 2016 at 5:41 AM, Duzongpeng <duzongpeng@huawei.com> wrote:











=
Hi, Michael,


=
=C2=A0


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Thanks for the reply. I ca=
n understand that intent is not in the current milestones, and you are work=
ing on several milestones drafts now.


=
=C2=A0


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 You are welcome to contrib=
ute to the intent draft at any time, and thanks in advance.


=
=C2=A0


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 According to your statemen=
ts, in my understanding, we can keep regarding the =E2=80=9Cintent=E2=80=9D=
 example in =E2=80=9Cprefix-management ID=E2=80=9D as an Autonomic intent.<=
u>


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 It=E2=80=99s semantics are=
 just like
=
=E2=80=9Call nodes of type x do this; type y does that=E2=80=9D, so it is a=
 high level policy.


=
=C2=A0


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 It really helps a lot. Tha=
nks.


=
=C2=A0


=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 I generally agree with you=
r statement, while some other personal understandings exist (shown below). =
But perhaps they can be discussed later.


=
=C2=A0


=
Best Regards=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0



=
Zongpeng Du


=
=C2=A0






From: Michael Behringe=
r (mbehring)
 [mailto:mbehring@c=
isco.com] 

Sent: Wednesday, March 16, 2016 4:17 PM

To: Duzongpeng; =
anima@ietf.org

Cc: Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang

Subject: RE: [Anima] Does Autonomic network need some intervention b=
eyond intent






=C2=A0


Hi Zongpeng,



=C2=A0


Yes, this is a recurring question and topic. Let me give my =
view on two levels:



=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
<=
u>


First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluenci=
ng=E2=80=9D or =E2=80=9Csteering=E2=80=9D a network, at least not initially=
. There is still traditional config, plus all the existing and emerging SDN=
 models.
 My view has always been that ANIMA sets a =E2=80=9Cdefault=E2=80=9D policy=
 in a network, and all other means are more specific and can override inten=
t. This is described in RFC7575. Main point here: ANIMA Intent isn=E2=80=99=
t the only tool in a network.



=C2=A0


[duzongpeng] I agree that inten=
t can coexist with traditional config.



I generally agree with the poli=
cy priorities, but I have some other concerns.


Some default policies/parameters set through =E2=
=80=9Cintent=E2=80=9D perhaps are fundamental for the uplayer services.<=
/u>


It should be prevented that hum=
an operator uses some CLIs to change them by mistake, and the uplayer servi=
ces are interrupted.


At least, some warning should b=
e reported to the human operator. [duzongpeng]


=
=C2=A0


Intent could perfectly well cover a high level policy such a=
s =E2=80=9Call nodes of type x do this; type y does that=E2=80=9D. (Which I=
 think is what you mention below.) When we started with Intent,
 the vision was to keep it very high level, more along the line =E2=80=9Cdo=
 the right thing=E2=80=9D
J=C2=A0 It has become clear that in today=E2=80=99s net=
working, people will want to push more =E2=80=9Cparameters=E2=80=9D than =
=E2=80=9Cpolicy=E2=80=9D,
 and I think we=E2=80=99ll just have to acknowledge that. 


=C2=A0


Personally I think it=E2=80=99ll pan out like this: Intent w=
ill have sections per autonomic function. An Intent parser will validate ov=
erall correctness (signature, etc), then segment the
 Intent into its bits: One part will be the autonomic fabric itself, then t=
here will be bits for each autonomic function.



=C2=A0


So the Intent parser on a node will grab intent, see there i=
s some Intent for autonomic function A, will see whether this is running lo=
cally, and if so, it=E2=80=99ll just pass that piece
 of Intent to that function. 


=C2=A0


In this way, a function owner defines both the intent itself=
, and how it is consumed by his/her function.



=
=C2=A0


[duzongpeng]Agree.[duzongpeng]<=
u>


=C2=A0


The downside is that there is no =E2=80=9Ccontrol=E2=80=9D a=
n=
//d
 people can do what they want, in the worst case even push config, which is=
 clearly NOT what Intent is meant to do. I=E2=80=99d hope that people don=
=E2=80=99t start writing Intent like =E2=80=9Cnode 17: here is your CLI; no=
de 23: here is your CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D. But
 what I describe above COULD be abused in this way. RFC7575 says this is NO=
T Intent.



=C2=A0


[duzongpeng]Agree. It is hard t=
o give an exact boundary.



However, I think we have two ch=
oices here.


The first one is to define a po=
licy first in every ASA, but with some open parameters that can be configur=
ed. What we need to do is just to distribute the network-level parameters f=
or that function. This way is suitable
 to some relatively static functions. 


The second is more advanced, in=
 the ideal case, the human operator can communicate with network using natu=
ral language or something similar. This way is more flexible, but harder to=
 realize.


[duzongpeng]


=
=C2=A0


In other words: Let=E2=80=99s not be too prescriptive, and h=
ope that folks will do more or less reasonable things.



=C2=A0


[duzongpeng]Agree.[duzongpeng]<=
u>


=
=C2=A0


I=E2=80=99ve been meaning to comment on the Intent drafts an=
d possibly contribute, I=E2=80=99m just struggling with priorities at the m=
oment (as you can see by my relative silence over the past
 weeks). 


=C2=A0


Michael


=C2=A0


=C2=A0


=C2=A0








From: Anima [mailto:anima-bounces@ietf.org]
On Behalf Of Duzongpeng

Sent: 16 March 2016 02:38

To: anima@ietf.o=
rg

Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; J=C3=A9ferson C=
ampos Nobre <j=
cnobre@inf.ufrgs.br>; Sheng Jiang <jiangsheng@huawei.com>

Subject: [Anima] Does Autonomic network need some intervention beyon=
d intent






=C2=A0


Hello, everyone


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 I understand that Autonomic network should be able to wo=
rk with as less configuration as possible. In the ideal case, an Autonomic =
network should be able to work well while only intent is needed from human =
operators.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 However, I wonder whether Autonomic network may need som=
e intervention from human operators through some low-level policies beyond =
intents.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 For example, in the 
https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it i=
s suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in IP RAN) can be assigned as an "intent". These configuration pa=
rameters need to be distributed in the autonomic
 domain to influence the detail configurations on each autonomic node.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 As intent is described as an abstract, declarative, high=
-level policy used to operate an autonomic domain, such as an enterprise ne=
twork. Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cp=
refix-management ID=E2=80=9D to some
 other words.


=C2=A0


Is =
=E2=80=9CASA parameters" ok here?



<=
/u>=C2=A0


I th=
ink some characteristics for this kind of low-level policies are<=
/u>


1.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Network level parameters, need to be distribute=
d by GRASP and ACP


2.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Relatively static compared to intent, perhaps o=
nly needed to be configured once


3.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Unavoidable in current network environment, mos=
tly for establishing the network infrastructure


4.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Rarely need coordinations with others parameter=
s, usually is closed related to one specific autonomic function


=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0 


Welc=
ome for comment. Your suggestions will be appreciated.=



<=
/u>=C2=A0


<=
/u>=C2=A0


Best Regards


Zongpeng Du









_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima





-- 
regards,
John




--001a11406aaeffb43a052ec35406--


From nobody Wed Mar 23 21:52:36 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3134912D617 for ; Wed, 23 Mar 2016 21:52:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9dGn3lVp8LgV for ; Wed, 23 Mar 2016 21:52:32 -0700 (PDT)
Received: from mail-pa0-x22e.google.com (mail-pa0-x22e.google.com [IPv6:2607:f8b0:400e:c03::22e]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B819D12D60D for ; Wed, 23 Mar 2016 21:52:32 -0700 (PDT)
Received: by mail-pa0-x22e.google.com with SMTP id td3so12653997pab.2 for ; Wed, 23 Mar 2016 21:52:32 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=ACQ/v/TbQc5XkASm9lumYKlwGMST7VhYchh40+f0ajE=; b=cqzhtbz9GpXI0c4tVrAtkBNuCSuvtQAH8E0Cs+fTdlZ1cR8QbKJBwjU00Ok0Vglkz7 KZZZNf0qq7GJfTTA90L6r4kG6fE06mPFfXTXhGtp3s68pkp6vKsGXs03iEbT21u/NPH3 CX1S3IigavXGyH/jdxH+kbj3JJuuy+puTDF2XPapjAw9JCCKSPM017aZe2laEQ4QDjvQ 35SY5aTCDIvLCqTxAgFuBuowj89L+VApoS5/SmDNcqwQ8ZS5i335fBNkd9FyrT4jvQVO a0P/bZLV93DhTUAUcj36fSDunZlo0zKj9vS9sms0lkdsCCLve25SDAMXYMxvRkKLz1+H mMEQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=ACQ/v/TbQc5XkASm9lumYKlwGMST7VhYchh40+f0ajE=; b=gSlN1r9YR9FseDVwHcFdeeKqA/nUIXHhbj6YB5EPsZHtLW4hiVsNDr1G9jPQWSpKP8 qrVgeYjYrfqeSJ1sIbM+CuPFI/zN+tJxZVyIJ3ax/mRu3TAF8Vgqi0PAoIMg0jPotrsA e87c1Nw3n1WEBl2qO1ifWFssaZiRxZKVVadQVmAFiJiGQ988CxYN41EsMUnV6xKx08QH CwqXs/aVFLXxa7G5mK9dYGiwhrhu34Zr2jPgG/2hcASSIIRoNI621ALwSCodA+zIDi4k 3UtO1U1mbENARjcTfsLSbDcuEb5Bayq2vBIbRmwo4w0MVvzSGVTNAbJJHcPkyZ7swBHw my+w==
X-Gm-Message-State: AD7BkJLhUlDlblWC4pdb/NESs3pBkNdZrh+wAMY1WPm2x/lCAjjYSOmHriNzsgXvb/+adw==
X-Received: by 10.66.254.168 with SMTP id aj8mr9824486pad.18.1458795152269; Wed, 23 Mar 2016 21:52:32 -0700 (PDT)
Received: from ?IPv6:2406:e007:4457:1:28cc:dc4c:9703:6781? ([2406:e007:4457:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id 19sm7578537pfp.66.2016.03.23.21.52.27 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 23 Mar 2016 21:52:30 -0700 (PDT)
To: John Strassner , Duzongpeng 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F37293.7040707@gmail.com>
Date: Thu, 24 Mar 2016 17:52:35 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "Michael Behringer \(mbehring\)" , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 04:52:35 -0000

On 24/03/2016 16:49, John Strassner wrote:
> I am confused as to why we would consider prefix-management as an auton=
omic
> intent.=20

It's the other way round. Prefix management was chosen as an autonomic us=
e
case, so a follow-up question is: what intent is appropriate for that use=
 case?

And the suggested answer is: the ISP's intent is that certain types
of edge network should get certain sizes of prefix. How the prefixes are
actually divided up and assigned is left to the autonomic nodes.

> To me, this breaks a fundamental rule of what is intent by
> requiring the intent writer to understand too many details about the ta=
rget
> of the intent. The analogy (at least in the ONF and TMF) was specifying=
 to
> a taxi company that you want a taxi, but you need to take a certain rou=
te,
> because you (the customer) know traffic patterns better than the taxi
> driver.
>=20
> Surely we could pick a more appropriate example?

By all means describe other use cases and what sort of intent they need.

    Brian
>=20
>=20
> regards,
> John
>=20
> On Wed, Mar 16, 2016 at 5:41 AM, Duzongpeng  wro=
te:
>=20
>> Hi, Michael,
>>
>>
>>
>>          Thanks for the reply. I can understand that intent is not in =
the
>> current milestones, and you are working on several milestones drafts n=
ow.
>>
>>
>>
>>          You are welcome to contribute to the intent draft at any time=
,
>> and thanks in advance.
>>
>>
>>
>>          According to your statements, in my understanding, we can kee=
p
>> regarding the =E2=80=9Cintent=E2=80=9D example in =E2=80=9Cprefix-mana=
gement ID=E2=80=9D as an Autonomic
>> intent.
>>
>>          It=E2=80=99s semantics are just like =E2=80=9Call nodes of ty=
pe x do this; type
>> y does that=E2=80=9D, so it is a high level policy.
>>
>>
>>
>>          It really helps a lot. Thanks.
>>
>>
>>
>>          I generally agree with your statement, while some other perso=
nal
>> understandings exist (shown below). But perhaps they can be discussed =
later.
>>
>>
>>
>> Best
>> Regards
>>
>> Zongpeng Du
>>
>>
>>
>> *From:* Michael Behringer (mbehring) [mailto:mbehring@cisco.com]
>> *Sent:* Wednesday, March 16, 2016 4:17 PM
>> *To:* Duzongpeng; anima@ietf.org
>> *Cc:* Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang
>> *Subject:* RE: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>
>>
>>
>> Hi Zongpeng,
>>
>>
>>
>> Yes, this is a recurring question and topic. Let me give my view on tw=
o
>> levels:
>>
>>
>>
>>
>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D=
 or =E2=80=9Csteering=E2=80=9D a network,
>> at least not initially. There is still traditional config, plus all th=
e
>> existing and emerging SDN models. My view has always been that ANIMA s=
ets a
>> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are=
 more specific and
>> can override intent. This is described in RFC7575. Main point here: AN=
IMA
>> Intent isn=E2=80=99t the only tool in a network.
>>
>>
>>
>> [duzongpeng] I agree that intent can coexist with traditional config.
>>
>> I generally agree with the policy priorities, but I have some other
>> concerns.
>>
>> Some default policies/parameters set through =E2=80=9Cintent=E2=80=9D =
perhaps are
>> fundamental for the uplayer services.
>>
>> It should be prevented that human operator uses some CLIs to change th=
em
>> by mistake, and the uplayer services are interrupted.
>>
>> At least, some warning should be reported to the human operator.
>> [duzongpeng]
>>
>>
>>
>> Intent could perfectly well cover a high level policy such as =E2=80=9C=
all nodes
>> of type x do this; type y does that=E2=80=9D. (Which I think is what y=
ou mention
>> below.) When we started with Intent, the vision was to keep it very hi=
gh
>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J  It =
has become clear
>> that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
>> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to ackno=
wledge that.
>>
>>
>>
>> Personally I think it=E2=80=99ll pan out like this: Intent will have s=
ections per
>> autonomic function. An Intent parser will validate overall correctness=

>> (signature, etc), then segment the Intent into its bits: One part will=
 be
>> the autonomic fabric itself, then there will be bits for each autonomi=
c
>> function.
>>
>>
>>
>> So the Intent parser on a node will grab intent, see there is some Int=
ent
>> for autonomic function A, will see whether this is running locally, an=
d if
>> so, it=E2=80=99ll just pass that piece of Intent to that function.
>>
>>
>>
>> In this way, a function owner defines both the intent itself, and how =
it
>> is consumed by his/her function.
>>
>>
>>
>> [duzongpeng]Agree.[duzongpeng]
>>
>>
>>
>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D an//d peopl=
e can do what they
>> want, in the worst case even push config, which is clearly NOT what In=
tent
>> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writi=
ng Intent like =E2=80=9Cnode
>> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: her=
e is your
>> CLI=E2=80=9D. But what I describe above COULD be abused in this way. R=
FC7575 says
>> this is NOT Intent.
>>
>>
>>
>> [duzongpeng]Agree. It is hard to give an exact boundary.
>>
>> However, I think we have two choices here.
>>
>> The first one is to define a policy first in every ASA, but with some =
open
>> parameters that can be configured. What we need to do is just to distr=
ibute
>> the network-level parameters for that function. This way is suitable t=
o
>> some relatively static functions.
>>
>> The second is more advanced, in the ideal case, the human operator can=

>> communicate with network using natural language or something similar. =
This
>> way is more flexible, but harder to realize.
>>
>> [duzongpeng]
>>
>>
>>
>> In other words: Let=E2=80=99s not be too prescriptive, and hope that f=
olks will do
>> more or less reasonable things.
>>
>>
>>
>> [duzongpeng]Agree.[duzongpeng]
>>
>>
>>
>> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly=
 contribute,
>> I=E2=80=99m just struggling with priorities at the moment (as you can =
see by my
>> relative silence over the past weeks).
>>
>>
>>
>> Michael
>>
>>
>>
>>
>>
>>
>>
>> *From:* Anima [mailto:anima-bounces@ietf.org ]=
 *On
>> Behalf Of *Duzongpeng
>> *Sent:* 16 March 2016 02:38
>> *To:* anima@ietf.org
>> *Cc:* Laurent Ciavaglia ; J=C3=A9ferson C=
ampos
>> Nobre ; Sheng Jiang 
>> *Subject:* [Anima] Does Autonomic network need some intervention beyon=
d
>> intent
>>
>>
>>
>> Hello, everyone
>>
>>
>>
>>          I understand that Autonomic network should be able to work wi=
th
>> as less configuration as possible. In the ideal case, an Autonomic net=
work
>> should be able to work well while only intent is needed from human
>> operators.
>>
>>
>>
>>          However, I wonder whether Autonomic network may need some
>> intervention from human operators through some low-level policies beyo=
nd
>> intents.
>>
>>
>>
>>          For example, in the
>> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it =
is
>> suggested that the prefix lengths for the CSG, ASG, RSG (different rol=
es in
>> IP RAN) can be assigned as an "intent". These configuration parameters=
 need
>> to be distributed in the autonomic domain to influence the detail
>> configurations on each autonomic node.
>>
>>
>>
>>          As intent is described as an abstract, declarative, high-leve=
l
>> policy used to operate an autonomic domain, such as an enterprise netw=
ork.
>> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cpre=
fix-management ID=E2=80=9D to some
>> other words.
>>
>>
>>
>> Is =E2=80=9CASA parameters" ok here?
>>
>>
>>
>> I think some characteristics for this kind of low-level policies are
>>
>> 1.       Network level parameters, need to be distributed by GRASP and=
 ACP
>>
>> 2.       Relatively static compared to intent, perhaps only needed to =
be
>> configured once
>>
>> 3.       Unavoidable in current network environment, mostly for
>> establishing the network infrastructure
>>
>> 4.       Rarely need coordinations with others parameters, usually is
>> closed related to one specific autonomic function
>>
>>
>>
>> Welcome for comment. Your suggestions will be appreciated.
>>
>>
>>
>>
>>
>> Best Regards
>>
>> Zongpeng Du
>>
>> _______________________________________________
>> Anima mailing list
>> Anima@ietf.org
>> https://www.ietf.org/mailman/listinfo/anima
>>
>>
>=20
>=20
>=20
>=20
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>=20


From nobody Wed Mar 23 23:58:45 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B3A2212D177 for ; Wed, 23 Mar 2016 23:58:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9gLYanvm14Q9 for ; Wed, 23 Mar 2016 23:58:39 -0700 (PDT)
Received: from mail-lb0-x231.google.com (mail-lb0-x231.google.com [IPv6:2a00:1450:4010:c04::231]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D354612D146 for ; Wed, 23 Mar 2016 23:58:38 -0700 (PDT)
Received: by mail-lb0-x231.google.com with SMTP id oe12so24496635lbc.0 for ; Wed, 23 Mar 2016 23:58:38 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=UbT0Q8ARx9croelZFfo7GNaf/pCdACpmBt74ut69v20=; b=b86CCAMBMDmtyZfMNxuuoEuBElmCBSSQir3FhHWlLhA3xC75NKoYw+NvkzfogDS4Ss GGG7FM1M6TKAyMjC2aT4Dgwi1K3TIVhFH/VLXV5QHMqSGlBUui6ByqO60xVzRXndG0AN 2efb2Au6gp5Q16baCIwCB5nyguR+17hFjOqSRhj6EOg429L0qmDvWFAZV1/WGHa+fIk3 7gc9GbBZHCiFSPbmw2TfToO33K5Y2H4nqg1bkbaawLpbm3KCsuWx8LMx27TU4KlXRVHD ixo+2HVqeoGNBuSRfUhlC4ITcpkIqM3dHM+jSU6Ul2BiCTw4RdziKySrcnqnyf9wynrt 9TbQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=UbT0Q8ARx9croelZFfo7GNaf/pCdACpmBt74ut69v20=; b=MvdYuztRFZ1FpItYUWLbd4ptTua7Tsaz5g0W3eSebZW+8HSGmh7b/jgFXfknv2rbpW OUIrWtyZIdbbpvk1yGvQ9oaSftbD7ZG+L7owImox0j5RHns/P9iNIR/TQXs7/sMdmph6 lSOo32cOvWJ23wchr2RjF9NyZEWYMLoYBHNk2/Rkis3xzVF68T8DLy45sfspCRnQVgxG MzGqIHk+EdcPieLEPsKhpJcZGuLjRHFP8MHvYdTF+voA4dagbFDlhnAP/R6yxAcTuVeS yIUwcJ8kUOU2L2z68Z8Yg4LkaQKNJr3ydiuMj3UBYAJU0jNtEYUHfKH/nW4Asp8kQM5S 5sOw==
X-Gm-Message-State: AD7BkJIFlOlMmKYur1BeFvLH/M3o6q5ADG/z2NXDgmEXujiXtsvFUnwRn8RLvGzJRLVgg6GUXocR8WaJe4w3rQ==
MIME-Version: 1.0
X-Received: by 10.112.170.68 with SMTP id ak4mr2247104lbc.94.1458802716972; Wed, 23 Mar 2016 23:58:36 -0700 (PDT)
Received: by 10.25.16.160 with HTTP; Wed, 23 Mar 2016 23:58:36 -0700 (PDT)
In-Reply-To: <782ad410ef844630a942c27f6279a49f@XCH-RCD-006.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com> <782ad410ef844630a942c27f6279a49f@XCH-RCD-006.cisco.com>
Date: Wed, 23 Mar 2016 23:58:36 -0700
Message-ID: 
From: John Strassner 
To: "Michael Behringer (mbehring)" , John Strassner 
Content-Type: multipart/alternative; boundary=001a11c259a8be1ed7052ec5f97c
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 06:58:44 -0000

--001a11c259a8be1ed7052ec5f97c
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

>> As Michael's yang outline showed, there will be network-wide aspects
>> to intent and also ASA-specific aspects. I think that in practice there
>> will be simple parameters in the ASA-specific parts, so (whether we
>> use yang or some other syntax) I see no advantage in having a
>> separate mechanism for the parameters - in a sense we never need
>> to answer the question, we simply treat all the information in the
>> same way, intent or parameter.

> Agree so far.

Sorry, I'm confused. Are we talking about intent having **parameters**?
Or are we talking about the output of the intent compiler generating
functions with parameters? Hopefully the latter...


>> Maybe we should think about levels of specialization. Top level,
>> network intent. Next level, intent for all ASAs supporting a given
>> objective. Those two levels would be flooded by GRASP. Next
>> level, intent for a single ASA. That would be transmitted by
>> unicast GRASP synchronization.

> Here I don't have the same view. The first two levels we agree on.
> Anything that's for a specific node doesn't qualify as "Intent" in my
> view. (Ignore that the flooding mechanism may contain unicast
> transmissions as part of the overall scheme).

Referring to the policy continuum, which I defined a long time ago
(let me know if you want references), there is a set of translations
from the most abstract level to a form suitable for consumption by
a device. This is because there are many different constituencies
that all **should be able to** work together, but have different
terminologies and concepts.

I'm not sure starting with "network intent" at the highest level is
correct. I'm not necessarily arguing for my original formulation
(which had business at the top and device implementation at the
bottom), but surely the greatest need for intent is for non *CIE
(25 cents to Michael) types?


> If Intent is a high-level policy (which I think we all agree on?)
> then I can think of two more "things" that are NOT Intent, to
> add to my list of "not-Intent":
>
> - anything that's for a specific node

Hmm. I **think** what you mean is if a specific node is directly
referenced in the intent statement, correct? (As opposed to if,
after the intent is compiled, the target is exactly one node.)

> - anything that's negotiated dynamically. (Intent is always
> originating from something non-technical - it's a policy).

Two observations:

    1) If it is non-technical (which I agree with), why are we
        talking about prefix-management? :-)
    2) This was a LONG, yet PAINFUL, discussion in both the
        ONF and the TMF. The specific counter-example used
        in the ONF was negotiating a taxi fare. One in the TMF
        was negotiating QoS and security. I think we need to
        be careful here about disallowing negotiation.

> - anything that needs to change frequently (high-level
> policy is normally pretty stable)

Agreed!

 > - anything that's not a "policy", ie doesn't come from a
> non-technical reasoning. (that's not a good way to define it,
> but can't think of a better one right now)

May be we should save this one for BA...

> I think you're suggesting an optimisation of Intent
> *distribution*, Brian, right? Where, if I know only a few
> nodes run a certain function, I just send that particular Intent
> to those nodes only? (is that what you're thinking of?)
>
> This would certainly be tempting, but I think it's a slippery
> slope, and would quickly get into conflict with current policy
> models. I'd prefer if we keep "our" Intent as "must be flooded
> everywhere". If it is really high level guidance, the optimization
> shouldn't be needed. Said differently: If we start pushing huge
> data models, maybe we're on the wrong track in the first place?

I think that the intent writer should be blissfully unaware of any
topology or specific autonomic function. It is up to the management
system (intent compiler plus appropriate logic) to figure out what
is meant by the intent, and where to push the intent.

Pushing huge data models seems unlikely at best. One exception
could be when an entire domain of diverse functions joins an
existing set of domains, AND the domains are expected to
interoperate and collaborate. I would view this as an exception,
and not the rule.

regards,
John

On Mon, Mar 21, 2016 at 2:31 AM, Michael Behringer (mbehring) <
mbehring@cisco.com> wrote:

> > -----Original Message-----
> > From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
> > Sent: 20 March 2016 20:17
> > To: Duzongpeng ; anima@ietf.org
> > Cc: Laurent Ciavaglia ; J=C3=A9ferson Camp=
os
> Nobre
> > ; Sheng Jiang ; Michael
> > Behringer (mbehring) 
> > Subject: Re: [Anima] Does Autonomic network need some intervention
> > beyond intent
> >
> > Hi Zongpeng,
> >
> > > Should these ASA parameters be part of ANIMA's work items just as
> > intent?
> >
> > As Michael's yang outline showed, there will be network-wide aspects to
> > intent and also ASA-specific aspects. I think that in practice there
> will be
> > simple parameters in the ASA-specific parts, so (whether we use yang or
> > some other syntax) I see no advantage in having a separate mechanism fo=
r
> > the parameters - in a sense we never need to answer the question, we
> > simply treat all the information in the same way, intent or parameter.
>
> Agree so far.
>
> > Maybe we should think about levels of specialization. Top level, networ=
k
> > intent. Next level, intent for all ASAs supporting a given objective.
> > Those two levels would be flooded by GRASP. Next level, intent for a
> single
> > ASA. That would be transmitted by unicast GRASP synchronization.
>
> Here I don't have the same view. The first two levels we agree on.
> Anything that's for a specific node doesn't qualify as "Intent" in my vie=
w.
> (Ignore that the flooding mechanism may contain unicast transmissions as
> part of the overall scheme).
>
> If Intent is a high-level policy (which I think we all agree on?) then I
> can think of two more "things" that are NOT Intent, to add to my list of
> "not-Intent":
>
> - anything that's for a specific node
> - anything that's negotiated dynamically. (Intent is always originating
> from something non-technical - it's a policy).
> - anything that needs to change frequently (high-level policy is normally
> pretty stable)
> - anything that's not a "policy", ie doesn't come from a non-technical
> reasoning. (that's not a good way to define it, but can't think of a bett=
er
> one right now)
>
> I think you're suggesting an optimisation of Intent *distribution*, Brian=
,
> right? Where, if I know only a few nodes run a certain function, I just
> send that particular Intent to those nodes only? (is that what you're
> thinking of?)
>
> This would certainly be tempting, but I think it's a slippery slope, and
> would quickly get into conflict with current policy models. I'd prefer if
> we keep "our" Intent as "must be flooded everywhere". If it is really hig=
h
> level guidance, the optimization shouldn't be needed. Said differently: I=
f
> we start pushing huge data models, maybe we're on the wrong track in the
> first place?
>
> Michael
>
> > Regards
> >    Brian
> >
> > On 18/03/2016 21:33, Duzongpeng wrote:
> > > Hi, Brian
> > >
> > >     Thanks for the response. I think we have agreed that "prefix-
> > management ID" can be an Autonomic intent.
> > >
> > >     However, some ASA parameters may appear to be more static, and
> > perhaps can fulfill the statement below:
> > >     1. Network level parameters, need to be distributed by GRASP and
> > ACP
> > >     2. Relatively static compared to intent, perhaps only needed to b=
e
> > configured once
> > >     3. Unavoidable in current network environment, mostly for
> > establishing the network infrastructure
> > >     4. Rarely need coordinations with others parameters, usually is
> > > closed related to one specific autonomic function
> > >
> > >     Some examples can be
> > >     What kind of IGP the Autonomic network domain should use (such as
> > IS-IS, OSPF);
> > >     What kind of transport layer network technology the Autonomic
> > network
> > > domain should use (such as MPLS, IP)
> > >
> > >     Should these ASA parameters be part of ANIMA's work items just as
> > intent?
> > >
> > > Best Regards
> > > Zongpeng Du
> > >
> > > -----Original Message-----
> > > From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
> > > Sent: Thursday, March 17, 2016 2:41 AM
> > > To: Duzongpeng; Michael Behringer (mbehring); anima@ietf.org
> > > Cc: Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang
> > > Subject: Re: [Anima] Does Autonomic network need some intervention
> > > beyond intent
> > >
> > > Yes, I think that whatever the "philosophy" of intent is, we should
> make
> > sure that the implementation is pragmatic and that we can adapt to
> > operator's requirements.
> > >
> > > Regards
> > >    Brian
> > >
> > > On 17/03/2016 01:41, Duzongpeng wrote:
> > >> Hi, Michael,
> > >>
> > >>          Thanks for the reply. I can understand that intent is not i=
n
> the current
> > milestones, and you are working on several milestones drafts now.
> > >>
> > >>          You are welcome to contribute to the intent draft at any
> time, and
> > thanks in advance.
> > >>
> > >>          According to your statements, in my understanding, we can
> keep
> > regarding the "intent" example in "prefix-management ID" as an Autonomi=
c
> > intent.
> > >>          It's semantics are just like "all nodes of type x do this;
> type y does
> > that", so it is a high level policy.
> > >>
> > >>          It really helps a lot. Thanks.
> > >>
> > >>          I generally agree with your statement, while some other
> personal
> > understandings exist (shown below). But perhaps they can be discussed
> > later.
> > >>
> > >> Best Regards
> > >> Zongpeng Du
> > >>
> > >> From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com]
> > >> Sent: Wednesday, March 16, 2016 4:17 PM
> > >> To: Duzongpeng; anima@ietf.org
> > >> Cc: Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang
> > >> Subject: RE: [Anima] Does Autonomic network need some intervention
> > >> beyond intent
> > >>
> > >> Hi Zongpeng,
> > >>
> > >> Yes, this is a recurring question and topic. Let me give my view on
> two
> > levels:
> > >>
> > >> First, ANIMA isn't the only thing "influencing" or "steering" a
> network, at
> > least not initially. There is still traditional config, plus all the
> existing and
> > emerging SDN models. My view has always been that ANIMA sets a "default=
"
> > policy in a network, and all other means are more specific and can
> override
> > intent. This is described in RFC7575. Main point here: ANIMA Intent
> isn't the
> > only tool in a network.
> > >>
> > >> [duzongpeng] I agree that intent can coexist with traditional config=
.
> > >> I generally agree with the policy priorities, but I have some other
> > concerns.
> > >> Some default policies/parameters set through "intent" perhaps are
> > fundamental for the uplayer services.
> > >> It should be prevented that human operator uses some CLIs to change
> > them by mistake, and the uplayer services are interrupted.
> > >> At least, some warning should be reported to the human operator.
> > >> [duzongpeng]
> > >>
> > >> Intent could perfectly well cover a high level policy such as "all
> nodes of
> > type x do this; type y does that". (Which I think is what you mention
> below.)
> > When we started with Intent, the vision was to keep it very high level,
> more
> > along the line "do the right thing" :)  It has become clear that in
> today's
> > networking, people will want to push more "parameters" than "policy",
> and I
> > think we'll just have to acknowledge that.
> > >>
> > >> Personally I think it'll pan out like this: Intent will have section=
s
> per
> > autonomic function. An Intent parser will validate overall correctness
> > (signature, etc), then segment the Intent into its bits: One part will
> be the
> > autonomic fabric itself, then there will be bits for each autonomic
> function.
> > >>
> > >> So the Intent parser on a node will grab intent, see there is some
> Intent
> > for autonomic function A, will see whether this is running locally, and
> if so, it'll
> > just pass that piece of Intent to that function.
> > >>
> > >> In this way, a function owner defines both the intent itself, and ho=
w
> it is
> > consumed by his/her function.
> > >>
> > >> [duzongpeng]Agree.[duzongpeng]
> > >>
> > >> The downside is that there is no "control" an//d people can do what
> they
> > want, in the worst case even push config, which is clearly NOT what
> Intent is
> > meant to do. I'd hope that people don't start writing Intent like "node
> 17:
> > here is your CLI; node 23: here is your CLI"; node 37: here is your
> CLI". But
> > what I describe above COULD be abused in this way. RFC7575 says this is
> NOT
> > Intent.
> > >>
> > >> [duzongpeng]Agree. It is hard to give an exact boundary.
> > >> However, I think we have two choices here.
> > >> The first one is to define a policy first in every ASA, but with som=
e
> open
> > parameters that can be configured. What we need to do is just to
> distribute
> > the network-level parameters for that function. This way is suitable to
> some
> > relatively static functions.
> > >> The second is more advanced, in the ideal case, the human operator c=
an
> > communicate with network using natural language or something similar.
> This
> > way is more flexible, but harder to realize.
> > >> [duzongpeng]
> > >>
> > >> In other words: Let's not be too prescriptive, and hope that folks
> will do
> > more or less reasonable things.
> > >>
> > >> [duzongpeng]Agree.[duzongpeng]
> > >>
> > >> I've been meaning to comment on the Intent drafts and possibly
> > contribute, I'm just struggling with priorities at the moment (as you
> can see
> > by my relative silence over the past weeks).
> > >>
> > >> Michael
> > >>
> > >>
> > >>
> > >> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng
> > >> Sent: 16 March 2016 02:38
> > >> To: anima@ietf.org
> > >> Cc: Laurent Ciavaglia
> > >> >;
> > >> J=C3=A9ferson Campos Nobre
> > >> >; Sheng Jiang
> > >> >
> > >> Subject: [Anima] Does Autonomic network need some intervention
> > beyond
> > >> intent
> > >>
> > >> Hello, everyone
> > >>
> > >>          I understand that Autonomic network should be able to work
> with as
> > less configuration as possible. In the ideal case, an Autonomic network
> > should be able to work well while only intent is needed from human
> > operators.
> > >>
> > >>          However, I wonder whether Autonomic network may need some
> > intervention from human operators through some low-level policies beyon=
d
> > intents.
> > >>
> > >>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-
> > prefix-management-00, it is suggested that the prefix lengths for the
> CSG,
> > ASG, RSG (different roles in IP RAN) can be assigned as an "intent".
> These
> > configuration parameters need to be distributed in the autonomic domain
> to
> > influence the detail configurations on each autonomic node.
> > >>
> > >>          As intent is described as an abstract, declarative,
> high-level policy
> > used to operate an autonomic domain, such as an enterprise network.
> > Perhaps we can rename the "intent" in the "prefix-management ID" to som=
e
> > other words.
> > >>
> > >> Is "ASA parameters" ok here?
> > >>
> > >> I think some characteristics for this kind of low-level policies are
> > >>
> > >> 1.       Network level parameters, need to be distributed by GRASP a=
nd
> > ACP
> > >>
> > >> 2.       Relatively static compared to intent, perhaps only needed t=
o
> be
> > configured once
> > >>
> > >> 3.       Unavoidable in current network environment, mostly for
> > establishing the network infrastructure
> > >>
> > >> 4.       Rarely need coordinations with others parameters, usually i=
s
> closed
> > related to one specific autonomic function
> > >>
> > >> Welcome for comment. Your suggestions will be appreciated.
> > >>
> > >>
> > >> Best Regards
> > >> Zongpeng Du
> > >>
> > >>
> > >>
> > >> _______________________________________________
> > >> Anima mailing list
> > >> Anima@ietf.org
> > >> https://www.ietf.org/mailman/listinfo/anima
> > >>
> > >
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>



--=20
regards,
John

--001a11c259a8be1ed7052ec5f97c
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


>> As Michael's yang outline showed, there =
will be network-wide aspects
>> to intent and also ASA-specific as=
pects. I think that in practice there
>> will be=C2=A0simple param=
eters in the ASA-specific parts, so (whether we
>> use yang or=C2=
=A0some other syntax) I see no advantage in having a
>> sep=
arate mechanism for=C2=A0the parameters - in a sense we never need
>&=
gt; to answer the question, we=C2=A0simply treat all the information in the=

>> same way, intent or parameter.

 > Agree so fa=
r.

Sorry, I'm confused. Are we talking about i=
ntent having **parameters**?
Or are we talking about the output o=
f the intent compiler generating
functions with parameters? Hopef=
ully the latter...


>> Maybe we should think about l=
evels of specialization. Top level,
>> network intent. Next=
 level, intent for all ASAs supporting a given
>> objective.=C2=A0=
Those two levels would be flooded by GRASP. Next
>> level, intent =
for a single=C2=A0ASA. That would be transmitted by
>> unic=
ast GRASP synchronization.

>=C2=A0Here I don't have the same =
view. The first two levels we agree on.
> Anything that's =
for a specific node doesn't qualify as "Intent" in my
>=C2=A0view. (Ignore that the flooding mechanism may contain unicast<=
/div>
>=C2=A0transmissions as part of the overall scheme).

Referring to the policy continuum, which I defined=C2=A0a=
=C2=A0long time ago
(let me know if you want references), there i=
s a set of translations
from the most abstract=C2=A0level to a fo=
rm suitable for consumption by
a device. This is because there ar=
e many different constituencies
that all **should be able to** wo=
rk together, but have different
terminologies and concepts.
=


I'm not sure starting with=C2=A0"network inten=
t" at the highest level is
correct. I'm not necessarily =
arguing for my original formulation
(which had business at the to=
p and device implementation at the
bottom), but=C2=A0surely the g=
reatest need for intent is for non *CIE
(25 cents to Michael) typ=
es?


>=C2=A0If Intent is a high-level policy (which I t=
hink we all agree on?)
>=C2=A0then I can think of two more &qu=
ot;things" that are NOT Intent, to
>=C2=A0add to my list =
of "not-Intent":
>
 > - anything that's for a spe=
cific node

Hmm. I **think** what you mean is if a=
=C2=A0specific node is directly
referenced in the intent statemen=
t, correct? (As opposed to if,
after the intent is compiled, the =
target is exactly one node.)

>=C2=A0- anything that's =
negotiated dynamically. (Intent is always
>=C2=A0originating f=
rom something non-technical - it's a policy).

=
Two observations:

=C2=A0=C2=A0=C2=A0 1) If it is n=
on-technical (which I agree with), why are we
=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 talking about prefix-management? :-)
=C2=
=A0=C2=A0=C2=A0 2) This was a LONG, yet PAINFUL, discussion in both the
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 ONF and the TMF. The spec=
ific counter-example used
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 in the=C2=A0ONF was negotiating a taxi=C2=A0fare. One in the TMF
<=
div>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 was negotiating QoS and secu=
rity. I think we need to
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 be careful here about disallowing negotiation.

>=C2=A0=
- anything that needs to change frequently (high-level
>=C2=A0=
policy is normally pretty stable)

Agreed!

=C2=A0> - anything that's not a "policy", ie doesn&#=
39;t come from a
> non-technical reasoning. (that's not a =
good way to define it,
>=C2=A0but can't think of a better =
one right now)

May be we should save this one for =
BA...

>=C2=A0I think you're suggesting an optimisation=
 of Intent
>=C2=A0*distribution*, Brian, right? Where, if I kn=
ow only a few
>=C2=A0nodes run a certain function, I just send=
 that particular Intent
>=C2=A0to those nodes only? (is that w=
hat you're thinking of?)
>
 > This would certainly be tempt=
ing, but I think it's a slippery
>=C2=A0slope, and would q=
uickly get into conflict with current policy
>=C2=A0models. I&=
#39;d prefer if we keep "our" Intent as "must be flooded
>=C2=A0everywhere". If it is really high level guidance, the=
 optimization
>=C2=A0shouldn't be needed. Said differently=
: If we start pushing huge
>=C2=A0data models, maybe we're=
 on the wrong track in the first place?

I think th=
at the intent writer should be blissfully unaware of any
topology=
 or specific autonomic function. It is up to the management
syste=
m (intent compiler plus appropriate logic) to figure out what
is =
meant by the intent, and where to push the intent.

Pushing huge data models seems unlikely at best. One exception
c=
ould be when an entire domain of diverse functions joins an
exist=
ing set of domains, AND the domains are expected to
interoperate =
and collaborate. I would view this as an exception,
and not the r=
ule.

regards,
John

On Mon, Mar 21, 2016 at=
 2:31 AM, Michael Behringer (mbehring) <mbehring@cisco.com>=
 wrote:
> -----Original Message-----
> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]

> Sent: 20 March 2016 20:17

> To: Duzongpeng <duzongpeng=
@huawei.com>; anima@ietf.org
> Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; J=C3=A9ferson Campos Nobre

> <jcnobre@inf.ufrgs.br&g=
t;; Sheng Jiang <jiangsheng@hua=
wei.com>; Michael

> Behringer (mbehring) <mbehrin=
g@cisco.com>

> Subject: Re: [Anima] Does Autonomic network need some intervention

> beyond intent

>

> Hi Zongpeng,

>

> > Should these ASA parameters be part of ANIMA's work items jus=
t as

> intent?

>

> As Michael's yang outline showed, there will be network-wide aspec=
ts to

> intent and also ASA-specific aspects. I think that in practice there w=
ill be

> simple parameters in the ASA-specific parts, so (whether we use yang o=
r

> some other syntax) I see no advantage in having a separate mechanism f=
or

> the parameters - in a sense we never need to answer the question, we
> simply treat all the information in the same way, intent or parameter.=




Agree so far.



> Maybe we should think about levels of specialization. Top level, netwo=
rk

> intent. Next level, intent for all ASAs supporting a given objective.<=
br>
> Those two levels would be flooded by GRASP. Next level, intent for a s=
ingle

> ASA. That would be transmitted by unicast GRASP synchronization.



Here I don't have the same view. The first two levels we agree on. Anyt=
hing that's for a specific node doesn't qualify as "Intent&quo=
t; in my view. (Ignore that the flooding mechanism may contain unicast tran=
smissions as part of the overall scheme).



If Intent is a high-level policy (which I think we all agree on?) then I ca=
n think of two more "things" that are NOT Intent, to add to my li=
st of "not-Intent":



- anything that's for a specific node

- anything that's negotiated dynamically. (Intent is always originating=
 from something non-technical - it's a policy).

- anything that needs to change frequently (high-level policy is normally p=
retty stable)

- anything that's not a "policy", ie doesn't come from a =
non-technical reasoning. (that's not a good way to define it, but can&#=
39;t think of a better one right now)



I think you're suggesting an optimisation of Intent *distribution*, Bri=
an, right? Where, if I know only a few nodes run a certain function, I just=
 send that particular Intent to those nodes only? (is that what you're =
thinking of?)



This would certainly be tempting, but I think it's a slippery slope, an=
d would quickly get into conflict with current policy models. I'd prefe=
r if we keep "our" Intent as "must be flooded everywhere&quo=
t;. If it is really high level guidance, the optimization shouldn't be =
needed. Said differently: If we start pushing huge data models, maybe we=
9;re on the wrong track in the first place?



Michael



> Regards

>=C2=A0 =C2=A0 Brian

>

> On 18/03/2016 21:33, Duzongpeng wrote:

> > Hi, Brian

> >

> >=C2=A0 =C2=A0 =C2=A0Thanks for the response. I think we have agree=
d that "prefix-

> management ID" can be an Autonomic intent.

> >

> >=C2=A0 =C2=A0 =C2=A0However, some ASA parameters may appear to be =
more static, and

> perhaps can fulfill the statement below:

> >=C2=A0 =C2=A0 =C2=A01. Network level parameters, need to be distri=
buted by GRASP and

> ACP

> >=C2=A0 =C2=A0 =C2=A02. Relatively static compared to intent, perha=
ps only needed to be

> configured once

> >=C2=A0 =C2=A0 =C2=A03. Unavoidable in current network environment,=
 mostly for

> establishing the network infrastructure

> >=C2=A0 =C2=A0 =C2=A04. Rarely need coordinations with others param=
eters, usually is

> > closed related to one specific autonomic function

> >

> >=C2=A0 =C2=A0 =C2=A0Some examples can be

> >=C2=A0 =C2=A0 =C2=A0What kind of IGP the Autonomic network domain =
should use (such as

> IS-IS, OSPF);

> >=C2=A0 =C2=A0 =C2=A0What kind of transport layer network technolog=
y the Autonomic

> network

> > domain should use (such as MPLS, IP)

> >

> >=C2=A0 =C2=A0 =C2=A0Should these ASA parameters be part of ANIMA&#=
39;s work items just as

> intent?

> >

> > Best Regards

> > Zongpeng Du

> >

> > -----Original Message-----

> > From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]

> > Sent: Thursday, March 17, 2016 2:41 AM

> > To: Duzongpeng; Michael Behringer (mbehring); anima@ietf.org

> > Cc: Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang
> > Subject: Re: [Anima] Does Autonomic network need some interventio=
n

> > beyond intent

> >

> > Yes, I think that whatever the "philosophy" of intent i=
s, we should make

> sure that the implementation is pragmatic and that we can adapt to

> operator's requirements.

> >

> > Regards

> >=C2=A0 =C2=A0 Brian

> >

> > On 17/03/2016 01:41, Duzongpeng wrote:

> >> Hi, Michael,

> >>

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 Thanks for the reply. I can=
 understand that intent is not in the current

> milestones, and you are working on several milestones drafts now.

> >>

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 You are welcome to contribu=
te to the intent draft at any time, and

> thanks in advance.

> >>

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 According to your statement=
s, in my understanding, we can keep

> regarding the "intent" example in "prefix-management ID=
" as an Autonomic

> intent.

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 It's semantics are just=
 like "all nodes of type x do this; type y does

> that", so it is a high level policy.

> >>

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 It really helps a lot. Than=
ks.

> >>

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 I generally agree with your=
 statement, while some other personal

> understandings exist (shown below). But perhaps they can be discussed<=
br>
> later.

> >>

> >> Best Regards

> >> Zongpeng Du

> >>

> >> From: Michael Behringer (mbehring) [mailto:mbehring@cisco.com]

> >> Sent: Wednesday, March 16, 2016 4:17 PM

> >> To: Duzongpeng; anima@ietf.=
org

> >> Cc: Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jian=
g

> >> Subject: RE: [Anima] Does Autonomic network need some interve=
ntion

> >> beyond intent

> >>

> >> Hi Zongpeng,

> >>

> >> Yes, this is a recurring question and topic. Let me give my v=
iew on two

> levels:

> >>

> >> First, ANIMA isn't the only thing "influencing"=
 or "steering" a network, at

> least not initially. There is still traditional config, plus all the e=
xisting and

> emerging SDN models. My view has always been that ANIMA sets a "d=
efault"

> policy in a network, and all other means are more specific and can ove=
rride

> intent. This is described in RFC7575. Main point here: ANIMA Intent is=
n't the

> only tool in a network.

> >>

> >> [duzongpeng] I agree that intent can coexist with traditional=
 config.

> >> I generally agree with the policy priorities, but I have some=
 other

> concerns.

> >> Some default policies/parameters set through "intent&quo=
t; perhaps are

> fundamental for the uplayer services.

> >> It should be prevented that human operator uses some CLIs to =
change

> them by mistake, and the uplayer services are interrupted.

> >> At least, some warning should be reported to the human operat=
or.

> >> [duzongpeng]

> >>

> >> Intent could perfectly well cover a high level policy such as=
 "all nodes of

> type x do this; type y does that". (Which I think is what you men=
tion below.)

> When we started with Intent, the vision was to keep it very high level=
, more

> along the line "do the right thing" :)=C2=A0 It has become c=
lear that in today's

> networking, people will want to push more "parameters" than =
"policy", and I

> think we'll just have to acknowledge that.

> >>

> >> Personally I think it'll pan out like this: Intent will h=
ave sections per

> autonomic function. An Intent parser will validate overall correctness=


> (signature, etc), then segment the Intent into its bits: One part will=
 be the

> autonomic fabric itself, then there will be bits for each autonomic fu=
nction.

> >>

> >> So the Intent parser on a node will grab intent, see there is=
 some Intent

> for autonomic function A, will see whether this is running locally, an=
d if so, it'll

> just pass that piece of Intent to that function.

> >>

> >> In this way, a function owner defines both the intent itself,=
 and how it is

> consumed by his/her function.

> >>

> >> [duzongpeng]Agree.[duzongpeng]

> >>

> >> The downside is that there is no "control" an//d pe=
ople can do what they

> want, in the worst case even push config, which is clearly NOT what In=
tent is

> meant to do. I'd hope that people don't start writing Intent l=
ike "node 17:

> here is your CLI; node 23: here is your CLI"; node 37: here is yo=
ur CLI". But

> what I describe above COULD be abused in this way. RFC7575 says this i=
s NOT

> Intent.

> >>

> >> [duzongpeng]Agree. It is hard to give an exact boundary.

> >> However, I think we have two choices here.

> >> The first one is to define a policy first in every ASA, but w=
ith some open

> parameters that can be configured. What we need to do is just to distr=
ibute

> the network-level parameters for that function. This way is suitable t=
o some

> relatively static functions.

> >> The second is more advanced, in the ideal case, the human ope=
rator can

> communicate with network using natural language or something similar. =
This

> way is more flexible, but harder to realize.

> >> [duzongpeng]

> >>

> >> In other words: Let's not be too prescriptive, and hope t=
hat folks will do

> more or less reasonable things.

> >>

> >> [duzongpeng]Agree.[duzongpeng]

> >>

> >> I've been meaning to comment on the Intent drafts and pos=
sibly

> contribute, I'm just struggling with priorities at the moment (as =
you can see

> by my relative silence over the past weeks).

> >>

> >> Michael

> >>

> >>

> >>

> >> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng

> >> Sent: 16 March 2016 02:38

> >> To: anima@ietf.org<m=
ailto:anima@ietf.org>

> >> Cc: Laurent Ciavaglia

> >> <laurent.ci=
avaglia@nokia.com<mailto:laurent.ciavaglia@nokia.com>>;

> >> J=C3=A9ferson Campos Nobre

> >> <jcnobre@inf.ufrgs=
.br<mailto:jcnobre@inf.ufrgs=
.br>>; Sheng Jiang

> >> <jiangsheng@huawe=
i.com<mailto:jiangsheng@hua=
wei.com>>

> >> Subject: [Anima] Does Autonomic network need some interventio=
n

> beyond

> >> intent

> >>

> >> Hello, everyone

> >>

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 I understand that Autonomic=
 network should be able to work with as

> less configuration as possible. In the ideal case, an Autonomic networ=
k

> should be able to work well while only intent is needed from human

> operators.

> >>

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 However, I wonder whether A=
utonomic network may need some

> intervention from human operators through some low-level policies beyo=
nd

> intents.

> >>

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 For example, in the https://tools.ietf.org/html/draft-ietf-anima-

> prefix-management-00, it is suggested that the prefix lengths for the =
CSG,

> ASG, RSG (different roles in IP RAN) can be assigned as an "inten=
t". These

> configuration parameters need to be distributed in the autonomic domai=
n to

> influence the detail configurations on each autonomic node.

> >>

> >>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 As intent is described as a=
n abstract, declarative, high-level policy

> used to operate an autonomic domain, such as an enterprise network.
> Perhaps we can rename the "intent" in the "prefix-manag=
ement ID" to some

> other words.

> >>

> >> Is "ASA parameters" ok here?

> >>

> >> I think some characteristics for this kind of low-level polic=
ies are

> >>

> >> 1.=C2=A0 =C2=A0 =C2=A0 =C2=A0Network level parameters, need t=
o be distributed by GRASP and

> ACP

> >>

> >> 2.=C2=A0 =C2=A0 =C2=A0 =C2=A0Relatively static compared to in=
tent, perhaps only needed to be

> configured once

> >>

> >> 3.=C2=A0 =C2=A0 =C2=A0 =C2=A0Unavoidable in current network e=
nvironment, mostly for

> establishing the network infrastructure

> >>

> >> 4.=C2=A0 =C2=A0 =C2=A0 =C2=A0Rarely need coordinations with o=
thers parameters, usually is closed

> related to one specific autonomic function

> >>

> >> Welcome for comment. Your suggestions will be appreciated.
> >>

> >>

> >> Best Regards

> >> Zongpeng Du

> >>

> >>

> >>

> >> _______________________________________________

> >> Anima mailing list

> >> Anima@ietf.org

> >> https://www.ietf.org/mailman/listinfo/anima=


> >>

> >



_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima





-- 
regards,
John




--001a11c259a8be1ed7052ec5f97c--


From nobody Thu Mar 24 00:59:32 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 853CA12D720 for ; Thu, 24 Mar 2016 00:59:31 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qPH4oJGJXpIu for ; Thu, 24 Mar 2016 00:59:29 -0700 (PDT)
Received: from rcdn-iport-3.cisco.com (rcdn-iport-3.cisco.com [173.37.86.74]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6988612D0C6 for ; Thu, 24 Mar 2016 00:59:29 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=3602; q=dns/txt; s=iport; t=1458806369; x=1460015969; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=E10sKdb2hql2I+84USSCZIxJM+Zi/ptIYwTzqdcNAvQ=; b=mDRLiVJT43efjJhB95u9UwCQuwOUa9Sl6yg5FRbtO2gcimbK2WCvjKqk jTD2SdZd0RsASbENQYzYEs4Ibri6pogcxe/fCGyCQxlmDb/0+7OuVAHSm U/OQJb2vFUeDN5jIXBdfI4YIdK0C9hhztx7CzMff7jAN+66dgt3BdJ9v1 g=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D6AQAnnfNW/5ldJa1egzNTfQaufItOA?= =?us-ascii?q?Q2BcBcKhWwCHIEaOBQBAQEBAQEBZCeEQQEBAQQBAQEgEToLDAQCAQgRBAEBAQI?= =?us-ascii?q?CIwMCAgIfBgsUAQgIAgQBDQUIiAoDEg6wIothDYR6AQEBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEBAQEBEQR8hSKERIFIdoFMSIJqglYFkmCETTEBjA6BboFtjSWGDoEmh1QBHgE?= =?us-ascii?q?BQoIwgTVsAYgYPX4BAQE?=
X-IronPort-AV: E=Sophos;i="5.24,384,1454976000"; d="scan'208";a="89725654"
Received: from rcdn-core-2.cisco.com ([173.37.93.153]) by rcdn-iport-3.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 24 Mar 2016 07:59:28 +0000
Received: from XCH-RCD-010.cisco.com (xch-rcd-010.cisco.com [173.37.102.20]) by rcdn-core-2.cisco.com (8.14.5/8.14.5) with ESMTP id u2O7xSaH016914 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 24 Mar 2016 07:59:28 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-010.cisco.com (173.37.102.20) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 24 Mar 2016 02:59:27 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Thu, 24 Mar 2016 02:59:27 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , Laurent Ciavaglia , "Liubing (Leo)" , Carsten Bormann 
Thread-Topic: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
Thread-Index: AQHRhLanV1CJbWvMW0GAElzHzEqps59nDdYAgACwpwCAAAXigIAAUmqAgAAkklA=
Date: Thu, 24 Mar 2016 07:59:27 +0000
Message-ID: <498df852d71145ad87eb42f68513fb65@XCH-RCD-006.cisco.com>
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com> <56F0E9CF.7070602@tzi.org> <56F1969D.1060804@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5E464@nkgeml514-mbx.china.huawei.com> <56F259F6.5000807@alcatel-lucent.com> <56F2EE26.9030107@gmail.com>  <56F33837.4020001@gmail.com>
In-Reply-To: <56F33837.4020001@gmail.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Cc: "anima@ietf.org" 
Subject: Re: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 07:59:31 -0000
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From nobody Thu Mar 24 01:15:39 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D3A1612D7B5 for ; Thu, 24 Mar 2016 01:15:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id b_eqKd1X_09x for ; Thu, 24 Mar 2016 01:15:36 -0700 (PDT)
Received: from alln-iport-1.cisco.com (alln-iport-1.cisco.com [173.37.142.88]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B8FF912D674 for ; Thu, 24 Mar 2016 01:15:35 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=2471; q=dns/txt; s=iport; t=1458807335; x=1460016935; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=tg1IZhkAWeuHBUqq7RmG+MW9fi/cA8qFFcTdqdyyW7Q=; b=grs+eUzH1LzinnhJydDmbAJbaNaSStVn+bivMGE0x7IrFjHI6uboxiUw Lu8GQ0bT+v2zf4vGU8MOGy5cGYUPMCcaI4gVj4NMohyF48U9LqI6YgreG 4zHdx3ma7qZwpkt+hTgBQzX6Rvc7P0v7RdeF2I5KovS70gVhAXqiy3GWU Q=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D5AQCtoPNW/4ENJK1egzNTfQa6SgENg?= =?us-ascii?q?XAXCoVsAoE3OBQBAQEBAQEBZCeEQQEBAQMBAQEBNzQQBwQCAQgRBAEBHwkHJws?= =?us-ascii?q?UCQgCBAESCIgXCA7BDwEBAQEBAQEBAQEBAQEBAQEBAQEBAREEhh6ERIQghXIFl?= =?us-ascii?q?14BjXyPEo8IAR4BAUKCMIE1bIhWfgEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,384,1454976000"; d="scan'208";a="253216957"
Received: from alln-core-9.cisco.com ([173.36.13.129]) by alln-iport-1.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 24 Mar 2016 08:15:34 +0000
Received: from XCH-ALN-006.cisco.com (xch-aln-006.cisco.com [173.36.7.16]) by alln-core-9.cisco.com (8.14.5/8.14.5) with ESMTP id u2O8FY7T008740 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 24 Mar 2016 08:15:34 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-006.cisco.com (173.36.7.16) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 24 Mar 2016 03:15:26 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Thu, 24 Mar 2016 03:15:26 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , Michael Richardson , "anima@ietf.org" 
Thread-Topic: [Anima] Who's a proxy? [was: need for some ASA definitions definitions to help bootstrap]
Thread-Index: AQHRhXnRxPNFvQfot0ut2UmYtQzhvZ9oPgEA
Date: Thu, 24 Mar 2016 08:15:26 +0000
Message-ID: <259458c3b8804541aabb0ddf0b151054@XCH-RCD-006.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com> <485.1458564774@obiwan.sandelman.ca> <56F0BDB4.1040904@gmail.com> <56F35918.4080701@gmail.com>
In-Reply-To: <56F35918.4080701@gmail.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Subject: Re: [Anima] Who's a proxy? [was: need for some ASA definitions definitions to help bootstrap]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 08:15:38 -0000

> -----Original Message-----
> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Brian E
> Carpenter
> Sent: 24 March 2016 04:04
> To: Michael Richardson ; anima@ietf.org
> Subject: [Anima] Who's a proxy? [was: need for some ASA definitions
> definitions to help bootstrap]
>=20
> >> These are things that *could* be determined by "power-on-above", i.e.
> >> a controller in the NOC could push (netconf-style or otherweise)
> >> configuration into a node to tell it, "You will be a bootstrap proxy
> >> on interfaces A,C and Q, and the registrar is at address BAR"
> >
> > That doesn't really seem right, because it assumes that the NOC has
> > perfect knowledge of the topology. Potentially, any node with multiple
> > link interfaces could be a proxy, and if the topology is
> > non-redundant, they would all need to be proxies.
>=20
> This has been niggling at me.
>=20
> 1. Does it matter if there is more than one proxy on a given LAN?

I don't see why it would matter. In fact, if you have many proxies, and if =
new devices pick at random, you'd even get some load balancing (if that mat=
ters).=20

Simplicity rules. I suggest:=20
- every device is always a proxy, by default
- admins have two options to stop the network expanding:=20
  1) Intent "this network shall not expand any further" - will turn all pro=
xies off everywhere
      (here is where the power of AN comes out!)=20
  2) Local override through traditional means (e.g., config)=20
     on an interface: "no autonomic"=20
     on device level: "no autonomic proxy"=20

Michael

> 2. If it does matter, the algorithm seems to be something like:
>=20
>    1. Do I have more than one LAN interface?
>       If no, I cannot be a proxy; exit
>    2. Is there an active proxy on any interface?
>       If yes, exclude that interface for now.
>    3. Can I discover the registrar via any interface?
>       If no, I cannot be a proxy; exit
>    4. Register myself as a proxy on the non-excluded interfaces.
>=20
> I don't think we need any input from a higher power. We just need the
> registrar to register itself with GRASP (so that it can be discovered) an=
d the
> proxy instance in each node to execute the above algorithm once in a whil=
e.
>=20
> Regards
>    Brian
>=20
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima


From nobody Thu Mar 24 06:51:27 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 03AA112DAB7 for ; Thu, 24 Mar 2016 06:51:26 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H4=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YcZOJpAqZiGC for ; Thu, 24 Mar 2016 06:51:23 -0700 (PDT)
Received: from alln-iport-5.cisco.com (alln-iport-5.cisco.com [173.37.142.92]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0AFFD12D8CA for ; Thu, 24 Mar 2016 06:49:59 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=14090; q=dns/txt; s=iport; t=1458827399; x=1460036999; h=from:to:cc:subject:date:message-id: content-transfer-encoding:mime-version; bh=J97lQJXD17KFQxFR7kZFxwpMed7/epUYyhFhdU9qJXU=; b=l+uBxMPtdanEQ9+h+NiTUzKY/+iJJQiZTZFLkgzmSppJdZIQ/+/J7kir 7tz1tCRngkMmUtXpZmv6eRmyYjZcZ08ptaKd8rmd709z074EgwpYJrEBM gPo9gGdRdMZgxwtm+9F0cmnpxJWOaZUMoruTp1oNrM/f7JYr8ITKIpfOi o=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0ASAgAL8PNW/49dJa1egzNTfQaufotOA?= =?us-ascii?q?Q2BcBcKhWwCHIEROBQBAQEBAQEBZCeEQQEBAQQBAQEgEToLDAYBGQQBAQECAiM?= =?us-ascii?q?DAgQfBgsUAQgJAQQBDQUIiAoDEg6wEYtiDYR6AQEBAQEBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEBEQR8hSKDRYM9gV8BATOCaoJWBZJghE0xAYEqhEaGHoFujxKHNIdUAR4BAUK?= =?us-ascii?q?DZWwBiBoHFwcWfgEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,385,1454976000"; d="scan'208";a="251369573"
Received: from rcdn-core-7.cisco.com ([173.37.93.143]) by alln-iport-5.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 24 Mar 2016 13:49:55 +0000
Received: from XCH-ALN-006.cisco.com (xch-aln-006.cisco.com [173.36.7.16]) by rcdn-core-7.cisco.com (8.14.5/8.14.5) with ESMTP id u2ODns7D028084 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 24 Mar 2016 13:49:55 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-006.cisco.com (173.36.7.16) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 24 Mar 2016 08:49:54 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Thu, 24 Mar 2016 08:49:54 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , John Strassner , Duzongpeng 
Thread-Topic: Another use case for Intent: "Freeze network enrolment" 
Thread-Index: AdGF07A8j9iFqNBlTOSNYD//018W1Q==
Date: Thu, 24 Mar 2016 13:49:54 +0000
Message-ID: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 13:51:26 -0000
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From nobody Thu Mar 24 07:03:35 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B628A12DC64 for ; Thu, 24 Mar 2016 07:03:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cuBz-Asx9SV8 for ; Thu, 24 Mar 2016 07:03:29 -0700 (PDT)
Received: from rcdn-iport-9.cisco.com (rcdn-iport-9.cisco.com [173.37.86.80]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0208612DBF2 for ; Thu, 24 Mar 2016 06:58:36 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=15002; q=dns/txt; s=iport; t=1458827916; x=1460037516; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=XjXjBHwpadgCGsXwHJzk7CQzwX27XVpZu6XnSGfMXsk=; b=KkA8xHbetetUmTn4EoWS3WSuJXC8w/EhCkWf16sIuFQjk5vV+oE/x1LW jDKuxWod3w7671Iy5tYYt90rr9MAmsL5X0qRuo553+WHywWqXCBu9p6DS a73WPSoKRLG2RRHxxhYAoKWbNmjT2z/YIh5EzKpW6/xgapoqUxl0RPher g=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0ASAgC28fNW/4UNJK1egzNTfQaufotOA?= =?us-ascii?q?Q2BcBcKhWwCHIEROBQBAQEBAQEBZCeEQQEBAQQBAQEgEToLDAQCAQgRBAEBAQI?= =?us-ascii?q?CIwMCAgIfBgsUAQgIAQEEAQ0FCIgKAxIOsBSLYg2EegEBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEBAQEBAREEfIUig0V/gj6CFIJqglYFly0xAYVwhh6Bbo8ShzSHVAEeAQFCg2V?= =?us-ascii?q?sAYgaJRZ+AQEB?=
X-IronPort-AV: E=Sophos;i="5.24,385,1454976000"; d="scan'208";a="84138252"
Received: from alln-core-11.cisco.com ([173.36.13.133]) by rcdn-iport-9.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 24 Mar 2016 13:58:35 +0000
Received: from XCH-RCD-010.cisco.com (xch-rcd-010.cisco.com [173.37.102.20]) by alln-core-11.cisco.com (8.14.5/8.14.5) with ESMTP id u2ODwZ0l010709 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 24 Mar 2016 13:58:35 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-010.cisco.com (173.37.102.20) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 24 Mar 2016 08:58:34 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Thu, 24 Mar 2016 08:58:34 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , John Strassner , Duzongpeng 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAAE3aEABktnAAAACNTOAAAhKozA=
Date: Thu, 24 Mar 2016 13:58:34 +0000
Message-ID: <580de989b17f49fa8fac32c4ba646e88@XCH-RCD-006.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>   <56F37293.7040707@gmail.com>
In-Reply-To: <56F37293.7040707@gmail.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 14:03:34 -0000
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From nobody Thu Mar 24 07:09:20 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CB58612DB7F for ; Thu, 24 Mar 2016 07:09:19 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H4=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id J56OHPrYQA2K for ; Thu, 24 Mar 2016 07:09:18 -0700 (PDT)
Received: from alln-iport-3.cisco.com (alln-iport-3.cisco.com [173.37.142.90]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9C1E612DBB0 for ; Thu, 24 Mar 2016 07:01:49 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=451; q=dns/txt; s=iport; t=1458828109; x=1460037709; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=HUCe/dHTQYLkym67TyXW4Q25I82YFSrt0y5/JNdwJmU=; b=jy2q1L1ZUqG6D5HifIld3RDVl/zvJrxVQVSRAM62nKiuK5v5F9FMXcpe M1ba2epLwcO2X2z5CDf1KQRwrORcUT8Iwi8/uYis82+5havXERKizUUxC QzwQPd7YNJdo5cxQGIDJPUV1NjBTy5fg2Vq+mv7S8d6/l+h5SP2/4cS+4 k=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AGAgA+8vNW/5JdJa1egzOBVrpMAQ2Bc?= =?us-ascii?q?IYNAoEtOBQBAQEBAQEBZCeEQQEBAQMBeQULAgEIRjIlAgQBDQ2IFwjBDAEBAQE?= =?us-ascii?q?BAQEBAQEBAQEBAQEBAQEBFoYehESKEgEEjXOJawGNfIFXARWNJYYOiHoBHgEBQ?= =?us-ascii?q?oNliUJ+AQEB?=
X-IronPort-AV: E=Sophos;i="5.24,385,1454976000"; d="scan'208";a="252505526"
Received: from rcdn-core-10.cisco.com ([173.37.93.146]) by alln-iport-3.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 24 Mar 2016 14:01:48 +0000
Received: from XCH-RCD-005.cisco.com (xch-rcd-005.cisco.com [173.37.102.15]) by rcdn-core-10.cisco.com (8.14.5/8.14.5) with ESMTP id u2OE1mAS026809 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 24 Mar 2016 14:01:48 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-005.cisco.com (173.37.102.15) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 24 Mar 2016 09:01:47 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Thu, 24 Mar 2016 09:01:47 -0500
From: "Michael Behringer (mbehring)" 
To: "Joel M. Halpern" , "Alexander Clemm (alex)" , John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAVKENoAADnV0sAAeZpOAAAGhrgAAHeyeIA==
Date: Thu, 24 Mar 2016 14:01:47 +0000
Message-ID: <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com>
In-Reply-To: <56F2E3DA.9030204@joelhalpern.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , Sheng Jiang , "anima@ietf.org" , =?iso-8859-1?Q?J=E9ferson_Campos_Nobre?= 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 14:09:20 -0000

> To some degree this also depend on how often we expect intents to be
> introduced, modified, or deleted.  If that rate is low, then flooding
> everywhere is probably cheap.  If that rate is high, I would want to be a=
ble to
> constrain the propagation.

Absolutely. And I cannot see how a "high level policy" would require a high=
 update rate. Can anyone? (If not we can declare that question solved!)

Michael

>=20
> Yours,
> Joel


From nobody Thu Mar 24 08:51:20 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D0DF512D1BD for ; Thu, 24 Mar 2016 08:51:18 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.952
X-Spam-Level: 
X-Spam-Status: No, score=-1.952 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, RCVD_IN_SORBS_WEB=0.77, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 42Ot-WM8ZNjB for ; Thu, 24 Mar 2016 08:51:09 -0700 (PDT)
Received: from mailb2.tigertech.net (mailb2.tigertech.net [208.80.4.154]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1DB7F12D543 for ; Thu, 24 Mar 2016 08:37:56 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by mailb2.tigertech.net (Postfix) with ESMTP id E408A1C00FB; Thu, 24 Mar 2016 08:37:55 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1458833875; bh=UPX4l1yupZVTVCMnjFl763F5CHzvh+miFAzAnKO8PAM=; h=Subject:To:References:Cc:From:Date:In-Reply-To:From; b=YNAH9Egz35xLZ52x1CUrEi2huYxLecZzPYfWcYxIFOzS3ert5CB5pHRsRpZSwaM9y WoMQHYwadwhUy/clJ8T0KL4SoYZeM6PskqGPHHhUL8+XiYrejO3mANpPXHsfkcDZI8 S2WUVyzalWzSDOI2RRCvJGpoFwz7ryKsJ4Y4TfrI=
X-Virus-Scanned: Debian amavisd-new at b2.tigertech.net
Received: from Joels-MacBook-Pro.local (unknown [192.165.126.74]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mailb2.tigertech.net (Postfix) with ESMTPSA id BAF6D40039C; Thu, 24 Mar 2016 08:37:54 -0700 (PDT)
To: "Michael Behringer (mbehring)" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com>
From: "Joel M. Halpern" 
Message-ID: <56F409D0.9030900@joelhalpern.com>
Date: Thu, 24 Mar 2016 11:37:52 -0400
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Duzongpeng , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 15:51:19 -0000

Well, I do see folks wanting high rate policy interfaces.  I think it is 
quite sensible for us to note in some descriptive document that our 
target does not include very high rate repeated modification of the high 
level policies.

At which point I can live with just flooding the intents.

Yours,
Joel

On 3/24/16 10:01 AM, Michael Behringer (mbehring) wrote:
>> To some degree this also depend on how often we expect intents to be
>> introduced, modified, or deleted.  If that rate is low, then flooding
>> everywhere is probably cheap.  If that rate is high, I would want to be able to
>> constrain the propagation.
>
> Absolutely. And I cannot see how a "high level policy" would require a high update rate. Can anyone? (If not we can declare that question solved!)
>
> Michael
>
>>
>> Yours,
>> Joel
>


From nobody Thu Mar 24 09:23:57 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 47C1812DB3F for ; Thu, 24 Mar 2016 09:23:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bCsdF-qfy3bN for ; Thu, 24 Mar 2016 09:23:50 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E8C1512DB6A for ; Thu, 24 Mar 2016 09:18:28 -0700 (PDT)
Received: from fr712umx3.dmz.alcatel-lucent.com (unknown [135.245.210.42]) by Websense Email Security Gateway with ESMTPS id DD4C9A873F57C; Thu, 24 Mar 2016 16:18:23 +0000 (GMT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (fr711usmtp1.zeu.alcatel-lucent.com [135.239.2.122]) by fr712umx3.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2OGIQ5c032718 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 24 Mar 2016 16:18:27 GMT
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id u2OGIQkF010895 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 24 Mar 2016 17:18:26 +0100
Received: from [135.224.194.183] (135.239.27.39) by FR712WXCHHUB03.zeu.alcatel-lucent.com (135.239.2.74) with Microsoft SMTP Server (TLS) id 14.3.195.1; Thu, 24 Mar 2016 17:18:26 +0100
To: "EXT Joel M. Halpern" , "Michael Behringer (mbehring)" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com> <56F409D0.9030900@joelhalpern.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56F41351.8080309@alcatel-lucent.com>
Date: Thu, 24 Mar 2016 17:18:25 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <56F409D0.9030900@joelhalpern.com>
Content-Type: multipart/alternative; boundary="------------050404000202080202060100"
X-Originating-IP: [135.239.27.39]
Archived-At: 
Cc: Duzongpeng , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 16:23:54 -0000

--------------050404000202080202060100
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit

Hello,

I think we should distinguish 2 aspects:
     -the rate at which intent policies changes.
     -the rate at which intent policies are flooded.

I agree that, intuitively (we might need a more thorough evaluation), 
intent policy may not change too often.

On the other side, the rate at which intent policies are flooded may 
become "high" if the list of recipients changes frequently, or if other 
events trigger the need to flood some/all intent policies again.
It is not because an intent policy is not changed, that recipients (e.g. 
new ASA instances) should not be informed about the intent policies, and 
thus generates a new broadcast.

For me, it is not clear yet how often the intent policies are flooded, 
to whom and the "size" of such intent/flood traffic.

And so before jumping to some conclusion about the design, I would 
prefer if we dig a bit further in the analysis and evaluation.

Best regards, Laurent.


On 24/03/2016 16:37, EXT Joel M. Halpern wrote:
> Well, I do see folks wanting high rate policy interfaces.  I think it 
> is quite sensible for us to note in some descriptive document that our 
> target does not include very high rate repeated modification of the 
> high level policies.
>
> At which point I can live with just flooding the intents.
>
> Yours,
> Joel
>
> On 3/24/16 10:01 AM, Michael Behringer (mbehring) wrote:
>>> To some degree this also depend on how often we expect intents to be
>>> introduced, modified, or deleted.  If that rate is low, then flooding
>>> everywhere is probably cheap.  If that rate is high, I would want to 
>>> be able to
>>> constrain the propagation.
>>
>> Absolutely. And I cannot see how a "high level policy" would require 
>> a high update rate. Can anyone? (If not we can declare that question 
>> solved!)
>>
>> Michael
>>
>>>
>>> Yours,
>>> Joel
>>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------050404000202080202060100
Content-Type: text/html; charset="windows-1252"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Hello,

      

      I think we should distinguish 2 aspects:

          -the rate at which intent policies changes.

          -the rate at which intent policies are flooded.

      

      I agree that, intuitively (we might need a more thorough
      evaluation), intent policy may not change too often.

      

      On the other side, the rate at which intent policies are flooded
      may become "high" if the list of recipients changes frequently, or
      if other events trigger the need to flood some/all intent policies
      again.

      It is not because an intent policy is not changed, that recipients
      (e.g. new ASA instances) should not be informed about the intent
      policies, and thus generates a new broadcast.

      

      For me, it is not clear yet how often the intent policies are
      flooded, to whom and the "size" of such intent/flood traffic.

      

      And so before jumping to some conclusion about the design, I would
      prefer if we dig a bit further in the analysis and evaluation.

      

      Best regards, Laurent.

      

    

    
On 24/03/2016 16:37, EXT Joel M.
      Halpern wrote:

    

    
Well,
      I do see folks wanting high rate policy interfaces.  I think it is
      quite sensible for us to note in some descriptive document that
      our target does not include very high rate repeated modification
      of the high level policies.
      

      

      At which point I can live with just flooding the intents.
      

      

      Yours,
      

      Joel
      

      

      On 3/24/16 10:01 AM, Michael Behringer (mbehring) wrote:
      

      

        
To some degree this also depend on how
          often we expect intents to be
          

          introduced, modified, or deleted.  If that rate is low, then
          flooding
          

          everywhere is probably cheap.  If that rate is high, I would
          want to be able to
          

          constrain the propagation.
          

        

        

        Absolutely. And I cannot see how a "high level policy" would
        require a high update rate. Can anyone? (If not we can declare
        that question solved!)
        

        

        Michael
        

        

        

          

          Yours,
          

          Joel
          

        

        

      

      

      _______________________________________________
      

      Anima mailing list
      

      Anima@ietf.org
      

      https://www.ietf.org/mailman/listinfo/anima
      

      

    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------050404000202080202060100--


From nobody Thu Mar 24 10:52:42 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 23DFA12D167 for ; Thu, 24 Mar 2016 10:52:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aK6Fp-_yuGDe for ; Thu, 24 Mar 2016 10:52:39 -0700 (PDT)
Received: from rcdn-iport-9.cisco.com (rcdn-iport-9.cisco.com [173.37.86.80]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8344E12D0D8 for ; Thu, 24 Mar 2016 10:52:39 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=4388; q=dns/txt; s=iport; t=1458841959; x=1460051559; h=from:to:subject:date:message-id: content-transfer-encoding:mime-version; bh=+P0Il53otK+Nz5g87xovPrbe/DdZesisgfRCG91oS5o=; b=LLIE1MzR4utwFWhEt0ZKuMW4b8ED8zwtG/qtew5PB2QC8JE0OlP5ivdo IZbfxCQ5Uc5Tzv2bJI0vqyQ8RUkNWWv1w1gZq0oJvN4B6yTI4NesirrzW 9UBB2zBlBaFhhP+M8dbKVbKID0Qrke/5zh4caG8ul6UinBfwVKj7jMJ5m g=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AaAgCyJ/RW/5tdJa1dgzNTfQa6TQENg?= =?us-ascii?q?XAdhyY4FAEBAQEBAQFkHAuESCcTUQE+QiYBBBsRiA4OoFWgQgEBAQcCAR2GHoR?= =?us-ascii?q?EihIFjTuKIwGFcIgMgW2HdIUxjwgBHgEBQoNlbIhWfgEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,386,1454976000"; d="scan'208";a="84221210"
Received: from rcdn-core-4.cisco.com ([173.37.93.155]) by rcdn-iport-9.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 24 Mar 2016 17:52:38 +0000
Received: from XCH-ALN-007.cisco.com (xch-aln-007.cisco.com [173.36.7.17]) by rcdn-core-4.cisco.com (8.14.5/8.14.5) with ESMTP id u2OHqcKX001700 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL) for ; Thu, 24 Mar 2016 17:52:38 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-007.cisco.com (173.36.7.17) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 24 Mar 2016 12:52:38 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Thu, 24 Mar 2016 12:52:37 -0500
From: "Michael Behringer (mbehring)" 
To: "anima@ietf.org" 
Thread-Topic: Reference Model Issues
Thread-Index: AdGF9G4U4OT93HH3SYixSiiMjjRm3A==
Date: Thu, 24 Mar 2016 17:52:37 +0000
Message-ID: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Subject: [Anima] Reference Model Issues
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 17:52:41 -0000

I just read through the entire document again, noting down things that need=
 fixing, or discussion. The below is my personal view, for discussion. I cl=
eaned up the github repository, and created a new file, corresponding to th=
e latest version plus some small edits I made today while reading through a=
gain.=20

One important point: Our chairs want us to take out or reduce drastically a=
ll the stuff that doesn't belong into the current charter. I agree that it =
is not always 100% clear if something is out of charter. But I'm having a h=
ard time reducing some sections. What if we make a new section "Future work=
" (which clearly indicates out of scope for this charter), where we collect=
 those points, but then are a bit more liberal about the volume of each sec=
tion? I think that would be a good compromise. Sheng?=20

I'll summarise the below points in the WG meeting in BA. Maybe we can alrea=
dy clarify some stuff before.=20

Michael




Reference model issues
24 Mar 2016

4.1 Naming:=20
- "It is recommended that the names are generated by the autonomic nodes
   themselves." - how? Should names / addresses not come from the registrar=
?=20
   Suggestion: Name *and* ACP address should be assigned by the registrar
   at enrolment time. zone-id is dynamic, rest is fixed.=20

- "a specific naming convention is out of scope":=20
   Suggestion: We define a default naming and addressing scheme.=20
   1st we settle on an addressing scheme from the ACP draft.=20
   That defines a "device-ID" (last n bits of ACP address).=20
   convert that to a string, and use it as a name. This would result in
   something like: 0123-4567-89ab-0001.example.com.=20
   The name would go into the domain certificate --> need to put this into=
=20
   the bootstrap draft as well.=20
  =20
- to write in the doc: "registrar picks a naming scheme; all registrars=20
  in a domain must use the same scheme. Example is: ...=20
  =20
4.4 ACP
- should move to the end of section 4, it breaks the flow where it is right=
 now.=20

4.7 Intent Distribution
- This section should only explain the basics, real work is out of scope fo=
r this phase.
- should point to the intent distribution draft
- should explain that intent is flooded to all nodes in a domain (if we agr=
ee),=20
  or other methods (if not).=20
- should explain that we expect Intent to have a long life time (months), t=
hus
  Intent distribution is expected to be very infrequent.=20
- should explain that the entire Intent file is flooded in one go (if we ag=
ree).=20

5. Functional Overview
- title is not good. This section really describes how an autonomic node be=
haves. maybe call it "Behaviour of an autonomic node"?=20
- we need to describe the bring-up better, specifically insecure discovery,=
=20
  ACP negotiation, ACP bringup, and subsequent operations, and which protoc=
ols to use
  where. Or should this go into the ACP draft??=20

6. Security and Trust Infrastructure
- needs a review from someone in the security space.
- we need to describe the certificate format; this should likely go into bo=
otstrap draft. This draft should explain where the certificate format is de=
scribed.=20
- the domain certificate should also contain the ACP IP address (where the =
zone bits are set to zero).=20

7.1=20
- broken paragraph; needs fixing.=20
- there is a mention of constrained devices - take out.=20

7.2 Specific ASAs
- should explain MUST implement
- generally, need to work through text in this section; relate back to sect=
ion 5.=20

8.1=20
- remove last reference to RFC7575.=20

8.2 Intent
- explain not part of current charter first.=20
- not sure whether we should keep this level of detail, or shorten further?=
=20

8.3 Aggregated Reporting
- explain not part of current charter first.=20
- my feeling: Should be reduced to 1-2 paragraphs, only high level idea.

General:=20
- should we combine all items that are not part of the current charter into=
 a separate section "Future Topics" or something like that? Then all (*) to=
pic would be nicely contained. Yes, I think I'd prefer that to taking them =
out completely, or shortening them to next-to-nothing.

10. Security Considerations
- needs to be worked over.=20

General:=20
- uppercase Intent everywhere in the doc.=20


From nobody Thu Mar 24 11:20:28 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2F53612D7B4 for ; Thu, 24 Mar 2016 11:20:26 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.53
X-Spam-Level: 
X-Spam-Status: No, score=-14.53 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QkeV3JXA8USz for ; Thu, 24 Mar 2016 11:20:23 -0700 (PDT)
Received: from rcdn-iport-8.cisco.com (rcdn-iport-8.cisco.com [173.37.86.79]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BF46812D71F for ; Thu, 24 Mar 2016 11:20:22 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=20448; q=dns/txt; s=iport; t=1458843622; x=1460053222; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=zFwwVRxolBvDJJLZOUuykyo7c1dnmTEIhwQD5k1WFv4=; b=BoYlfXcS48mlEVIe65PIqnvly9xti40qOUzM1NNEavE1kN1iAmEKxJfs 0hY9vEVHoJgeIzjZhEKSO+lNZrwYheF3Q/CEo4tq2u9pwseEn1q0HtW1q 8mEmJRocwJzzaRFXqifi3h87tlER3ZI9k7lb+AZBYRZEonVQKgUPGsVVp Q=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0BiBQAEI/RW/4cNJK1dgmhMU4EDtV+Ef?= =?us-ascii?q?IFwhg0CHIEYOhIBAQEBAQEBZCeEQgEBBCMKTBACAQgONAICAjAlAgQODYgfsDe?= =?us-ascii?q?QaAEBAQEBAQEBAQEBAQEBAQEBAQEBARWKYoRJgnOCVgWXXgGNfIFthEyIWY8IA?= =?us-ascii?q?ScCOYIDGYFJiUIBfQEBAQ?=
X-IronPort-AV: E=Sophos; i="5.24,386,1454976000"; d="scan'208,217"; a="84553769"
Received: from alln-core-2.cisco.com ([173.36.13.135]) by rcdn-iport-8.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 24 Mar 2016 18:20:21 +0000
Received: from XCH-RTP-007.cisco.com (xch-rtp-007.cisco.com [64.101.220.147]) by alln-core-2.cisco.com (8.14.5/8.14.5) with ESMTP id u2OIKL3a020024 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 24 Mar 2016 18:20:21 GMT
Received: from xch-rtp-001.cisco.com (64.101.220.141) by XCH-RTP-007.cisco.com (64.101.220.147) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 24 Mar 2016 14:20:20 -0400
Received: from xch-rtp-001.cisco.com ([64.101.220.141]) by XCH-RTP-001.cisco.com ([64.101.220.141]) with mapi id 15.00.1104.009; Thu, 24 Mar 2016 14:20:20 -0400
From: "Alexander Clemm (alex)" 
To: John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAVBrxYAAHjH7AAAF5yGgAB0lm4AAFALkQA==
Date: Thu, 24 Mar 2016 18:20:20 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> 
In-Reply-To: 
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [128.107.151.61]
Content-Type: multipart/alternative; boundary="_000_c0148e95d49a4d4a97ad70aa5a771fa8XCHRTP001ciscocom_"
MIME-Version: 1.0
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 18:20:26 -0000

--_000_c0148e95d49a4d4a97ad70aa5a771fa8XCHRTP001ciscocom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_c0148e95d49a4d4a97ad70aa5a771fa8XCHRTP001ciscocom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_c0148e95d49a4d4a97ad70aa5a771fa8XCHRTP001ciscocom_--


From nobody Thu Mar 24 11:28:30 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 102FE12D7E6 for ; Thu, 24 Mar 2016 11:28:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PQcHnjmwtqrU for ; Thu, 24 Mar 2016 11:28:20 -0700 (PDT)
Received: from mail-lf0-x232.google.com (mail-lf0-x232.google.com [IPv6:2a00:1450:4010:c07::232]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BDC3F12D18F for ; Thu, 24 Mar 2016 11:28:19 -0700 (PDT)
Received: by mail-lf0-x232.google.com with SMTP id c62so38292501lfc.1 for ; Thu, 24 Mar 2016 11:28:19 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=Iu1THCOnD16wGfAth8u33JQ47TZzVgYKdyQTP4BprPk=; b=OuRCCCZ/jw4zKPUomhdHmXbPEYPI4/X5rY3QnnRFiwTqZE1d3Pcd6WJ6GyKcFtuyZh 4cSD1/7kkeOTTt65pXoqXn0nL6WJHP5csYxWaogkZuOXlPFuRmfL7FWGZjDNyFw5Xm8d pjA6h7kIFkXa6kzxBbWB79KmF1eK2NRtJFK7FhPhH77X5njlSAE4HGuc9Bf3MfiSUizJ AnXHXqwxTBj9L6m5q1TclrSKiFbI5HetxX2JxJlV1GzsUtnsQs6h2/IZUhif2xwte6Uv /a0nQB+s69Iqf+vr3QSc7QG1wAy6pWfFeAurJ44veg1+MGLjF8dtgImV/W/RRglhNCVE KI3A==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=Iu1THCOnD16wGfAth8u33JQ47TZzVgYKdyQTP4BprPk=; b=mAy84yLLYfBH1Kq5e6YG8VMBv+kUpKxtu3+voDLOoGvziidCB3sj8XuHNpEeafhwhU ACfc1NTiqFt6EaZQI5S91tSebm8Qk5ibG2yp9Zu/CIPGfY0T3r3AF9kxGX4IM+7aQTRz 4mI9RIT8gLehhbWNEha0bBTHOIJLxkflUVt7XsSm0Jv5a/3slArxNli9EDhiGHYXmTdQ 5L5Vq950nJwLrQux7zlsp0CNbS9zEa+xMRFBYfBc9S5F+B7lM+qs425KLW6tztt9w/DN oiMT2iJlVXdAWFdVVc/bZ2T5LZd8Yebvaae5YsFrCksc4Vd34KcU/g1jVJzFDy7UqbaY xlgw==
X-Gm-Message-State: AD7BkJJasDmdXxDO5bH9/jmRsQD8ob7HV5JTsY8W3Rql00Ofwfv5quhKPZYCR58BO4ZohXTFa6DqYaE8rOS+7Q==
MIME-Version: 1.0
X-Received: by 10.25.205.7 with SMTP id d7mr4202411lfg.70.1458844097990; Thu, 24 Mar 2016 11:28:17 -0700 (PDT)
Received: by 10.25.16.234 with HTTP; Thu, 24 Mar 2016 11:28:17 -0700 (PDT)
In-Reply-To: <56F409D0.9030900@joelhalpern.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com> <56F409D0.9030900@joelhalpern.com>
Date: Thu, 24 Mar 2016 11:28:17 -0700
Message-ID: 
From: John Strassner 
To: "Joel M. Halpern" , John Strassner 
Content-Type: multipart/alternative; boundary=001a1142091a3e5835052ecf9cf0
Archived-At: 
Cc: Duzongpeng , "anima@ietf.org" , "Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 18:28:28 -0000

--001a1142091a3e5835052ecf9cf0
Content-Type: text/plain; charset=UTF-8

Note that policy (intent or otherwise) is not **just** about configuration.
I have used policy to define what should be monitored, and how to refine
the monitoring to get the data needed. This is a much higher rate of
running policies than adjusting them for configuration.

Note also that such policies can return a **lot** of data. For example, the
original example I gave (modified by Michael Richardson) was:

    Arrange VM guest distribution so that (CPU) utilization is < 70%

This does a potentially large amount of work, and could trigger additional
policies (intent or otherwise) depending on its results.

regards,
John

On Thu, Mar 24, 2016 at 8:37 AM, Joel M. Halpern 
wrote:

> Well, I do see folks wanting high rate policy interfaces.  I think it is
> quite sensible for us to note in some descriptive document that our target
> does not include very high rate repeated modification of the high level
> policies.
>
> At which point I can live with just flooding the intents.
>
> Yours,
> Joel
>
> On 3/24/16 10:01 AM, Michael Behringer (mbehring) wrote:
>
>> To some degree this also depend on how often we expect intents to be
>>> introduced, modified, or deleted.  If that rate is low, then flooding
>>> everywhere is probably cheap.  If that rate is high, I would want to be
>>> able to
>>> constrain the propagation.
>>>
>>
>> Absolutely. And I cannot see how a "high level policy" would require a
>> high update rate. Can anyone? (If not we can declare that question solved!)
>>
>> Michael
>>
>>
>>> Yours,
>>> Joel
>>>
>>
>>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>



-- 
regards,
John

--001a1142091a3e5835052ecf9cf0
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Note that policy (intent or otherwise) is not **just*=
* about configuration. I have used policy to define what should be monitore=
d, and how to refine the monitoring to get the data needed. This is a much =
higher rate of running policies than adjusting them for configuration.

Note also that such policies can return a **lot** of d=
ata. For example, the original example I gave (modified by Michael Richards=
on) was:

=C2=A0=C2=A0=C2=A0 Arrange VM guest distr=
ibution so that (CPU) utilization is < 70%

This=
 does a potentially large amount of work, and could trigger additional poli=
cies (intent or otherwise) depending on its results.

regards,
John

On Thu, Mar 24, 2016 at 8:37 AM, Joel M. Halpern <j=
mh@joelhalpern.com> wrote:
=
Well, I do see folks wanting high rate policy interfaces.=C2=A0 I think it =
is quite sensible for us to note in some descriptive document that our targ=
et does not include very high rate repeated modification of the high level =
policies.



At which point I can live with just flooding the intents.



Yours,

Joel



On 3/24/16 10:01 AM, Michael Behringer (mbehring) wrote:



To some degree this also depend on how often we expect intents to be

introduced, modified, or deleted.=C2=A0 If that rate is low, then flooding<=
br>
everywhere is probably cheap.=C2=A0 If that rate is high, I would want to b=
e able to

constrain the propagation.





Absolutely. And I cannot see how a "high level policy" would requ=
ire a high update rate. Can anyone? (If not we can declare that question so=
lved!)



Michael







Yours,

Joel









_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima





-- 
regards,
John




--001a1142091a3e5835052ecf9cf0--


From nobody Thu Mar 24 11:52:02 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E379D12D1A6 for ; Thu, 24 Mar 2016 11:52:00 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cnCXqs9dO_Xb for ; Thu, 24 Mar 2016 11:51:58 -0700 (PDT)
Received: from mail-lb0-x231.google.com (mail-lb0-x231.google.com [IPv6:2a00:1450:4010:c04::231]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id DCF9612D19C for ; Thu, 24 Mar 2016 11:51:57 -0700 (PDT)
Received: by mail-lb0-x231.google.com with SMTP id bc4so36865675lbc.2 for ; Thu, 24 Mar 2016 11:51:57 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=tXsp3lh9kITK4GSiFh24yfGHtYKhXYS+4kArnoDIFAQ=; b=R36wNKbKC0XRg6gv8oLzy2OhMEc0X7KJCfjwukoVtzDMA4xRhnw5YZwW82mth4NGhv ZR3cVrS97dStfGEM4jTohbdTCVzSa6rSzwazNu5ILLUikPf6lOKdvupd5aPkedW74LEz 24BqAugn5Z9g/jlai8ATIYLWWB9FzVxkPTdoqyG2kLuSMXqnCt+KypdR0tnhOskANc+V KWcyvsQyzEpMfop0dZV72tFG4ehcrTPK7dUkkHCNpLAn3fYYAniBZDc1ly0DRAOCucFN nkGwiX8rBIzQJUNE+NQnTNKuNlbGb+hE7seFKRBYxbXWNTgDHBbSfRvSGy1J3th48VcQ fqjA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=tXsp3lh9kITK4GSiFh24yfGHtYKhXYS+4kArnoDIFAQ=; b=js6uW/TKFylyGm4n6xiOfL1jcIEXGNBsfv4jrbDCGZVxa5yP1zausBQrx5IttePFuB 8Q5wiBIK22k8BtMx7TsmKILEoE9B0WksDp3nOat+vRtY3zFE45uGa/jtnh+ZngQQG3Qc XmZ+hmrFLfof/UlPnRizFT83xATnCijLGdmFpm4wprgpXrmgiRU/X/iTOTMc1pe/cNvA mx6fqt0IjMsVljx76l4Yumfw3uRpEpXn88NRRvjczjKzFak5SIgVRq/d5IGSzUZB7Lk9 V3tpFcpBVqiD367/LqyXOhGfEueJ+u+D9Jz2Fwek3XQWYYZPjEfUULYefVMpr8Uhm0FW /DXQ==
X-Gm-Message-State: AD7BkJIQiB7Co/Yd7dV5KE6DxrNb0gCqbiau893YKO+DzkfSK/SxS9WFl1HV9hbp0WcKZamHo5RYKI9EKSBfcA==
MIME-Version: 1.0
X-Received: by 10.112.171.70 with SMTP id as6mr4100228lbc.85.1458845516098; Thu, 24 Mar 2016 11:51:56 -0700 (PDT)
Received: by 10.25.16.234 with HTTP; Thu, 24 Mar 2016 11:51:55 -0700 (PDT)
In-Reply-To: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>  
Date: Thu, 24 Mar 2016 11:51:55 -0700
Message-ID: 
From: John Strassner 
To: "Alexander Clemm (alex)" , John Strassner 
Content-Type: multipart/alternative; boundary=001a11c38b94c4f23e052ecff0d2
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 18:52:01 -0000

--001a11c38b94c4f23e052ecff0d2
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Alex,

thanks for your replies, my responses in ..


*... snip ...*

I really like your suggestion of flooding intent to all nodes (even if
there may be optimizations later).  It provides a level of conceptual
clarity.  Every node should be aware of overall intent (and need to decide
for itself if there are actions for it to take).



Why should **every** node be aware of all intent? Many times, this won't be
applicable.



The only reason I can see for doing this is to enable collaboration and/or
emergent behavior (in the AI sense). However, if you want emergent
behavior from autonomic nodes, then each node that is part of that
collaboration must interact. I don't think we know enough about this to
support or prohibit such behavior, other than this places an extra
requirement on the computational and reasoning abilities of the node.



If you have a different purpose, please do explain.



  Why should they _*not*_ be aware?  Now, granted, not every piece of
intent affects every node.  That said, if we are talking about high-level
intent here, and want to make it as easy as possible for users to interact
with the Autonomic Network, why would we bother users or some other
components with having to figure out which piece of intent might apply
where?

The analogy is human conversations. Assume I am a manager. Why would I care
about **every** conversation my employees have?
Look at the example I gave (that Michael Richardson formulated better):

    Arrange VM guest distribution so that (CPU) utilization is < 70%

Why would nodes care about this if it wasn't applicable to them?



A node will be able to determine on its own whether or not it needs to act
on intent.



In general, I disagree. That means that each node has to be able to
interpret intent. That is not going to happen for IoT devices. Even for
more computationally capable devices, how would they do this? Intent by
definition is high-level. I doubt that you and I (as practitioners in the
art) would come up with the same intent formulation.

Instead, I think that a management entity (e.g., an intent compiler) or a
set of such entities (possibly per domain, possibly spanning domains, this
doesn't matter right now) dedicated to the task of translating intent could
output a device-consumable formulation. That's totally reasonable, but is
different from what you said.



And yes, if you do want to implement emergent behavior with inter-node
collaboration, you can introduce this without needing to expose how intent
is implemented.

All this assumes that intent is not updated too frequently (the other
discussion thread), and that the volume of intent is not prohibitive.
However, if that were the case, arguably we are no longer talking about
=E2=80=9Cintent=E2=80=9D any more but other management mechanisms.

We need to be careful. If the above intent example I gave is run over a
large data center, it could generate a lot of data. Note that the intent
itself is probably not run too frequently.



If there is some intent that applies only to some parts of the network and
not others, that can be addressed by using different Autonomic Domains.
And, of course, Autonomic Domains may be associated with certain Autonomic
Functions.  Intervention beyond intent should NOT be allowed for the
Autonomic Function(s) being served an Autonomic Domain, but of course other
interventions (for different functions, hence outside the scope) are still
possible.



I think that Autonomic Domains should (must?) be hidden from the
user/intent writer. They are clearly of huge internal benefit, but why
should the user of intent know about them? In addition, I think that we
should use roles as an indirection mechanism for scalability when working
with Autonomic Domains.



 

What you are saying is that you want your network to be a single autonomic
domain that applies to all functions.  The main think is that I think you
would want to avoid having to compute dependency graphs between objectives
and trying to figure out which nodes within the domain should get one piece
of intent but not another and at which point in time (as the autonomic
network itself and the dependency graph may themselves undergo change).  In
that case, allowing nodes to participate in a handful of separate domains
may be conceptually a simpler model.

 

That is not what I'm saying. Let me restate:

In order for autonomic networks to be of maximum utility to a user, a user
should not have to worry about which autonomic domains are present, and
which autonomic functions are contained in which autonomic domain.

This has nothing to do with compressing the internal topology to a single
autonomic domain - it is the **user model** that I am talking about.



regards,
John

--001a11c38b94c4f23e052ecff0d2
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi Alex,

thanks for your rep=
lies, my responses in <jcs2>..</jcs2>

=

... snip ...
I really like your =
suggestion of flooding intent to all nodes (even if there may be optimizati=
ons later).=C2=A0 It provides a level of conceptual clarity.=C2=A0 Every no=
de should be aware of overall intent (and need to decide for itself if ther=
e are actions for it to take).
<jcs>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 
Why should **every** node be aware of all inte=
nt? Many times, this won't be applicable.
=C2=A0
The only rea=
son I can see for doing this is to enable collaboration and/or emergent beh=
avior (in the AI sense). However, if you want emergent behavior=C2=A0from a=
utonomic nodes, then each node that is part of that collaboration must inte=
ract. I don't think we know enough about this to support or prohibit su=
ch behavior, other than this places an extra requirement on the computation=
al and reasoning abilities of the node.
=C2=A0
I=
f you have a different purpose, please do explain.
<=
p class=3D"MsoNormal"></jcs>
=
=C2=A0<ALEX> Why should they _not_ be aware?=C2=A0 Now, granted, not every piece of intent affects e=
very node.=C2=A0 That said, if we are talking about high-level intent here,=
 and want to make it as easy as possible for users to interact with the Aut=
onomic Network, why would we bother users or some other components with hav=
ing to figure out which piece of intent might apply where?=C2=A0 
<jcs2>
The analogy is human conversations. Assu=
me I am a manager. Why would I care about **every** conversation my employe=
es have?
Look at the example I gave (that Michael Richardson form=
ulated better):

=C2=A0=C2=A0=C2=A0 Arrange VM gues=
t distribution so that (CPU) utilization is < 70%

Why would nodes care about this if it wasn't applicable to them?
</jcs2>


A node will be able to determine on its own whether or not i=
t needs to act on intent.=C2=A0 
</ALEX>
<jcs2><=
/div>
In general, I disagree. That means that each node has to be able =
to interpret intent. That is not going to happen for IoT devices. Even for =
more computationally capable devices, how would they do this? Intent by def=
inition is high-level. I doubt that you and I (as practitioners in the art)=
 would come up with the same intent formulation.

I=
nstead, I think that a management entity (e.g., an intent compiler) or a se=
t of such entities (possibly per domain, possibly spanning domains, this do=
esn't matter right now) dedicated to the task of translating intent cou=
ld output a device-consumable formulation. That's totally reasonable, b=
ut is different from what you said.
</jcs2>
<=
br>

And yes, if you do wan=
t to implement emergent behavior with inter-node collaboration, you can int=
roduce this without needing to expose how intent is implemented.=C2=A0 <=
/u>
All this assume=
s that intent is not updated too frequently (the other discussion thread), =
and that the volume of intent is not prohibitive. However, if that were the=
 case, arguably we are no longer talking about =E2=80=9Cintent=E2=80=9D any=
 more but other management mechanisms.=C2=A0 
<=
/div>
<jcs2>
We need to be careful. If the above intent=
 example I gave is run over a large data center, it could generate a lot of=
 data. Note that the intent itself is probably not run too frequently.
<=
/div>
</jcs2>


=

If there is som=
e intent that applies only to some parts of the network and not others, tha=
t can be addressed by using different Autonomic Domains.=C2=A0 And, of cour=
se, Autonomic Domains may be associated with certain Autonomic Functions.=
=C2=A0 Intervention beyond intent should NOT be allowed for the Autonomic F=
unction(s) being served an Autonomic Domain, but of course other interventi=
ons (for different functions, hence outside the scope) are still possible.=
=C2=A0
<jcs>=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 
I think that Autonomic Domains should (must?) be hidden from the user=
/intent writer. They are clearly of huge internal benefit, but why should t=
he user of intent know about them? In addition, I think that we should use =
roles as an indirection mechanism for scalability when working with Autonom=
ic Domains.
</jcs><=
/u>
=C2=A0&l=
t;ALEX>
What you are saying is that you want your network to be a single autonom=
ic domain that applies to all functions.=C2=A0 The main think is that I thi=
nk you would want to avoid having to compute dependency graphs between obje=
ctives and trying to figure out which nodes within the domain should get on=
e piece of intent but not another and at which point in time (as the autono=
mic network itself and the dependency graph may themselves undergo change).=
=C2=A0 In that case, allowing nodes to participate in a handful of separate=
 domains may be conceptually a simpler model.=C2=A0 
=C2=A0</AL=
EX>
<jcs2>
That is no=
t what I'm saying. Let me restate:

In order fo=
r autonomic networks to be of maximum utility to a user, a user should not =
have to worry about which autonomic domains are present, and which autonomi=
c functions are contained in which autonomic domain.

This has nothing to do with compressing the internal topology to a singl=
e autonomic domain - it is the **user model** that I am talking about.
<=
/div>
</jcs2>


regards,
John


--001a11c38b94c4f23e052ecff0d2--


From nobody Thu Mar 24 12:30:25 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 95FE012D816 for ; Thu, 24 Mar 2016 12:30:15 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UglEEvPYO0Sz for ; Thu, 24 Mar 2016 12:30:03 -0700 (PDT)
Received: from mail-pa0-x22a.google.com (mail-pa0-x22a.google.com [IPv6:2607:f8b0:400e:c03::22a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5A16D12D6AA for ; Thu, 24 Mar 2016 12:29:58 -0700 (PDT)
Received: by mail-pa0-x22a.google.com with SMTP id fe3so29718903pab.1 for ; Thu, 24 Mar 2016 12:29:58 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=UqmxqoVigjpIFWgDXS7vjkgczeZ16JfUqs3/F+VnqNo=; b=XcfHnUmrnkOHYsAXSL7uUnet31ZKrlHUOrJV+TGrD07Djo0JKRPijDIy6YBr5TywEl Ha27PaxImOtck3O/qiILpko5PRAWqB5qK9w8C7Cay+xibOEOXYED86D82cmU7y2+xmAq r4TyGUxyY2wjiFQk/SYq2vL3QWXZMtlKoLj3R/THWfd2vd1SfTWjE56N0cHlnao17+fT D33W1C080LzdVrQJg8QCbtSNQqUFsccisl0PJXyKbcZFmm4/9VQlEte8oxifBO2UC3yZ EaZuQt1F/95g2TQMpfw9bVvmPlb36GGgmd24oC/wz+dFguG5J8Pqd0P1UNFVY5QKAXau XfuA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=UqmxqoVigjpIFWgDXS7vjkgczeZ16JfUqs3/F+VnqNo=; b=TzziEc6z+RXyQXFgOhvNjKX2vQKEV5iyDvuwCPgqi0NANE7VzqMLTmvAmTQdjKw/M4 NnDqzPDDZfoUJe0MP6IWL0WSrt2RV+R3wbUx97mU9paMzintzhG5s/1NvPx2sJ3N29Zv xIALQXeOR20BN4Wa+hKhquCLoQ2o1sZrfN1v2CunEqLwCLvDps0wyyg9B15oruV261IF 05eLlzZhOqeSF4I6/aY/do5Lb3hVjyKZ3POAN3zfjjCbjeBX39/itCL6f/Y7mZQG/Mgz WwTk0gJqBPeMS2/TDSFdCD9ra6BKJHH+QwjCFEX2nk3vag9fcwu/wyIVxIWIFV8//ubg Uoew==
X-Gm-Message-State: AD7BkJKm28p7WvI6aPbTKWGKWPWQ0sV2WrRR2RBIvox57rRQ/A2jviRn+bql3oFhhFAx8Q==
X-Received: by 10.66.167.145 with SMTP id zo17mr15040265pab.73.1458847797970;  Thu, 24 Mar 2016 12:29:57 -0700 (PDT)
Received: from ?IPv6:2406:e007:46cc:1:28cc:dc4c:9703:6781? ([2406:e007:46cc:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id ez6sm11856154pab.12.2016.03.24.12.29.54 (version=TLSv1/SSLv3 cipher=OTHER); Thu, 24 Mar 2016 12:29:56 -0700 (PDT)
To: "Michael Behringer (mbehring)" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , Laurent Ciavaglia , "Liubing (Leo)" , Carsten Bormann 
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com> <56F0E9CF.7070602@tzi.org> <56F1969D.1060804@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5E464@nkgeml514-mbx.china.huawei.com> <56F259F6.5000807@alcatel-lucent.com> <56F2EE26.9030107@gmail.com>  <56F33837.4020001@gmail.com> <498df852d71145ad87eb42f68513fb65@XCH-RCD-006.cisco.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F4403A.1030007@gmail.com>
Date: Fri, 25 Mar 2016 08:30:02 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <498df852d71145ad87eb42f68513fb65@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: "anima@ietf.org" 
Subject: Re: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 19:30:15 -0000

There could be a 3rd case for partial updates: there are multiple sources=
 for
intent in the network.

I don't really have a feeling for frequency, but I do agree that it shoul=
d
be relatively rare.

"Huge" is another discussion. If it does get to MB we might as well distr=
ibute
it with FTP :-). I was hoping for kB.

    Brian
On 24/03/2016 20:59, Michael Behringer (mbehring) wrote:
> Let me challenge that: Why do we need partial updates? I claim full upd=
ate should do. let's keep it as simple as possible.=20
>=20
> I think this statement is true if:=20
> - updates are infrequent (most importantly)
> - Intent doesn't become huge.=20
>=20
> If we really mean what we say ("high level guidance"), then it will cha=
nge once a month at most. Perfectly ok to flood even a few MBy with that =
frequency.=20
>=20
> If not, which assumption is wrong and why?=20
>=20
> Michael
>=20
>> -----Original Message-----
>> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
>> Sent: 24 March 2016 01:44
>> To: J=C3=A9ferson Campos Nobre ; Laurent Ciav=
aglia
>> ; Liubing (Leo) ;=

>> Carsten Bormann 
>> Cc: Michael Behringer (mbehring) ; anima@ietf.org
>> Subject: Re: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE=
:
>> Intent yang model - some thoughts
>>
>> On 24/03/2016 08:48, J=C3=A9ferson Campos Nobre wrote:
>>> Em qua, 23 de mar de 2016 =C3=A0s 16:27, Brian E Carpenter <
>>> brian.e.carpenter@gmail.com> escreveu:
>>>
>>>> On 23/03/2016 21:55, Laurent Ciavaglia wrote:
>>>> ...
>>>>>> b)  Sophisticated operation on Intent, e.g. partial update,
>>>>>> deleting,
>>>> query etc.
>>>>
>>>> I'm pretty sure we will need partial update and deletion. I don't
>>>> expect we will need query.
>>>>
>>>
>>> Probably, query would be needed for (self?) monitoring and intent
>>> verification and consistency checks.
>>
>> Actually the query in that case could just be a Request Synch for the =
relevant
>> subset of intent. The objective "Intent.PrefixManager" is the example =
used
>> in draft-ietf-anima-prefix-management.
>>
>>     Brian
>>>
>>>
>>>>
>>>> ...
>>>>> As for RESTful "operations", this is a possible approach but we
>>>>> should
>>>> probably first discuss the requirements before choosing a type of
>>>> communication scheme (cf characteristics of autonomic networks).
>>>>
>>>> I agree. RESTful has become a sort of religion recently but it's
>>>> really just a way of almost obtaining transactional integrity across=

>>>> proxies and caches.
>>>> It may not be what we need (for example, we might need semantics
>> more
>>>> like DNCP).
>>>> It may be that simple flooding (already supported by GRASP) will be
>>>> sufficient, if the update and deletion operations are atomic.
>>>>
>>>>
>>> + 1
>>>
>>>
>>>>    Brian
>>>>
>>>> _______________________________________________
>>>> Anima mailing list
>>>> Anima@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>
>>>
>=20


From nobody Thu Mar 24 12:32:03 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 48AB612D766 for ; Thu, 24 Mar 2016 12:32:00 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.53
X-Spam-Level: 
X-Spam-Status: No, score=-14.53 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id dOxKYhnYcMug for ; Thu, 24 Mar 2016 12:31:57 -0700 (PDT)
Received: from alln-iport-7.cisco.com (alln-iport-7.cisco.com [173.37.142.94]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 57A7C12D80F for ; Thu, 24 Mar 2016 12:31:51 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=29564; q=dns/txt; s=iport; t=1458847911; x=1460057511; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=aqxjbKr2h1Bv+JScFPsGiWwNJXidXGLqLpFVVh2M9U4=; b=Qb6JdvPh6XxRk5uLDLFMqoBwiFldssZvpZXg5Bjs19GNEUlbeUUV7jaF G4uSBqpApo0vKBYNUQxB1p/xHkNDwQgdbnQZZ4FROUDhm9VYzin4xO0mo zhlofWVjczxWceryxCzx3eCiCBLsIh5hv04AI5W2Xlwojc9qdP3ewINFV 0=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AcAgDQP/RW/4ENJK1egmdMU30Grn+LT?= =?us-ascii?q?gENgXCGDQIcgRo4FAEBAQEBAQFkJ4RBAQEBBCMKTBACAQgOAwQBASgDAgICHxE?= =?us-ascii?q?UCQgCBA4FCIgKAxKwXYteDYR3AQEBAQEBAQEBAQEBAQEBAQEBAQEBFYpigj6CC?= =?us-ascii?q?wkggkqCVgWXLTEBiGKDLIFugW2ETIhZhzSHVAEeAQFCggMZgUlsiFYBfQEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,386,1454976000";  d="scan'208,217";a="253278694"
Received: from alln-core-9.cisco.com ([173.36.13.129]) by alln-iport-7.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 24 Mar 2016 19:31:50 +0000
Received: from XCH-RTP-007.cisco.com (xch-rtp-007.cisco.com [64.101.220.147]) by alln-core-9.cisco.com (8.14.5/8.14.5) with ESMTP id u2OJVnTQ029840 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 24 Mar 2016 19:31:49 GMT
Received: from xch-rtp-001.cisco.com (64.101.220.141) by XCH-RTP-007.cisco.com (64.101.220.147) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 24 Mar 2016 15:31:48 -0400
Received: from xch-rtp-001.cisco.com ([64.101.220.141]) by XCH-RTP-001.cisco.com ([64.101.220.141]) with mapi id 15.00.1104.009; Thu, 24 Mar 2016 15:31:48 -0400
From: "Alexander Clemm (alex)" 
To: John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAVBrxYAAHjH7AAAF5yGgAB0lm4AAFALkQAAL0MSAAAeRBRA=
Date: Thu, 24 Mar 2016 19:31:48 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>   
In-Reply-To: 
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [128.107.151.61]
Content-Type: multipart/alternative; boundary="_000_aaa2a755387847269703235103edb4adXCHRTP001ciscocom_"
MIME-Version: 1.0
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 19:32:00 -0000

--_000_aaa2a755387847269703235103edb4adXCHRTP001ciscocom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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=

--_000_aaa2a755387847269703235103edb4adXCHRTP001ciscocom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_aaa2a755387847269703235103edb4adXCHRTP001ciscocom_--


From nobody Thu Mar 24 12:32:33 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3A5E512D7AA for ; Thu, 24 Mar 2016 12:32:32 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bd_t8gDVvVXi for ; Thu, 24 Mar 2016 12:32:29 -0700 (PDT)
Received: from mail-pf0-x234.google.com (mail-pf0-x234.google.com [IPv6:2607:f8b0:400e:c00::234]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 24BF712D136 for ; Thu, 24 Mar 2016 12:32:29 -0700 (PDT)
Received: by mail-pf0-x234.google.com with SMTP id u190so65674827pfb.3 for ; Thu, 24 Mar 2016 12:32:29 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=bjTWuCsnDtM/O22S8zGUPCNPbwn38fONMhS+24SuBp8=; b=fLRGERAgnqytVwGR4Wp5No6EJ3mZV+rj94uexWRLgTcRK6J/ViqICh9Tc3z2+QHQGH Hkuj9Il6mgLL3I6oJJoE+JbkrRwsTOoGD/cfVkIv2y3KpqyuZ6suta7jBZABilIIQLWz 94J+YImC6IQQisBjyf2VOI7LGiLA0OAqy27t+t6Pyf+36Oeaww3/ZsxYt3dy4mmbixHs AKPj50y4YhS1WZQBREjJGPT1hJWJFSFF9Iy56HwjXrZ56N4co5XXhg1bvjP+/PY8E77a cRupG0RiLNKV11Tpruzkw5kqaTWznnB3HvRtlbgmnA7nC7M7nkXL7b9XWmdlXNnYj8bz 8dPA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=bjTWuCsnDtM/O22S8zGUPCNPbwn38fONMhS+24SuBp8=; b=ZibLmUPZlDoEc4yC4vl5nabWtLWKAhXVBX7mGrd31K6CCPvwq79P3B0OCl/L8B0FWa FlYhbS8X5HujfZomYMRWKWQpbN2EmQZI0pRyCf1YoqKtEP6wHLTvhVkbl1/9F2H04Bz0 RatGbWn2jeDh/BlI5m/2CN3PQJyFgdJqUJ1XsW/fyajG1t2vo1H5jClA4Z89BkJFObbE Cgy1QHLH9DGXZojm+EKdFcJtLg0p4uYRyv8d3eTUuIgc9l1PB72eDW6POykkZNiqpXBK IYM3PXfbvMiGNGPSwkupz5Fj+BJ5bhUpTFE+lc6w+Q0Zkq7mkWaXPdQ6os0FrbbQ5hTf nayw==
X-Gm-Message-State: AD7BkJL63XA3x1R1ZJdK/9CcXr66KnX/fRK6ElW2+myzP9F5BZu21q48H+Pvr2xhSfr7JA==
X-Received: by 10.98.16.7 with SMTP id y7mr15073644pfi.111.1458847948629; Thu, 24 Mar 2016 12:32:28 -0700 (PDT)
Received: from ?IPv6:2406:e007:46cc:1:28cc:dc4c:9703:6781? ([2406:e007:46cc:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id o69sm11833826pfi.14.2016.03.24.12.32.24 (version=TLSv1/SSLv3 cipher=OTHER); Thu, 24 Mar 2016 12:32:27 -0700 (PDT)
To: "Michael Behringer (mbehring)" , John Strassner , Duzongpeng 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F440D1.3060702@gmail.com>
Date: Fri, 25 Mar 2016 08:32:33 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 19:32:32 -0000

On 25/03/2016 02:49, Michael Behringer (mbehring) wrote:
> Maybe we should combine two threads here. I think there is a perfect us=
e case for Intent I  mentioned in another mail, which would be even appli=
cable to what we do today:=20
>=20
> "Freeze network enrolment".=20
>=20
> Since any node can be a proxy node for bootstrap, it makes perfect sens=
e to flood this to all nodes.=20
>=20
> Local override possible via traditional config / netconf / etc.=20
>=20
> No parameters needed. Can't get any simpler.=20

Right, but that one sure sounds like a partial update. The same would app=
ly
to any other form of emergency signal.

   Brian

>=20
> Michael
>=20
>=20
>> -----Original Message-----
>> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
>> Sent: 24 March 2016 05:53
>> To: John Strassner ; Duzongpeng
>> 
>> Cc: Laurent Ciavaglia ; J=C3=A9ferson Cam=
pos Nobre
>> ; anima@ietf.org; Michael Behringer (mbehring)
>> ; Sheng Jiang 
>> Subject: Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>
>> On 24/03/2016 16:49, John Strassner wrote:
>>> I am confused as to why we would consider prefix-management as an
>>> autonomic intent.
>>
>> It's the other way round. Prefix management was chosen as an autonomic=

>> use case, so a follow-up question is: what intent is appropriate for t=
hat use
>> case?
>>
>> And the suggested answer is: the ISP's intent is that certain types of=
 edge
>> network should get certain sizes of prefix. How the prefixes are actua=
lly
>> divided up and assigned is left to the autonomic nodes.
>>
>>> To me, this breaks a fundamental rule of what is intent by requiring
>>> the intent writer to understand too many details about the target of
>>> the intent. The analogy (at least in the ONF and TMF) was specifying
>>> to a taxi company that you want a taxi, but you need to take a certai=
n
>>> route, because you (the customer) know traffic patterns better than
>>> the taxi driver.
>>>
>>> Surely we could pick a more appropriate example?
>>
>> By all means describe other use cases and what sort of intent they nee=
d.
>>
>>     Brian
>>>
>>>
>>> regards,
>>> John
>>>
>>> On Wed, Mar 16, 2016 at 5:41 AM, Duzongpeng
>>  wrote:
>>>
>>>> Hi, Michael,
>>>>
>>>>
>>>>
>>>>          Thanks for the reply. I can understand that intent is not i=
n
>>>> the current milestones, and you are working on several milestones dr=
afts
>> now.
>>>>
>>>>
>>>>
>>>>          You are welcome to contribute to the intent draft at any
>>>> time, and thanks in advance.
>>>>
>>>>
>>>>
>>>>          According to your statements, in my understanding, we can
>>>> keep regarding the =E2=80=9Cintent=E2=80=9D example in =E2=80=9Cpref=
ix-management ID=E2=80=9D as an
>>>> Autonomic intent.
>>>>
>>>>          It=E2=80=99s semantics are just like =E2=80=9Call nodes of =
type x do this;
>>>> type y does that=E2=80=9D, so it is a high level policy.
>>>>
>>>>
>>>>
>>>>          It really helps a lot. Thanks.
>>>>
>>>>
>>>>
>>>>          I generally agree with your statement, while some other
>>>> personal understandings exist (shown below). But perhaps they can be=

>> discussed later.
>>>>
>>>>
>>>>
>>>> Best
>>>> Regards
>>>>
>>>> Zongpeng Du
>>>>
>>>>
>>>>
>>>> *From:* Michael Behringer (mbehring) [mailto:mbehring@cisco.com]
>>>> *Sent:* Wednesday, March 16, 2016 4:17 PM
>>>> *To:* Duzongpeng; anima@ietf.org
>>>> *Cc:* Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Jiang
>>>> *Subject:* RE: [Anima] Does Autonomic network need some intervention=

>>>> beyond intent
>>>>
>>>>
>>>>
>>>> Hi Zongpeng,
>>>>
>>>>
>>>>
>>>> Yes, this is a recurring question and topic. Let me give my view on
>>>> two
>>>> levels:
>>>>
>>>>
>>>>
>>>>
>>>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=
=9D or =E2=80=9Csteering=E2=80=9D a
>>>> network, at least not initially. There is still traditional config,
>>>> plus all the existing and emerging SDN models. My view has always
>>>> been that ANIMA sets a =E2=80=9Cdefault=E2=80=9D policy in a network=
, and all other
>>>> means are more specific and can override intent. This is described i=
n
>>>> RFC7575. Main point here: ANIMA Intent isn=E2=80=99t the only tool i=
n a network.
>>>>
>>>>
>>>>
>>>> [duzongpeng] I agree that intent can coexist with traditional config=
=2E
>>>>
>>>> I generally agree with the policy priorities, but I have some other
>>>> concerns.
>>>>
>>>> Some default policies/parameters set through =E2=80=9Cintent=E2=80=9D=
 perhaps are
>>>> fundamental for the uplayer services.
>>>>
>>>> It should be prevented that human operator uses some CLIs to change
>>>> them by mistake, and the uplayer services are interrupted.
>>>>
>>>> At least, some warning should be reported to the human operator.
>>>> [duzongpeng]
>>>>
>>>>
>>>>
>>>> Intent could perfectly well cover a high level policy such as =E2=80=
=9Call
>>>> nodes of type x do this; type y does that=E2=80=9D. (Which I think i=
s what
>>>> you mention
>>>> below.) When we started with Intent, the vision was to keep it very
>>>> high level, more along the line =E2=80=9Cdo the right thing=E2=80=9D=
 J  It has become
>>>> clear that in today=E2=80=99s networking, people will want to push m=
ore
>>>> =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I th=
ink we=E2=80=99ll just have to acknowledge
>> that.
>>>>
>>>>
>>>>
>>>> Personally I think it=E2=80=99ll pan out like this: Intent will have=
 sections
>>>> per autonomic function. An Intent parser will validate overall
>>>> correctness (signature, etc), then segment the Intent into its bits:=

>>>> One part will be the autonomic fabric itself, then there will be bit=
s
>>>> for each autonomic function.
>>>>
>>>>
>>>>
>>>> So the Intent parser on a node will grab intent, see there is some
>>>> Intent for autonomic function A, will see whether this is running
>>>> locally, and if so, it=E2=80=99ll just pass that piece of Intent to =
that function.
>>>>
>>>>
>>>>
>>>> In this way, a function owner defines both the intent itself, and ho=
w
>>>> it is consumed by his/her function.
>>>>
>>>>
>>>>
>>>> [duzongpeng]Agree.[duzongpeng]
>>>>
>>>>
>>>>
>>>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D an//d peo=
ple can do what
>>>> they want, in the worst case even push config, which is clearly NOT
>>>> what Intent is meant to do. I=E2=80=99d hope that people don=E2=80=99=
t start writing
>>>> Intent like =E2=80=9Cnode
>>>> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: h=
ere is
>>>> your CLI=E2=80=9D. But what I describe above COULD be abused in this=
 way.
>>>> RFC7575 says this is NOT Intent.
>>>>
>>>>
>>>>
>>>> [duzongpeng]Agree. It is hard to give an exact boundary.
>>>>
>>>> However, I think we have two choices here.
>>>>
>>>> The first one is to define a policy first in every ASA, but with som=
e
>>>> open parameters that can be configured. What we need to do is just t=
o
>>>> distribute the network-level parameters for that function. This way
>>>> is suitable to some relatively static functions.
>>>>
>>>> The second is more advanced, in the ideal case, the human operator
>>>> can communicate with network using natural language or something
>>>> similar. This way is more flexible, but harder to realize.
>>>>
>>>> [duzongpeng]
>>>>
>>>>
>>>>
>>>> In other words: Let=E2=80=99s not be too prescriptive, and hope that=
 folks
>>>> will do more or less reasonable things.
>>>>
>>>>
>>>>
>>>> [duzongpeng]Agree.[duzongpeng]
>>>>
>>>>
>>>>
>>>> I=E2=80=99ve been meaning to comment on the Intent drafts and possib=
ly
>>>> contribute, I=E2=80=99m just struggling with priorities at the momen=
t (as you
>>>> can see by my relative silence over the past weeks).
>>>>
>>>>
>>>>
>>>> Michael
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> *From:* Anima [mailto:anima-bounces@ietf.org
>>>> ] *On Behalf Of *Duzongpeng
>>>> *Sent:* 16 March 2016 02:38
>>>> *To:* anima@ietf.org
>>>> *Cc:* Laurent Ciavaglia ; J=C3=A9ferson=

>>>> Campos Nobre ; Sheng Jiang
>>>> 
>>>> *Subject:* [Anima] Does Autonomic network need some intervention
>>>> beyond intent
>>>>
>>>>
>>>>
>>>> Hello, everyone
>>>>
>>>>
>>>>
>>>>          I understand that Autonomic network should be able to work
>>>> with as less configuration as possible. In the ideal case, an
>>>> Autonomic network should be able to work well while only intent is
>>>> needed from human operators.
>>>>
>>>>
>>>>
>>>>          However, I wonder whether Autonomic network may need some
>>>> intervention from human operators through some low-level policies
>>>> beyond intents.
>>>>
>>>>
>>>>
>>>>          For example, in the
>>>> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, i=
t
>>>> is suggested that the prefix lengths for the CSG, ASG, RSG (differen=
t
>>>> roles in IP RAN) can be assigned as an "intent". These configuration=

>>>> parameters need to be distributed in the autonomic domain to
>>>> influence the detail configurations on each autonomic node.
>>>>
>>>>
>>>>
>>>>          As intent is described as an abstract, declarative,
>>>> high-level policy used to operate an autonomic domain, such as an
>> enterprise network.
>>>> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cp=
refix-management ID=E2=80=9D to
>>>> some other words.
>>>>
>>>>
>>>>
>>>> Is =E2=80=9CASA parameters" ok here?
>>>>
>>>>
>>>>
>>>> I think some characteristics for this kind of low-level policies are=

>>>>
>>>> 1.       Network level parameters, need to be distributed by GRASP a=
nd
>> ACP
>>>>
>>>> 2.       Relatively static compared to intent, perhaps only needed t=
o be
>>>> configured once
>>>>
>>>> 3.       Unavoidable in current network environment, mostly for
>>>> establishing the network infrastructure
>>>>
>>>> 4.       Rarely need coordinations with others parameters, usually i=
s
>>>> closed related to one specific autonomic function
>>>>
>>>>
>>>>
>>>> Welcome for comment. Your suggestions will be appreciated.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Best Regards
>>>>
>>>> Zongpeng Du
>>>>
>>>> _______________________________________________
>>>> Anima mailing list
>>>> Anima@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>
>>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Anima mailing list
>>> Anima@ietf.org
>>> https://www.ietf.org/mailman/listinfo/anima
>>>
>=20


From nobody Thu Mar 24 12:43:56 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0C2D412D65E for ; Thu, 24 Mar 2016 12:43:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id sZe7cd5QUoNF for ; Thu, 24 Mar 2016 12:43:52 -0700 (PDT)
Received: from mail-pa0-x22d.google.com (mail-pa0-x22d.google.com [IPv6:2607:f8b0:400e:c03::22d]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 63EB912D19D for ; Thu, 24 Mar 2016 12:43:52 -0700 (PDT)
Received: by mail-pa0-x22d.google.com with SMTP id tt10so30069815pab.3 for ; Thu, 24 Mar 2016 12:43:52 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=ODhoc9Zlbc9I43V3udw93fPYEU37c3CIWR8cwnIoxeM=; b=sA6RugpIKUQi1zxKAtRVRMacCPghBcl+P4Mcfjo7cgT/JMF4RTREDhEU90SJ4MhjmZ /HnS78P/zdiKihlYH+OT5Gmu0cWu0v5rodbt1G8iTMmMKRAdIcArF16nOY3fkUI6bder p54JKe0ikhBU7eyRLeHi427KBS5QFU94bYZHBhQoa/ERqdyow4BE32b4dTZYlE7HGgp5 X3uHpjQyBJvYHR3RLtm+tEP9uAdpsdCQRZNN8at7jitOIm+K08NON4+jhOyXGCaygW/K 7+84ZrglBbZ+tgkicASrCoLbsxVbZUMVnDJoQAVLpREsYRtf9yQCfZsNSiFUOJWjeh8Y hjyA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=ODhoc9Zlbc9I43V3udw93fPYEU37c3CIWR8cwnIoxeM=; b=DXqTRdJON7WLko0ZSJVNrlGDwRuPqQCT0IpUjy8rWIIbf2Z43mRsE53CPBBTg4M873 ZVkhb+Hj4Uqv422Of5EBj/CuDDM6hfBJrAJJ8s1+sHnu8cFgC6r5Ge/+6WjMTm6xZS9O KlxMvrYc/5264zR8hwrd8SLumLf8ntAf7wppgm9upcDvDj7Om4AZSjQEz5jtD7/wdP5r LX7YM/+JTfoIWyKNyer1B0KZ8JCVWVx47eyB7xunQwouac+9iAI4ProEUszV5D+1Qo79 E/+euEj/0i6PSUP30eIIJKRAmT6jB0mf02j4pkoKXb/070pm77nP0bp3foUijoEaoZJj cVkg==
X-Gm-Message-State: AD7BkJJJZ6kEA7SunubKFWWAYp1M2LdKZZqs8d8jbfkhzuSDj7t9+m41ZJ97RbImA8PXag==
X-Received: by 10.66.124.167 with SMTP id mj7mr15260625pab.153.1458848632000;  Thu, 24 Mar 2016 12:43:52 -0700 (PDT)
Received: from ?IPv6:2406:e007:46cc:1:28cc:dc4c:9703:6781? ([2406:e007:46cc:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id z13sm11869964pfi.5.2016.03.24.12.43.48 (version=TLSv1/SSLv3 cipher=OTHER); Thu, 24 Mar 2016 12:43:50 -0700 (PDT)
To: Laurent Ciavaglia , "EXT Joel M. Halpern" , "Michael Behringer (mbehring)" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com> <56F409D0.9030900@joelhalpern.com> <56F41351.8080309@alcatel-lucent.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F4437C.7000803@gmail.com>
Date: Fri, 25 Mar 2016 08:43:56 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56F41351.8080309@alcatel-lucent.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Duzongpeng , "anima@ietf.org" , Sheng Jiang 
Subject: [Anima] Re-flooding [was: Does Autonomic network need some intervention beyond intent]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 19:43:54 -0000

On 25/03/2016 05:18, Laurent Ciavaglia wrote:
...
> It is not because an intent policy is not changed, that recipients (e.g. new ASA instances) should not be informed about the intent policies, and thus generates a new broadcast. 

A newly started ASA can (and probably should) perform the following logic:

 - discover the Intent server (just another ASA, of course)
 - request unicast synchronization of the latest intent

Actually the same applies to sleeping nodes: at some frequency
they should wake up and synchronize intent, or any other flooded
objective that they care about. This is already specified in GRASP:
  "Sleeping nodes that require an objective subject to flooding SHOULD
   periodically request unicast synchronization for that objective."

That doesn't mean than periodic re-flooding is forbidden, but the frequency
can be kept quite low.

    Brian




From nobody Thu Mar 24 12:50:36 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AC95012D1BB for ; Thu, 24 Mar 2016 12:50:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id o9xcbljSQZn7 for ; Thu, 24 Mar 2016 12:50:34 -0700 (PDT)
Received: from mail-pf0-x22d.google.com (mail-pf0-x22d.google.com [IPv6:2607:f8b0:400e:c00::22d]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 34A6012D19D for ; Thu, 24 Mar 2016 12:50:34 -0700 (PDT)
Received: by mail-pf0-x22d.google.com with SMTP id 4so65958355pfd.0 for ; Thu, 24 Mar 2016 12:50:34 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=sf+wMJ+LNontWsLewdXg2bAtqfkKQdNVC4UuNZOysPE=; b=LrJDN76OkAZxkjgKMhKsZSv2YAlWjgf2OE3YcvIJhDwWVhaKqWozNlcrD1CQH2plsu 4vevw1Dqz1L9O/LyXjkflEGGU6KrgXFR6I+GxSRHERj0IDoIKHQ8Ne2BDSyIt4yLff/F kiMCZUFsVkQfeIBfx92e/u25AmvLDitamnjs9XVXvnv1jZIf1x/yqfAKHjmC5vf0KJ4d /lUUCdHfWJ17P7CIVq9cs9+Pgf0TKVPnWA3a9xvDNTu2PYTsQMZMF1XIBJViwjP+rhVu aB2TXPrXtyHrbS2CKagwP9vwlIqrQ8TLIvoEenPD+2u5RRpaulykLB6o4bfMDxm3Qz/v Cp2Q==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=sf+wMJ+LNontWsLewdXg2bAtqfkKQdNVC4UuNZOysPE=; b=HsziPvb9pFvd42BrN3GP1TJ8147Le9Dubsl9PQuvDI/lBRLXz5Qi2fFNmNiibDOCRY R5OeChoIpwTwmYvzlHXsuzoAt3QtjV+lpJxHEvVP9RwOpVbDLfhkyal0wCcT3XI/sW5v meQL+Tq7ohFLUciqIoXGJxJhh7m6ia+2tqYH2tq6ceCSawgE78d+cfFWSI8ZPZSRHj9e Gg146la8UMwyzYDIPgaYEarqvxTN1ey4HzjPqntnbrXPGjQN9LmLsS26C9iHc8KmNJKf ty9g0a44tdAI4jDknn792zpy5C+9GwH+Mb4Cy69hEo/LzF0rFGg3UmVVfjaT6Y++Hrde x9bQ==
X-Gm-Message-State: AD7BkJITAx6rl3Fw6MRXv7QCWnGn9iWiO29p8Guh+Pojo7zTSxnjQJfRe/rvzVjphounPQ==
X-Received: by 10.98.42.211 with SMTP id q202mr15258211pfq.13.1458849033846; Thu, 24 Mar 2016 12:50:33 -0700 (PDT)
Received: from ?IPv6:2406:e007:46cc:1:28cc:dc4c:9703:6781? ([2406:e007:46cc:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id o7sm11881878pfa.37.2016.03.24.12.50.29 (version=TLSv1/SSLv3 cipher=OTHER); Thu, 24 Mar 2016 12:50:32 -0700 (PDT)
To: "Alexander Clemm (alex)" , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>  
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F4450E.9070405@gmail.com>
Date: Fri, 25 Mar 2016 08:50:38 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: [Anima] Relevant intent [was: Does Autonomic network need some intervention beyond intent]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 19:50:35 -0000

On 25/03/2016 07:20, Alexander Clemm (alex) wrote:
...
>  Why should they _not_ be aware?  Now, granted, not every piece of intent affects every node. 

We don't know in advance. At any time, a new ASA might start in any node that
depends on some intent item not previously needed in that node. Now, as I just
said to Laurent, that ASA needs to synchronize relevant intent. If it has previously
been flooded to all nodes, this should be possible without any new messages.
(Hmm. I need to check whether the GRASP rules about caching flooded data
guarantee that already, but I think so.)

    Brian


From nobody Thu Mar 24 12:55:30 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C1C8112D52B for ; Thu, 24 Mar 2016 12:55:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6Wmch_9x8hZd for ; Thu, 24 Mar 2016 12:55:27 -0700 (PDT)
Received: from mail-pf0-x234.google.com (mail-pf0-x234.google.com [IPv6:2607:f8b0:400e:c00::234]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0964212D590 for ; Thu, 24 Mar 2016 12:55:27 -0700 (PDT)
Received: by mail-pf0-x234.google.com with SMTP id u190so66052336pfb.3 for ; Thu, 24 Mar 2016 12:55:27 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=ejyRK6nwkZqDrSCxDY+4+Yjmfos7jYKQIqyLHGxyK0A=; b=F1+QI2QyJzOgpsI9LgodA71GIhkUM/sfVR+nic+9A/9uqxfWTjCdJ92jIIprTM19zX QAL+1oZWEUmjo5lAqzAMpFaiR1nmV45wdwdac4j6ZVQ5SpPj0YMeJGfK+SDMAKj71Mbv bw9Mqv8XnRxPNRap/AduoOveJLi1AH8F75/NM4LDsB97VJ+qlMIik0yL/w6RHn3OtSLf XLb+xFVwNqfPc9gdT4CuINd0eaKnv0MAhFpu2QmuoEWk1/h4OoBHxXhoCq0ieaBaHbUS IHxHjcaM6S2CD4VhAFvo2wapqzjoOWncPCccUAxDbMeSkzHbCIv0wCn0qQmjSs+Mqs3J A84g==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=ejyRK6nwkZqDrSCxDY+4+Yjmfos7jYKQIqyLHGxyK0A=; b=Ag2qJYc+z6jIE7U3rUBwh5vIaoM6jZCCaK8UEC1yJVva/pjx9tboehmFZwENzxQhEH 1wlC3m5eYCMOdxKEu0kbfMWTBTolBWbCKIuXa4BhP1yiRSKkxwJPtpvlgVbXLG44qIGG lhma4bo9zyrrCKtLJQ3MJhhxC6blfqTUI9pVBvgABX8Z1+lgY4iQT8me0qE1er7Yrnyg jxc/rf0vjrZNzwxMs0fVd40Xw6gn4Hqiqk3wHLDTdSYfNDlrA+dgYuqiTKLiSzXDhI7T 3N41eR5/ue6npgD6b1HOrIN55aYTksq5CaWBMLalGxMURnBDlBxGktkqYAgu+fDF73oo KXSw==
X-Gm-Message-State: AD7BkJJcJ8LuDlcDqA4TBQDMn8szu8PMmTS3To9sIGt4iXN/Ft3E3K0pCyInphjxqvbNnQ==
X-Received: by 10.98.89.22 with SMTP id n22mr15383030pfb.55.1458849326631; Thu, 24 Mar 2016 12:55:26 -0700 (PDT)
Received: from ?IPv6:2406:e007:46cc:1:28cc:dc4c:9703:6781? ([2406:e007:46cc:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id b74sm11894499pfd.24.2016.03.24.12.55.22 (version=TLSv1/SSLv3 cipher=OTHER); Thu, 24 Mar 2016 12:55:25 -0700 (PDT)
To: John Strassner , "Alexander Clemm (alex)" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>   
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F44632.7070808@gmail.com>
Date: Fri, 25 Mar 2016 08:55:30 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 19:55:29 -0000

On 25/03/2016 07:51, John Strassner wrote:
...
> A node will be able to determine on its own whether or not it needs to act on intent. 
> 
> 
>> 
>> In general, I disagree. That means that each node has to be able to interpret intent. That is not going to happen for IoT devices.

IoT devices are out of scope here - Anima is for devices that can actually support
quite complex software. Intent for managing a bunch of IoT devices is in scope,
of course.

    Brian


From nobody Thu Mar 24 13:08:21 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7260F12D52C for ; Thu, 24 Mar 2016 13:08:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YJ1OAfn7WF2J for ; Thu, 24 Mar 2016 13:08:17 -0700 (PDT)
Received: from mail-lf0-x22e.google.com (mail-lf0-x22e.google.com [IPv6:2a00:1450:4010:c07::22e]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BF5D312D0A8 for ; Thu, 24 Mar 2016 13:08:16 -0700 (PDT)
Received: by mail-lf0-x22e.google.com with SMTP id c62so39990465lfc.1 for ; Thu, 24 Mar 2016 13:08:16 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=qa7zY2myNv9KRJXlm6UKOLxX9MLqwExIBAJkiJ5oeA4=; b=nt6nTyNvJGohtN6YfvRLvE/C6Bvto4zsd4ctpQFdJ7ccysNGKvjqMonE+EKSKAsGWf huWv6L3+6sF35hOyUqpDgglEL/VQErjChR6Ier3JlxcQ2KVNS0ti6PBm8rSwYsI8LjcA WBbZoexbPKmrnrqZVixECOr/WIprdLAmIeEKmBXCn6239OZd7jsF4oIOIU9IF15a82yI FVY0CqSWOTmmO2xbIufZH23nJqIDavnQa6t7Ahrgrq5ql7Xb6w9QHaysP9BsiMlbjtuk MD9j7x8eu1cVDReKaQ9lLm56S4WQAjPkGO8NseUhj7ka55QRuTo6oDxkbHi2121clcMI 3JBw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=qa7zY2myNv9KRJXlm6UKOLxX9MLqwExIBAJkiJ5oeA4=; b=mRuSE3pg2+LoqG0A0XiHdy22XTOvEVTH0WgJHpGU12hvuBjS/nsqPEcaEGMki/C5P+ 3QaJIeKpqm0ZwPDeNQ85FdpSfuRouZFW0MY+u52NJvNGAQIBNQ8T1Vu0uOYq4/Mbk9oe rQ2E4NL5xFkfASNzqg4yVDbxpqP/JaOpzh8jcnRoWucoC3a6hC6F6qaz4Bc7S5/1ohgq ABgP4nVBY28Px0I6f737AhJPEEbysawHcV74fBpUcjSwt/avOAF/OcdL0gBqVyEdvDwl v6fIAVXE2HDesN5SwXbcy6u7jz/Lck+ZZDEoh44j8zWZxd22fM6/WBEBzO7fQtyw3o5M ifpA==
X-Gm-Message-State: AD7BkJLQSwEDXwUw4mgIbwkIppP3kxl7gEVhoTNglGXyi1Y3gF5sTZRfahXrw+Ad9BTv1V/VivceT0R4ufHkDA==
MIME-Version: 1.0
X-Received: by 10.25.165.140 with SMTP id o134mr4441549lfe.35.1458850094798; Thu, 24 Mar 2016 13:08:14 -0700 (PDT)
Received: by 10.25.16.234 with HTTP; Thu, 24 Mar 2016 13:08:14 -0700 (PDT)
In-Reply-To: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>    
Date: Thu, 24 Mar 2016 13:08:14 -0700
Message-ID: 
From: John Strassner 
To: "Alexander Clemm (alex)" 
Content-Type: multipart/alternative; boundary=001a11410a0eae5a16052ed1014a
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 20:08:20 -0000

--001a11410a0eae5a16052ed1014a
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Thanks Alex, I think we do indeed agree again.

regards,
John

On Thu, Mar 24, 2016 at 12:31 PM, Alexander Clemm (alex) 
wrote:

> Hi John
>
> inline,
>
> --- Alex
>
>
>
> *From:* John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* Thursday, March 24, 2016 11:52 AM
> *To:* Alexander Clemm (alex) ; John Strassner <
> strazpdj@gmail.com>
> *Cc:* Michael Behringer (mbehring) ; Duzongpeng <
> duzongpeng@huawei.com>; Laurent Ciavaglia ;
> Sheng Jiang ; anima@ietf.org; J=C3=A9ferson Campos
> Nobre 
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> Hi Alex,
>
>
>
> thanks for your replies, my responses in ..
>
>
>
>
>
> *... snip ...*
>
> I really like your suggestion of flooding intent to all nodes (even if
> there may be optimizations later).  It provides a level of conceptual
> clarity.  Every node should be aware of overall intent (and need to decid=
e
> for itself if there are actions for it to take).
>
> 
>
> Why should **every** node be aware of all intent? Many times, this won't
> be applicable.
>
>
>
> The only reason I can see for doing this is to enable collaboration and/o=
r
> emergent behavior (in the AI sense). However, if you want emergent
> behavior from autonomic nodes, then each node that is part of that
> collaboration must interact. I don't think we know enough about this to
> support or prohibit such behavior, other than this places an extra
> requirement on the computational and reasoning abilities of the node.
>
>
>
> If you have a different purpose, please do explain.
>
> 
>
>   Why should they _*not*_ be aware?  Now, granted, not every piece
> of intent affects every node.  That said, if we are talking about
> high-level intent here, and want to make it as easy as possible for users
> to interact with the Autonomic Network, why would we bother users or some
> other components with having to figure out which piece of intent might
> apply where?
>
> 
>
> The analogy is human conversations. Assume I am a manager. Why would I
> care about **every** conversation my employees have?
>
> Look at the example I gave (that Michael Richardson formulated better):
>
>
>
>     Arrange VM guest distribution so that (CPU) utilization is < 70%
>
>
>
> Why would nodes care about this if it wasn't applicable to them?
>
> 
>
>
>
>
>
> A node will be able to determine on its own whether or not it needs to ac=
t
> on intent.
>
> 
>
> 
>
> In general, I disagree. That means that each node has to be able to
> interpret intent. That is not going to happen for IoT devices. Even for
> more computationally capable devices, how would they do this? Intent by
> definition is high-level. I doubt that you and I (as practitioners in the
> art) would come up with the same intent formulation.
>
>
>
> Instead, I think that a management entity (e.g., an intent compiler) or a
> set of such entities (possibly per domain, possibly spanning domains, thi=
s
> doesn't matter right now) dedicated to the task of translating intent cou=
ld
> output a device-consumable formulation. That's totally reasonable, but is
> different from what you said.
>
> 
>
>
>
>  The goals with autonomic networks is to make it as easy as possible
> to manage/=E2=80=9Dguide=E2=80=9D network behavior.  I am just looking at=
 it from the user
> perspective:  Anything that requires me to differentiate what I need to
> communicate to whom makes things a little more difficult =E2=80=93 I may =
have
> forgotten someone, I may have inadvertently communicated conflicting
> intent, etc.  You are correct in that there are practical considerations
> that make it hard to implement this.
>
>
>
> You are right to point out that some devices may not be able to support
> intent.  This is why there is still policy-based management etc; that
> doesn=E2=80=99t go away and continues to exist. One aspect that is probab=
ly needed
> is a reporting function that tells a user which devices support intent an=
d
> which don=E2=80=99t and need to be managed using other means .
>
> 
>
>
>
> And yes, if you do want to implement emergent behavior with inter-node
> collaboration, you can introduce this without needing to expose how inten=
t
> is implemented.
>
> All this assumes that intent is not updated too frequently (the other
> discussion thread), and that the volume of intent is not prohibitive.
> However, if that were the case, arguably we are no longer talking about
> =E2=80=9Cintent=E2=80=9D any more but other management mechanisms.
>
> 
>
> We need to be careful. If the above intent example I gave is run over a
> large data center, it could generate a lot of data. Note that the intent
> itself is probably not run too frequently.
>
> 
>
>
>
>
>
> If there is some intent that applies only to some parts of the network an=
d
> not others, that can be addressed by using different Autonomic Domains.
> And, of course, Autonomic Domains may be associated with certain Autonomi=
c
> Functions.  Intervention beyond intent should NOT be allowed for the
> Autonomic Function(s) being served an Autonomic Domain, but of course oth=
er
> interventions (for different functions, hence outside the scope) are stil=
l
> possible.
>
> 
>
> I think that Autonomic Domains should (must?) be hidden from the
> user/intent writer. They are clearly of huge internal benefit, but why
> should the user of intent know about them? In addition, I think that we
> should use roles as an indirection mechanism for scalability when working
> with Autonomic Domains.
>
> 
>
>  
>
> What you are saying is that you want your network to be a single autonomi=
c
> domain that applies to all functions.  The main think is that I think you
> would want to avoid having to compute dependency graphs between objective=
s
> and trying to figure out which nodes within the domain should get one pie=
ce
> of intent but not another and at which point in time (as the autonomic
> network itself and the dependency graph may themselves undergo change).  =
In
> that case, allowing nodes to participate in a handful of separate domains
> may be conceptually a simpler model.
>
>  
>
> 
>
> That is not what I'm saying. Let me restate:
>
>
>
> In order for autonomic networks to be of maximum utility to a user, a use=
r
> should not have to worry about which autonomic domains are present, and
> which autonomic functions are contained in which autonomic domain.
>
>
>
> This has nothing to do with compressing the internal topology to a single
> autonomic domain - it is the **user model** that I am talking about.
>
> 
>
>
>
>  OK, I think we are agreeing on to making it easy on the user.  We do
> not want to have users having to worry about distinctions between differe=
nt
> devices /nodes in an autonomic network etc when communicating network
> intent.
>
>
>
> regards,
>
> John
>



--=20
regards,
John

--001a11410a0eae5a16052ed1014a
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Thanks Alex, I think we do indeed agree again.
<=
div>
regards,
John

On Thu, Mar 24, 2016 at 12:31 PM, Alex=
ander Clemm (alex) <alex@cisco.com> wrote:











Hi John


inline,<AC>


--- Alex

=C2=A0


From: John Strassner [mailto:strazpdj@gmail.com]


Sent: Thursday, March 24, 2016 11:52 AM

To: Alexander Clemm (alex) <alex@cisco.com>; John Strassner <strazpdj@gmail.com>

Cc: Michael Behringer (mbehring) <mbehring@cisco.com>; Duzongpeng <duzongpeng@huawei.com>; Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; Sheng Jiang <jiangsheng@huawei.=
com>; anima@ietf=
.org; J=C3=A9ferson Campos Nobre <jcnobre@inf.ufrgs.br>

Subject: Re: [Anima] Does Autonomic network need some intervention b=
eyond intent


=C2=A0






Hi Alex,






=C2=A0






thanks for your replies, my responses in <jcs2>=
;..</jcs2>






=C2=A0






=C2=A0






... snip ...


I really like y=
our suggestion of flooding intent to all nodes (even if there may be optimi=
zations later).=C2=A0 It provides a level of conceptual clarity.=C2=A0
 Every node should be aware of overall intent (and need to decide for itsel=
f if there are actions for it to take).


<jcs>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0



Why should **every** node be aware of all intent? Ma=
ny times, this won't be applicable.


=C2=A0




The only reason I can see for doing this is to enabl=
e collaboration and/or emergent behavior (in the AI sense). However, if you=
 want emergent behavior=C2=A0from autonomic nodes, then
 each node that is part of that collaboration must interact. I don't th=
ink we know enough about this to support or prohibit such behavior, other t=
han this places an extra requirement on the computational and reasoning abi=
lities of the node.






=C2=A0






If you have a different purpose, please do explain.<=
u>




</jcs>


=C2=A0<ALEX&=
gt; Why should they _not_ be aware?=C2=A0 Now, granted, not every pi=
ece of intent affects every node.=C2=A0 That said, if we are talking about =
high-level
 intent here, and want to make it as easy as possible for users to interact=
 with the Autonomic Network, why would we bother users or some other compon=
ents with having to figure out which piece of intent might apply where?=C2=
=A0







<jcs2>






The analogy is human conversations. Assume I am a ma=
nager. Why would I care about **every** conversation my employees have?<=
/u>






Look at the example I gave (that Michael Richardson =
formulated better):






=C2=A0






=C2=A0=C2=A0=C2=A0 Arrange VM guest distribution so =
that (CPU) utilization is < 70%






=C2=A0






Why would nodes care about this if it wasn't app=
licable to them?






</jcs2>






=C2=A0






=C2=A0








A node will be =
able to determine on its own whether or not it needs to act on intent.=C2=
=A0



</ALEX>






<jcs2>






In general, I disagree. That means that each node ha=
s to be able to interpret intent. That is not going to happen for IoT devic=
es. Even for more computationally capable devices, how would they do this? =
Intent by definition is high-level.
 I doubt that you and I (as practitioners in the art) would come up with th=
e same intent formulation.






=C2=A0






Instead, I think that a management entity (e.g., an =
intent compiler) or a set of such entities (possibly per domain, possibly s=
panning domains, this doesn't matter right now) dedicated to the task o=
f translating intent could output a device-consumable
 formulation. That's totally reasonable, but is different from what you=
 said.






</jcs2>






=C2=A0






<AC> The =
goals with autonomic networks is to make it as easy as possible to manage/=
=E2=80=9Dguide=E2=80=9D network behavior.=C2=A0 I am just looking at it fro=
m the user perspective:=C2=A0 Anything that requires me to differentiate wh=
at
 I need to communicate to whom makes things a little more difficult =E2=80=
=93 I may have forgotten someone, I may have inadvertently communicated con=
flicting intent, etc.=C2=A0 You are correct in that there are practical con=
siderations that make it hard to implement this.=C2=A0



=C2=A0


You are right t=
o point out that some devices may not be able to support intent.=C2=A0 This=
 is why there is still policy-based management etc; that doesn=E2=80=99t go=
 away and continues to exist. One aspect that is probably
 needed is a reporting function that tells a user which devices support int=
ent and which don=E2=80=99t and need to be managed using other means .=C2=
=A0



<AC> =C2=
=A0


=C2=A0






And yes, if you=
 do want to implement emergent behavior with inter-node collaboration, you =
can introduce this without needing to expose how intent is implemented.=C2=
=A0



All this assumes that intent is =
not updated too frequently (the other discussion thread), and that
 the volume of intent is not prohibitive. However, if that were the case, a=
rguably we are no longer talking about =E2=80=9Cintent=E2=80=9D any more bu=
t other management mechanisms.=C2=A0







<jcs2>






We need to be careful. If the above intent example I=
 gave is run over a large data center, it could generate a lot of data. Not=
e that the intent itself is probably not run too frequently.<=
/p>





</jcs2>






=C2=A0








=C2=A0








If there is som=
e intent that applies only to some parts of the network and not others, tha=
t can be addressed by using different Autonomic Domains.=C2=A0 And,
 of course, Autonomic Domains may be associated with certain Autonomic Func=
tions.=C2=A0 Intervention beyond intent should NOT be allowed for the Auton=
omic Function(s) being served an Autonomic Domain, but of course other inte=
rventions (for different functions, hence
 outside the scope) are still possible.=C2=A0


<jcs>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0



I think that Autonomic Domains should (must?) be hid=
den from the user/intent writer. They are clearly of huge internal benefit,=
 but why should the user of intent know about them?
 In addition, I think that we should use roles as an indirection mechanism =
for scalability when working with Autonomic Domains.


</jcs>


=C2=A0<ALEX&=
gt;


What you are saying is that you =
want your network to be a single autonomic domain that applies to all
 functions.=C2=A0 The main think is that I think you would want to avoid ha=
ving to compute dependency graphs between objectives and trying to figure o=
ut which nodes within the domain should get one piece of intent but not ano=
ther and at which point in time (as the
 autonomic network itself and the dependency graph may themselves undergo c=
hange).=C2=A0 In that case, allowing nodes to participate in a handful of s=
eparate domains may be conceptually a simpler model.=C2=A0



=C2=A0</ALEX=
>






<jcs2>






That is not what I'm saying. Let me restate:<=
/u>






=C2=A0






In order for autonomic networks to be of maximum uti=
lity to a user, a user should not have to worry about which autonomic domai=
ns are present, and which autonomic functions are contained in which autono=
mic domain.






=C2=A0






This has nothing to do with compressing the internal=
 topology to a single autonomic domain - it is the **user model** that I am=
 talking about.






</jcs2>






=C2=A0


<AC> OK, I think we are ag=
reeing on to making it easy on the user.=C2=A0 We do not want to have users=
 having to worry about distinctions between different devices
 /nodes in an autonomic network etc when communicating network intent.</=
AC>






=C2=A0






regards,






John
















-- 
regards,
John




--001a11410a0eae5a16052ed1014a--


From nobody Thu Mar 24 13:13:36 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A48B012D827 for ; Thu, 24 Mar 2016 13:13:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id b4j_zBE_2WCU for ; Thu, 24 Mar 2016 13:13:33 -0700 (PDT)
Received: from mail-lb0-x22c.google.com (mail-lb0-x22c.google.com [IPv6:2a00:1450:4010:c04::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 64B4512D0CC for ; Thu, 24 Mar 2016 13:13:33 -0700 (PDT)
Received: by mail-lb0-x22c.google.com with SMTP id bc4so38048978lbc.2 for ; Thu, 24 Mar 2016 13:13:33 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=clVVV9+5uU+0VCw2ssCNymozt42Y4lpvz2zNludf3Co=; b=uew5uBI/Jp7+igzlOQmABh5IYPP3eE/V+1KAeJ7G6a1xMLSek2hh6ndN9hax4iHSOl KTJnn3bbYyxBf1yaR1cRvOlWxPPMsuS9J9Va94CodnoQUA/cnGDxWlZt6EpMR/YzXH9d ls//lNOw5ANZslb06C4+K2DmtZlbEO1Q2lHrEAbH7gu/cWYF4KoKPErmp8HdEAwp5/CZ C1mS+Vd6iOpPF21QT+pLXnoyJzeW03U6SFaETQgehWdDvG+mvR1eqt9kqa/Jh0jFAnfh dHmeC5L+ch8FLbJjUfgpq4/2oUcJrEaviBiKdA+uCSuFu6OzIn46Vreed+fevsSUqfG1 vnzg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=clVVV9+5uU+0VCw2ssCNymozt42Y4lpvz2zNludf3Co=; b=cYPSdDkFo2FUphsOuktPlhEi7Lh4NvcTp9SQFO0NXchNrBp2miHsuII6tJCI5yfCcT oAQ+EWORd8m9yM12QUQLOqYnS/D8SpPD3z1trbVFybTmWJoeEfbVBUgaTcTNF5z4L2n3 43/p6Y/NKzB2t++poF2IkCBIBkTVgMbWxccnSdxEjngN0eKs/s0IzzNyIyJF+/aXEkJ0 UuQmt6YlM/WYfMW/Fb5rXLY7jPmy8IiViO/5/DI5isVD3lF26i9PydSryhZ88vka0hlu 8lftO6x3dslGiSs8Vf50n9Tkrtt/jfrn+Q7+yeiAn52ky/RraiGWg2EumU1BSQrGZDL3 5JMw==
X-Gm-Message-State: AD7BkJIlg4mOkvr5hF0ycA4RtXtsRjbYR5DpX1PGgUTFOZmzccuBxPfXPrQAbVE0FEVDkkFB8d0fmXr376ZjZg==
MIME-Version: 1.0
X-Received: by 10.112.52.230 with SMTP id w6mr3542700lbo.144.1458850411446; Thu, 24 Mar 2016 13:13:31 -0700 (PDT)
Received: by 10.25.16.234 with HTTP; Thu, 24 Mar 2016 13:13:31 -0700 (PDT)
In-Reply-To: <56F44632.7070808@gmail.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>    <56F44632.7070808@gmail.com>
Date: Thu, 24 Mar 2016 13:13:31 -0700
Message-ID: 
From: John Strassner 
To: Brian E Carpenter , John Strassner 
Content-Type: multipart/alternative; boundary=001a11c3eab48e0b0a052ed11474
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Alexander Clemm \(alex\)" , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 20:13:35 -0000

--001a11c3eab48e0b0a052ed11474
Content-Type: text/plain; charset=UTF-8

I may be getting old, but I seem to recall concern over "constrained
nodes" being able to use ANIMA being discussed at several
previous meetings and on the mail list. I re-read the charter, and do
not find such devices explicitly mentioned.

However, the point was that if intent is indeed high-level, then
without a standard lexicon, it will be virtually impossible to ensure
that different devices could process intent directly from the user.

regards,
John

On Thu, Mar 24, 2016 at 12:55 PM, Brian E Carpenter <
brian.e.carpenter@gmail.com> wrote:

> On 25/03/2016 07:51, John Strassner wrote:
> ...
> > A node will be able to determine on its own whether or not it needs to
> act on intent.
> >
> > 
> >> 
> >> In general, I disagree. That means that each node has to be able to
> interpret intent. That is not going to happen for IoT devices.
>
> IoT devices are out of scope here - Anima is for devices that can actually
> support
> quite complex software. Intent for managing a bunch of IoT devices is in
> scope,
> of course.
>
>     Brian
>
>


-- 
regards,
John

--001a11c3eab48e0b0a052ed11474
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


I may be getting old, but I seem to recall concern ov=
er "constrained
nodes" being able to use ANIMA being di=
scussed at several
previous meetings and on the mail list. I re-r=
ead the charter, and do
not find such devices explicitly mentione=
d.

However, the point was that if intent is indeed=
 high-level, then
without a standard lexicon, it will be virtuall=
y impossible to ensure
that different devices could process inten=
t directly from the user.

regards,
John<=
/div>

On Thu=
, Mar 24, 2016 at 12:55 PM, Brian E Carpenter <brian.e.carpenter=
@gmail.com> wrote:
On 25/03=
/2016 07:51, John Strassner wrote:

...

> A node will be able to determine on its own whether or not it needs to=
 act on intent.

>

> </ALEX>

>> <jcs2>

>> In general, I disagree. That means that each node has to be able t=
o interpret intent. That is not going to happen for IoT devices.



IoT devices are out of scope here - Anima is for devices that can actually =
support

quite complex software. Intent for managing a bunch of IoT devices is in sc=
ope,

of course.



=C2=A0 =C2=A0 Brian







-- 
regards,
John




--001a11c3eab48e0b0a052ed11474--


From nobody Thu Mar 24 14:21:17 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 13C8C12D612 for ; Thu, 24 Mar 2016 14:21:13 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id G8QOOrTMZxbQ for ; Thu, 24 Mar 2016 14:21:11 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0786E12D66C for ; Thu, 24 Mar 2016 14:21:11 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id CFD939AF1D9CC; Thu, 24 Mar 2016 21:21:04 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2OLL8Po027432 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 24 Mar 2016 21:21:08 GMT
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2OLL8aQ024974 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 24 Mar 2016 22:21:08 +0100
Received: from [135.224.194.183] (135.239.27.41) by FR712WXCHHUB03.zeu.alcatel-lucent.com (135.239.2.74) with Microsoft SMTP Server (TLS) id 14.3.195.1; Thu, 24 Mar 2016 22:21:08 +0100
To: EXT Brian E Carpenter , Laurent Ciavaglia , "EXT Joel M. Halpern" , "Michael Behringer (mbehring)" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com> <56F409D0.9030900@joelhalpern.com> <56F41351.8080309@alcatel-lucent.com> <56F4437C.7000803@gmail.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56F45A42.4090207@alcatel-lucent.com>
Date: Thu, 24 Mar 2016 22:21:06 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <56F4437C.7000803@gmail.com>
Content-Type: multipart/alternative; boundary="------------050806000900080203020509"
X-Originating-IP: [135.239.27.41]
Archived-At: 
Cc: Duzongpeng , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Re-flooding [was: Does Autonomic network need some intervention beyond intent]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 21:21:13 -0000

--------------050806000900080203020509
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 7bit

Brian,


On 24/03/2016 20:43, EXT Brian E Carpenter wrote:
> On 25/03/2016 05:18, Laurent Ciavaglia wrote:
> ...
>> It is not because an intent policy is not changed, that recipients (e.g. new ASA instances) should not be informed about the intent policies, and thus generates a new broadcast.
> A newly started ASA can (and probably should) perform the following logic:
>
>   - discover the Intent server (just another ASA, of course)
>   - request unicast synchronization of the latest intent
>
> Actually the same applies to sleeping nodes: at some frequency
> they should wake up and synchronize intent, or any other flooded
> objective that they care about. This is already specified in GRASP:
>    "Sleeping nodes that require an objective subject to flooding SHOULD
>     periodically request unicast synchronization for that objective."
>
> That doesn't mean than periodic re-flooding is forbidden, but the frequency
> can be kept quite low.

Agreed.
So, there are at least two mechanisms involved concerning intent 
"distribution":
     -flooding (when intent is created, updated, etc.)
and
     -unicast or pub/sub or ... for the cases we described.

Michael B./all: do you agree with the above? does it make a case for 
considering more than only flooding as the mechanism for intent 
distribution?

Thanks, best regards,
Laurent.

--------------050806000900080203020509
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Brian,

      

    

    
On 24/03/2016 20:43, EXT Brian E
      Carpenter wrote:

    

    

      
On 25/03/2016 05:18, Laurent Ciavaglia wrote:
...


      

        
It is not because an intent policy is not changed, that recipients (e.g. new ASA instances) should not be informed about the intent policies, and thus generates a new broadcast. 


      

      
A newly started ASA can (and probably should) perform the following logic:

 - discover the Intent server (just another ASA, of course)
 - request unicast synchronization of the latest intent

Actually the same applies to sleeping nodes: at some frequency
they should wake up and synchronize intent, or any other flooded
objective that they care about. This is already specified in GRASP:
  "Sleeping nodes that require an objective subject to flooding SHOULD
   periodically request unicast synchronization for that objective."

That doesn't mean than periodic re-flooding is forbidden, but the frequency
can be kept quite low.

    

    

    Agreed.

    So, there are at least two mechanisms involved concerning intent
    "distribution":

        -flooding (when intent is created, updated, etc.)

    and    

        -unicast or pub/sub or ... for the cases we described.

    

    Michael B./all: do you agree with the above? does it make a case for
    considering more than only flooding as the mechanism for intent
    distribution?

    

    Thanks, best regards,

    Laurent.
  


--------------050806000900080203020509--


From nobody Thu Mar 24 14:38:07 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7341F12D914 for ; Thu, 24 Mar 2016 14:38:05 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vuCtGfHwnXdw for ; Thu, 24 Mar 2016 14:38:04 -0700 (PDT)
Received: from mail-pa0-x22e.google.com (mail-pa0-x22e.google.com [IPv6:2607:f8b0:400e:c03::22e]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E34AD12D59F for ; Thu, 24 Mar 2016 14:38:03 -0700 (PDT)
Received: by mail-pa0-x22e.google.com with SMTP id fe3so31743353pab.1 for ; Thu, 24 Mar 2016 14:38:03 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=VRmf5z+bJS5bDb6PkLMTVVs8EP48I0w+DlPvp3vgfPE=; b=trG9ryWnHR28vAIU1lPRt55W+uutRwDppOwLQm7lZb39s+adk6TLxncYeUD0N4+hKW 5shVKA5Vi5sQMJreyvcN3+xBN+pXbhPppikIuEKiyhEC+4f4FEKBoetCYnTT//fv9vhE eLMwr0mEBYIh+f+pNTp4f13E8Gh2X2ux7vRSSz5bQhWR/wuKoOWqzbQ3EbV/ilsrB/PG 1HG7ebI7jR7c55+JSUHLhxkv1YUbmsWDkXsfLdb+odlACU9Dl81KWptUkuWM6PtgJSTJ cMzK/qcIy0552dGd4m5bi4CGksLeDHmxiMr0JNTttjHxkeqNiYfzcvuOiHUXz7xXIKcQ 1QYw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=VRmf5z+bJS5bDb6PkLMTVVs8EP48I0w+DlPvp3vgfPE=; b=DNEGh/+ngGqOZEkGwMUNwzVqpr8mb6Rd3sgx+zLnKMHAsV0Nr9jC+2jEo4cKHjjlLZ NtnlK2/QLEzB0uOHw89xQl0XZh2MZt3E+rbIuHv+Y5n17O6sxyxHLBcciZ+z7UDOvCqT EaDEdjoQmzKJNkTDLFdD6PCr4dMmSRLu/TMOYZ7+RY2fTcq4c/pqSBHebRTf3UsF1PBR V+MVfywe5SwmEBSMEbfmgvCRVrhogvhHZSnWhTp0bav2pitylIEHaJzTorWcpU2hsqX7 Nt7pLmvh78WB4aCKVuIvugMb1mAqCOouD+S1Mmj4oUdTdOGa5GYabfGLbkKX2jR44Rk+ PGag==
X-Gm-Message-State: AD7BkJJilpcjgjelVwt4TjTRieuRJoe/MRuTUok0UyzBRQFwYGvBAVKny8Z+MM8/CQQ77w==
X-Received: by 10.67.23.161 with SMTP id ib1mr15409019pad.156.1458855483463; Thu, 24 Mar 2016 14:38:03 -0700 (PDT)
Received: from ?IPv6:2406:e007:46cc:1:28cc:dc4c:9703:6781? ([2406:e007:46cc:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id wy7sm12172964pab.5.2016.03.24.14.37.59 (version=TLSv1/SSLv3 cipher=OTHER); Thu, 24 Mar 2016 14:38:02 -0700 (PDT)
To: John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>    <56F44632.7070808@gmail.com> 
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F45E40.4070609@gmail.com>
Date: Fri, 25 Mar 2016 10:38:08 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Alexander Clemm \(alex\)" , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 21:38:05 -0000

On 25/03/2016 09:13, John Strassner wrote:
> I may be getting old, but I seem to recall concern over "constrained
> nodes" being able to use ANIMA being discussed at several
> previous meetings and on the mail list. I re-read the charter, and do
> not find such devices explicitly mentioned.

Of course it's hard to define the dividing line, but I've always
thought: light switches, no; lighting controllers, maybe;
building management controllers, yes. I certainly don't expect
an intelligent light switch to process intent items like "energy
policy = maximum savings".

> However, the point was that if intent is indeed high-level, then
> without a standard lexicon, it will be virtually impossible to ensure
> that different devices could process intent directly from the user.

Agreed. We have to get to that point (but later, according to the
WG Charter).

   Brian

> 
> regards,
> John
> 
> On Thu, Mar 24, 2016 at 12:55 PM, Brian E Carpenter <
> brian.e.carpenter@gmail.com> wrote:
> 
>> On 25/03/2016 07:51, John Strassner wrote:
>> ...
>>> A node will be able to determine on its own whether or not it needs to
>> act on intent.
>>>
>>> 
>>>> 
>>>> In general, I disagree. That means that each node has to be able to
>> interpret intent. That is not going to happen for IoT devices.
>>
>> IoT devices are out of scope here - Anima is for devices that can actually
>> support
>> quite complex software. Intent for managing a bunch of IoT devices is in
>> scope,
>> of course.
>>
>>     Brian
>>
>>
> 
> 


From nobody Thu Mar 24 14:54:01 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 77AE512D8F8 for ; Thu, 24 Mar 2016 14:54:00 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id FR2gz1NmD8r6 for ; Thu, 24 Mar 2016 14:53:58 -0700 (PDT)
Received: from mail-lb0-x236.google.com (mail-lb0-x236.google.com [IPv6:2a00:1450:4010:c04::236]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2EA5312D8F3 for ; Thu, 24 Mar 2016 14:53:58 -0700 (PDT)
Received: by mail-lb0-x236.google.com with SMTP id k12so39360890lbb.1 for ; Thu, 24 Mar 2016 14:53:58 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=ny8s4/+OimlgbEK+12SG4OJcfqAupwiyfXLzD4xEUZ4=; b=j142RorBAilG8CrwTeGuBfYmFvkyEeRb3rP9ApepY6VcTsSRk+hudjgKg9cT4sv/dt T57Zhq4d2qPnGP4JRzdTTBGvRaEzl+CqP0Se9KbI0j0strxHcjHsoLwXLYEKJwrJueTt vx/+5NZoH7pPq3JUueeyCU2NUeOwzwKoP16G/cISuKyfNSF6vB7GHKYssQt5txOPg+lI tYBODePyIl9pfePLO3VF6Q/KAdo2/LzX7kP/WSyTmhFScYQwZOhONXACl03CCB3P4v6Q wnAp8ndbExQa2t6fS0IAsKvJLNBpEcEgOyCW8MipGElvK4KqcNz0buslWM7V3SI3/seO UuzQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=ny8s4/+OimlgbEK+12SG4OJcfqAupwiyfXLzD4xEUZ4=; b=ih8Ru+ahSFVhdcnn8cXv0GNp6YOFb40xwnQyyGTkVLupRz77Pn3Y4H/Ft6nW4MXEEn 7QFwtONWf0kR23tKz7exbx3QcGk8C0IKM2xlQMTspmO21TDyWZ79yu6sCCOWB+9RKY9z qOmSIG3EwLD33b+O9EKeZQc1KTKs5A+ekWiA/rhPl0jVDgVnu9nnTdH1s602twS5WHn6 7eG74NXeItIwXPOJwa0YyKN2IsrHG3+OWBrB7IOr7zQB4hFecm0FkAEn5+h1YqOttUH/ D9W5R3JL1ykDFwYIsunoz98+FyPJ6o4fO0BeUHNIrRFHX9AhIW2NHKdvyI2a4cicyOUa kEMg==
X-Gm-Message-State: AD7BkJJTlEq3MqCA+6ve/jH+I+GbZrM7mTP3HtVGcEVu3/e8PCIWoPQLg1EN95a1SRE/kKyGG5RKEuJWG+171Q==
MIME-Version: 1.0
X-Received: by 10.112.139.202 with SMTP id ra10mr4317906lbb.41.1458856436380;  Thu, 24 Mar 2016 14:53:56 -0700 (PDT)
Received: by 10.25.16.234 with HTTP; Thu, 24 Mar 2016 14:53:56 -0700 (PDT)
In-Reply-To: <56F45E40.4070609@gmail.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>    <56F44632.7070808@gmail.com>  <56F45E40.4070609@gmail.com>
Date: Thu, 24 Mar 2016 14:53:56 -0700
Message-ID: 
From: John Strassner 
To: Brian E Carpenter 
Content-Type: multipart/alternative; boundary=001a11c23888ab385d052ed27bbd
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Alexander Clemm \(alex\)" , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 21:54:00 -0000

--001a11c23888ab385d052ed27bbd
Content-Type: text/plain; charset=UTF-8

OK, thanks Brian, the light bulb switched on. :-)

regards,
John

On Thu, Mar 24, 2016 at 2:38 PM, Brian E Carpenter <
brian.e.carpenter@gmail.com> wrote:

> On 25/03/2016 09:13, John Strassner wrote:
> > I may be getting old, but I seem to recall concern over "constrained
> > nodes" being able to use ANIMA being discussed at several
> > previous meetings and on the mail list. I re-read the charter, and do
> > not find such devices explicitly mentioned.
>
> Of course it's hard to define the dividing line, but I've always
> thought: light switches, no; lighting controllers, maybe;
> building management controllers, yes. I certainly don't expect
> an intelligent light switch to process intent items like "energy
> policy = maximum savings".
>
> > However, the point was that if intent is indeed high-level, then
> > without a standard lexicon, it will be virtually impossible to ensure
> > that different devices could process intent directly from the user.
>
> Agreed. We have to get to that point (but later, according to the
> WG Charter).
>
>    Brian
>
> >
> > regards,
> > John
> >
> > On Thu, Mar 24, 2016 at 12:55 PM, Brian E Carpenter <
> > brian.e.carpenter@gmail.com> wrote:
> >
> >> On 25/03/2016 07:51, John Strassner wrote:
> >> ...
> >>> A node will be able to determine on its own whether or not it needs to
> >> act on intent.
> >>>
> >>> 
> >>>> 
> >>>> In general, I disagree. That means that each node has to be able to
> >> interpret intent. That is not going to happen for IoT devices.
> >>
> >> IoT devices are out of scope here - Anima is for devices that can
> actually
> >> support
> >> quite complex software. Intent for managing a bunch of IoT devices is in
> >> scope,
> >> of course.
> >>
> >>     Brian
> >>
> >>
> >
> >
>



-- 
regards,
John

--001a11c23888ab385d052ed27bbd
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


OK, thanks Brian, the light bulb switched on. :-)

regards,
John

On Thu, Mar 24, 2016 at 2:38 PM, Br=
ian E Carpenter <brian.e.carpenter@gmail.com> wrot=
e:
On 25/03/2016 09:13, John Strassner wr=
ote:

> I may be getting old, but I seem to recall concern over "constrai=
ned

> nodes" being able to use ANIMA being discussed at several

> previous meetings and on the mail list. I re-read the charter, and do<=
br>
> not find such devices explicitly mentioned.



Of course it's hard to define the dividing line, but I've always
thought: light switches, no; lighting controllers, maybe;

building management controllers, yes. I certainly don't expect

an intelligent light switch to process intent items like "energy

policy =3D maximum savings".



> However, the point was that if intent is indeed high-level, then

> without a standard lexicon, it will be virtually impossible to ensure<=
br>
> that different devices could process intent directly from the user.


Agreed. We have to get to that point (but later, according to the

WG Charter).



=C2=A0 =C2=A0Brian



>

> regards,

> John

>

> On Thu, Mar 24, 2016 at 12:55 PM, Brian E Carpenter <

> brian.e.carpenter@gmail=
.com> wrote:

>

>> On 25/03/2016 07:51, John Strassner wrote:

>> ...

>>> A node will be able to determine on its own whether or not it =
needs to

>> act on intent.

>>>

>>> </ALEX>

>>>> <jcs2>

>>>> In general, I disagree. That means that each node has to b=
e able to

>> interpret intent. That is not going to happen for IoT devices.

>>

>> IoT devices are out of scope here - Anima is for devices that can =
actually

>> support

>> quite complex software. Intent for managing a bunch of IoT devices=
 is in

>> scope,

>> of course.

>>

>>=C2=A0 =C2=A0 =C2=A0Brian

>>

>>

>

>





-- 
regards,
John




--001a11c23888ab385d052ed27bbd--


From nobody Thu Mar 24 16:00:38 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A2D4612D94C for ; Thu, 24 Mar 2016 16:00:36 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id l_K48wplVT68 for ; Thu, 24 Mar 2016 16:00:33 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 180DB12D55C for ; Thu, 24 Mar 2016 16:00:32 -0700 (PDT)
Received: from fr712umx3.dmz.alcatel-lucent.com (unknown [135.245.210.42]) by Websense Email Security Gateway with ESMTPS id 7946E218F1004; Thu, 24 Mar 2016 23:00:26 +0000 (GMT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (fr711usmtp1.zeu.alcatel-lucent.com [135.239.2.122]) by fr712umx3.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2ON0U42001282 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 24 Mar 2016 23:00:30 GMT
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id u2ON0Ubi013596 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 25 Mar 2016 00:00:30 +0100
Received: from [135.224.194.183] (135.239.27.41) by FR711WXCHHUB02.zeu.alcatel-lucent.com (135.239.2.112) with Microsoft SMTP Server (TLS) id 14.3.195.1; Fri, 25 Mar 2016 00:00:30 +0100
To: "EXT Michael Behringer (mbehring)" , "anima@ietf.org" 
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56F4718D.8010104@alcatel-lucent.com>
Date: Fri, 25 Mar 2016 00:00:29 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
Content-Type: multipart/alternative; boundary="------------020106030406040104000004"
X-Originating-IP: [135.239.27.41]
Archived-At: 
Subject: Re: [Anima] Reference Model Issues
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 23:00:36 -0000

--------------020106030406040104000004
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit

Dear Michael, all,

Thanks for your continuous care of the document and taking some hindsight!

I am also more in favor of grouping non-chartered aspects into a future 
work section, and allow the text there to be meaningful for 1) the 
relationship with (chartered) ANIMA components, 2) the overall picture, 
and 3) describing the implications of these future items with enough depth.

I think the reference document is a pivot one in ANIMA to prepare what 
will come next (in ANIMA or elsewhere) and improve our collective 
understanding on future areas to investigate.
I think it would be a mistake to neglect this opportunity by restricting 
the document on chartered items.

Thanks, Laurent.
ps: of course my above support has nothing to do with the fact that I'm 
too lazy busy to condense the section I am editing ;)

On 24/03/2016 18:52, EXT Michael Behringer (mbehring) wrote:
> I just read through the entire document again, noting down things that need fixing, or discussion. The below is my personal view, for discussion. I cleaned up the github repository, and created a new file, corresponding to the latest version plus some small edits I made today while reading through again.
>
> One important point: Our chairs want us to take out or reduce drastically all the stuff that doesn't belong into the current charter. I agree that it is not always 100% clear if something is out of charter. But I'm having a hard time reducing some sections. What if we make a new section "Future work" (which clearly indicates out of scope for this charter), where we collect those points, but then are a bit more liberal about the volume of each section? I think that would be a good compromise. Sheng?
>
> I'll summarise the below points in the WG meeting in BA. Maybe we can already clarify some stuff before.
>
> Michael
>
>
>
>
> Reference model issues
> 24 Mar 2016
>
> 4.1 Naming:
> - "It is recommended that the names are generated by the autonomic nodes
>     themselves." - how? Should names / addresses not come from the registrar?
>     Suggestion: Name *and* ACP address should be assigned by the registrar
>     at enrolment time. zone-id is dynamic, rest is fixed.
>
> - "a specific naming convention is out of scope":
>     Suggestion: We define a default naming and addressing scheme.
>     1st we settle on an addressing scheme from the ACP draft.
>     That defines a "device-ID" (last n bits of ACP address).
>     convert that to a string, and use it as a name. This would result in
>     something like: 0123-4567-89ab-0001.example.com.
>     The name would go into the domain certificate --> need to put this into
>     the bootstrap draft as well.
>     
> - to write in the doc: "registrar picks a naming scheme; all registrars
>    in a domain must use the same scheme. Example is: ...
>     
> 4.4 ACP
> - should move to the end of section 4, it breaks the flow where it is right now.
>
> 4.7 Intent Distribution
> - This section should only explain the basics, real work is out of scope for this phase.
> - should point to the intent distribution draft
> - should explain that intent is flooded to all nodes in a domain (if we agree),
>    or other methods (if not).
> - should explain that we expect Intent to have a long life time (months), thus
>    Intent distribution is expected to be very infrequent.
> - should explain that the entire Intent file is flooded in one go (if we agree).
>
> 5. Functional Overview
> - title is not good. This section really describes how an autonomic node behaves. maybe call it "Behaviour of an autonomic node"?
> - we need to describe the bring-up better, specifically insecure discovery,
>    ACP negotiation, ACP bringup, and subsequent operations, and which protocols to use
>    where. Or should this go into the ACP draft??
>
> 6. Security and Trust Infrastructure
> - needs a review from someone in the security space.
> - we need to describe the certificate format; this should likely go into bootstrap draft. This draft should explain where the certificate format is described.
> - the domain certificate should also contain the ACP IP address (where the zone bits are set to zero).
>
> 7.1
> - broken paragraph; needs fixing.
> - there is a mention of constrained devices - take out.
>
> 7.2 Specific ASAs
> - should explain MUST implement
> - generally, need to work through text in this section; relate back to section 5.
>
> 8.1
> - remove last reference to RFC7575.
>
> 8.2 Intent
> - explain not part of current charter first.
> - not sure whether we should keep this level of detail, or shorten further?
>
> 8.3 Aggregated Reporting
> - explain not part of current charter first.
> - my feeling: Should be reduced to 1-2 paragraphs, only high level idea.
>
> General:
> - should we combine all items that are not part of the current charter into a separate section "Future Topics" or something like that? Then all (*) topic would be nicely contained. Yes, I think I'd prefer that to taking them out completely, or shortening them to next-to-nothing.
>
> 10. Security Considerations
> - needs to be worked over.
>
> General:
> - uppercase Intent everywhere in the doc.
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------020106030406040104000004
Content-Type: text/html; charset="windows-1252"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Dear Michael, all,

      

      Thanks for your continuous care of the document and taking some
      hindsight!

      

      I am also more in favor of grouping non-chartered aspects into a
      future work section, and allow the text there to be meaningful for
      1) the relationship with (chartered) ANIMA components, 2) the
      overall picture, and 3) describing the implications of these
      future items with enough depth.

      

      I think the reference document is a pivot one in ANIMA to prepare
      what will come next (in ANIMA or elsewhere) and improve our
      collective understanding on future areas to investigate.

      I think it would be a mistake to neglect this opportunity by
      restricting the document on chartered items.

      

      Thanks, Laurent.

      ps: of course my above support has nothing to do with the fact
      that I'm too lazy busy to condense the section I
      am editing ;)

    

    
On 24/03/2016 18:52, EXT Michael
      Behringer (mbehring) wrote:

    

    

      
I just read through the entire document again, noting down things that need fixing, or discussion. The below is my personal view, for discussion. I cleaned up the github repository, and created a new file, corresponding to the latest version plus some small edits I made today while reading through again. 

One important point: Our chairs want us to take out or reduce drastically all the stuff that doesn't belong into the current charter. I agree that it is not always 100% clear if something is out of charter. But I'm having a hard time reducing some sections. What if we make a new section "Future work" (which clearly indicates out of scope for this charter), where we collect those points, but then are a bit more liberal about the volume of each section? I think that would be a good compromise. Sheng? 

I'll summarise the below points in the WG meeting in BA. Maybe we can already clarify some stuff before. 

Michael




Reference model issues
24 Mar 2016

4.1 Naming: 
- "It is recommended that the names are generated by the autonomic nodes
   themselves." - how? Should names / addresses not come from the registrar? 
   Suggestion: Name *and* ACP address should be assigned by the registrar
   at enrolment time. zone-id is dynamic, rest is fixed. 

- "a specific naming convention is out of scope": 
   Suggestion: We define a default naming and addressing scheme. 
   1st we settle on an addressing scheme from the ACP draft. 
   That defines a "device-ID" (last n bits of ACP address). 
   convert that to a string, and use it as a name. This would result in
   something like: 0123-4567-89ab-0001.example.com. 
   The name would go into the domain certificate --> need to put this into 
   the bootstrap draft as well. 
   
- to write in the doc: "registrar picks a naming scheme; all registrars 
  in a domain must use the same scheme. Example is: ... 
   
4.4 ACP
- should move to the end of section 4, it breaks the flow where it is right now. 

4.7 Intent Distribution
- This section should only explain the basics, real work is out of scope for this phase.
- should point to the intent distribution draft
- should explain that intent is flooded to all nodes in a domain (if we agree), 
  or other methods (if not). 
- should explain that we expect Intent to have a long life time (months), thus
  Intent distribution is expected to be very infrequent. 
- should explain that the entire Intent file is flooded in one go (if we agree). 

5. Functional Overview
- title is not good. This section really describes how an autonomic node behaves. maybe call it "Behaviour of an autonomic node"? 
- we need to describe the bring-up better, specifically insecure discovery, 
  ACP negotiation, ACP bringup, and subsequent operations, and which protocols to use
  where. Or should this go into the ACP draft?? 

6. Security and Trust Infrastructure
- needs a review from someone in the security space.
- we need to describe the certificate format; this should likely go into bootstrap draft. This draft should explain where the certificate format is described. 
- the domain certificate should also contain the ACP IP address (where the zone bits are set to zero). 

7.1 
- broken paragraph; needs fixing. 
- there is a mention of constrained devices - take out. 

7.2 Specific ASAs
- should explain MUST implement
- generally, need to work through text in this section; relate back to section 5. 

8.1 
- remove last reference to RFC7575. 

8.2 Intent
- explain not part of current charter first. 
- not sure whether we should keep this level of detail, or shorten further? 

8.3 Aggregated Reporting
- explain not part of current charter first. 
- my feeling: Should be reduced to 1-2 paragraphs, only high level idea.

General: 
- should we combine all items that are not part of the current charter into a separate section "Future Topics" or something like that? Then all (*) topic would be nicely contained. Yes, I think I'd prefer that to taking them out completely, or shortening them to next-to-nothing.

10. Security Considerations
- needs to be worked over. 

General: 
- uppercase Intent everywhere in the doc. 

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima



    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------020106030406040104000004--


From nobody Thu Mar 24 16:37:26 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3B47112D625 for ; Thu, 24 Mar 2016 16:37:23 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.939
X-Spam-Level: 
X-Spam-Status: No, score=-5.939 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_FONT_FACE_BAD=0.981, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bII3v7D32ahM for ; Thu, 24 Mar 2016 16:37:18 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0F61512D0A6 for ; Thu, 24 Mar 2016 16:37:12 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id B9841960A5A88; Thu, 24 Mar 2016 23:37:04 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2ONb8iQ017788 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 24 Mar 2016 23:37:08 GMT
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2ONb6Iu006579 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 25 Mar 2016 00:37:07 +0100
Received: from [135.224.194.183] (135.239.27.39) by FR711WXCHHUB02.zeu.alcatel-lucent.com (135.239.2.112) with Microsoft SMTP Server (TLS) id 14.3.195.1; Fri, 25 Mar 2016 00:37:06 +0100
To: EXT John Strassner , Laurent Ciavaglia 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <56F28403.6060401@alcatel-lucent.com> 
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56F47A21.8070404@alcatel-lucent.com>
Date: Fri, 25 Mar 2016 00:37:05 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: multipart/alternative; boundary="------------020101060608030109080505"
X-Originating-IP: [135.239.27.39]
Archived-At: 
Cc: Duzongpeng , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , "EXT Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 23:37:23 -0000

--------------020101060608030109080505
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit

Hi John,

Thanks for your comments.
Please see below [lc2][/lc2].

On 24/03/2016 04:22, EXT John Strassner wrote:
>
> Hi Laurent,
>
> Thanks for your note. I'd be happy to join in a virtual meeting. 
> Please see .. for my observations.
>
> On 23/03/2016 11:57, EXT Michael Behringer (mbehring) wrote:
>
> John,
>
> /[stripped]/
>
>
> More fundamentally, I think it is misleading to imagine that intent 
> policies will be contained in one "file" (it is possible to do so but 
> impractical/cumbersome to manage). cf also the next comment.
> [/LC]
> 
> See my previous reply. I am more worried about the user model - 
> constructing a single file means that the constructor has to know 
> implementation details. I think that this is unreasonable for the 
> intent writer. However, I see no reason why the intent compiler could 
> not produce a single file, or sets of files that group similar intent 
> (e.g., invoking multiple autonomic functions in the same ASA) together.
> < /jcs>
[lc2]: Fully agree. I was thinking about the user model. The output of 
the intent compiler could be a single file, I am fine with that.[/lc2]

> My suggestion: Intent is flooded to all nodes in a domain.
>
> [LC]
> It is the "resolution" of the intent clauses that determines the 
> domain/set of target "nodes".
> So in that order, yes the intent will be flooded to all the relevant 
> nodes in the "intent policy domain". This domain may be all nodes in 
> an ANIMA network, or a sub-set, only one node, or zero. This is 
> contextual.
> There will be *_many_* intent policies for a given ANIMA network.
> [/LC]
>
> 
>
> Let's use a concept of:
>
>    +--------+     +----------+ +-----------+
>
>    | Intent |---->|  Intent  |---->| Autonomic |
>
>    | Writer |     | Compiler |     | Network  |
>
>    +--------+     +----------+ +-----------+
>
> I **think** what Laurent is saying is that once the Intent Compiler 
> resolves the intent of the user/intent writer, the target of the 
> intent policy is then identified. I agree with this. Even for the 
> broad case of "intent fabric", I don't think that an intent writer 
> need be conscious of this (vs. an autonomic function). Why should he 
> or she care?
>
> Laurent, please correct me if I am wrong here.
>
> 
>
[lc2]: You read me right.[/lc2]
>
> The ASA owners define the content of what they want to push in their 
> bit of
> the Intent. If we insist of full flooding, we encourage the right 
> behaviour. And
> if someone really puts into his Intent “node A: do this; node B: do 
> the other; node C: do the third†it is awkward enough for people to 
> take a step back and think whether they’re doing the right thing J
>
> [LC]
> I think we have a disconnect on what an intent look like and how it 
> can be used.
> In my "mental model", there is not one "big" intent with sub-parts for 
> each ASA, the ANI, etc...
> Each ASA designer (owner) will define the content / specificity of its 
> ASA functionality or links to the existing elements of the ANIMA 
> intent policy model, but it does not mean that we must aggregate all 
> parts into a single "file".
> What would be the reason to do that?
>
> In addition, there will not only be intent policies per functions, or 
> for the ANI.
> There will be intent policies conveying the Service level, the 
> Business level... these will not resolve/map 1-to-1 in a single ASA 
> "policy".
> [/LC]
>
> 
>
> I don't think that you and Michael are really disagreeing (other than 
> one implementation (the file), which could be one option of many).
>
> Indeed, in FOCALE and other work I have done, once intent was 
> compiled, we did group things together for a number of reasons (e.g., 
> dependencies). I think that the Autonomic Functions should advertise 
> their capabilities, or else, how does the Intent Compiler have a clue 
> as to what gets shipped to what (not to mention, how does it translate 
> intent into a form that is consumable by ASAs). This is even more 
> important for Services - one could imagine that a single Service 
> intent would need multiple Autonomic Functions to be implemented.
>
> 
>
[lc2]:I fully agree, weused a same approach in UNIVERSELF for the policy 
continuum from the operator high-level policy input to the service- and 
network-levels policy translation.
and indeed, for the service-level this is a key aspect (mapping the 
autonomic functions contributing/involved/impacting the services).[/lc2]

>
> jcs>
> Do you mean that the intent writer will **know the difference**
> between the fabric and each autonomic function? I hope not,
> but perhaps I'm misunderstanding you...
> < /jcs>
>
> Not sure I understand your concern. Yes, in my model there is a 
> section for the fabric, and sections for the various functions. Seems 
> a natural structure?!?
>
> [LC]
> It is OK to have different intent-policies for the fabric and for the 
> functions, but why put them all into one file/policy?
> [/LC]
> _
> _
>
> 
>
> See my previous reply. In short, I was worried about this requiring 
> the intent writer to be aware of which autonomic functions were 
> present, and the nature of the autonomic fabric. I don't care that 
> much about the structure of the input.
>
> 
>
[lc2]: I agree, it is still early to discuss the structure/format of the 
intent, but my point was to start thinking about what a common intent 
format would look like for the ASAs, to ensure a certain level of 
interoperability and automation in the processing of the intent. This is 
linked to what you say below/in another message on the ASA capabilities 
being advertised/registered.[/lc2]
>
>
> “Function ownerâ€: Yes, I’m suggesting to structure Intent around 
> Autonomic Functions. This is not optimal from a scientific point of 
> view, but aligns very well with the way how things are organised. 
> Probably we need to discuss that in BA. But refer to my YANG model 
> sent a few days ago.
>
> [LC]
> I fully agree on the focus per autonomic function, but it does not 
> imply to have one single intent policy with all parts inside.
> [/LC]
>
> 
>
> My point is that the intent writer should not have to have detailed 
> knowledge about how the autonomic network works, or what its 
> constituent autonomic functions are. In my previous post, I talked 
> about an "autonomic catalog" that the intent writer could use. This 
> would hide the fact that a particular service in the autonomic 
> function had a 1:1 or 1:n correspondence to a set of autonomic functions.
>
> 
>
[lc2]: I agree with the catalog approach. This relates to what we have 
tried to push in different drafts with the life-cycle of ASA, where ASA 
need to register or be discovered, for governance, coordination or 
information/knowledge purposes.[/lc2]

>
> We should schedule some discussion time outside the ANIMA meeting 
> window...
>
> [LC]
> Yes. And a white board too!
> Unfortunately, I cannot attend physically this time.
> Can we schedule a face-to-face ad-hoc meeting?
> [/LC]
>
> 
>
> Count me in.
>
> 
>
[lc]: ok. I will see how to set this up, and will keep the WG/interested 
participants posted.[/lc2]
thank, best regards, Laurent.

> regards,
>
> John
>
>
> On Wed, Mar 23, 2016 at 4:54 AM, Laurent Ciavaglia 
> > wrote:
>
>     Michael,
>
>     There are some aspects I am not sure to fully understand / agree
>     with (marked below with [LC]).
>
>     Best regards, Laurent.
>
>
>     On 23/03/2016 11:57, EXT Michael Behringer (mbehring) wrote:
>>
>>     John,
>>
>>     /[stripped]/
>>
>>     The “parameter†versus “policy†discussion: My view is that we
>>     shouldn’t try to define exactly **what** intent allows and
>>     doesn’t allow, since we’ve heard from many sides that the
>>     boundary is not clear, and it may be impossible to have a really
>>     clear, usable definition. So my point was: Let’s focus on the
>>     distribution model: Intent is flooded to all nodes. Let’s NOT do
>>     selective flooding, unicast Intent to specific nodes, etc,
>>     because that would blur the boundary even worse.
>>
>
>     [LC]
>     Before stating that intent policies should be flooded to all
>     nodes, can we first compare/evaluate the different options?
>     -broadcast is simple...but how important is it compared to other
>     criteria...?
>     I see at least two issues:
>     1. flooding a "big" intent file (cf your current tentative
>     example) may be bandwidth hungry, if one considers how frequently
>     such an intent file could be distributed (ASA stop/start, device
>     join/leave, policy change/update...) compared to distributing only
>     what is required / where it is required.
>     2. security: flooding the entire file reveal information to target
>     (possibly undesirable/rogue) entities that do not need it (and
>     thus may mis-use it).
>
>     More fundamentally, I think it is misleading to imagine that
>     intent policies will be contained in one "file" (it is possible to
>     do so but impractical/cumbersome to manage). cf also the next comment.
>     [/LC]
>
>>     My suggestion: Intent is flooded to all nodes in a domain.
>>
>     [LC]
>     It is the "resolution" of the intent clauses that determines the
>     domain/set of target "nodes".
>     So in that order, yes the intent will be flooded to all the
>     relevant nodes in the "intent policy domain". This domain may be
>     all nodes in an ANIMA network, or a sub-set, only one node, or
>     zero. This is contextual.
>     There will be _*many*_ intent policies for a given ANIMA network.
>     [/LC]
>
>>     The ASA owners define the content of what they want to push in
>>     their bit of the Intent. If we insist of full flooding, we
>>     encourage the right behaviour. And if someone really puts into
>>     his Intent “node A: do this; node B: do the other; node C: do the
>>     third†it is awkward enough for people to take a step back and
>>     think whether they’re doing the right thing J
>>
>     [LC]
>     I think we have a disconnect on what an intent look like and how
>     it can be used.
>     In my "mental model", there is not one "big" intent with sub-parts
>     for each ASA, the ANI, etc...
>     Each ASA designer (owner) will define the content / specificity of
>     its ASA functionality or links to the existing elements of the
>     ANIMA intent policy model, but it does not mean that we must
>     aggregate all parts into a single "file".
>     What would be the reason to do that?
>
>     In addition, there will not only be intent policies per functions,
>     or for the ANI.
>     There will be intent policies conveying the Service level, the
>     Business level... these will not resolve/map 1-to-1 in a single
>     ASA "policy".
>     [/LC]
>
>>     
>>
>>     Do you mean that the intent writer will **know the difference**
>>
>>     between the fabric and each autonomic function? I hope not,
>>
>>     but perhaps I'm misunderstanding you...
>>
>>     
>>
>>     Not sure I understand your concern. Yes, in my model there is a
>>     section for the fabric, and sections for the various functions.
>>     Seems a natural structure?!?
>>
>     [LC]
>     It is OK to have different intent-policies for the fabric and for
>     the functions, but why put them all into one file/policy?
>     [/LC]
>>
>>     “Function ownerâ€: Yes, I’m suggesting to structure Intent around
>>     Autonomic Functions. This is not optimal from a scientific point
>>     of view, but aligns very well with the way how things are
>>     organised. Probably we need to discuss that in BA. But refer to
>>     my YANG model sent a few days ago.
>>
>     [LC]
>     I fully agree on the focus per autonomic function, but it does not
>     imply to have one single intent policy with all parts inside.
>     [/LC]
>
>>     We should schedule some discussion time outside the ANIMA meeting
>>     window...
>>
>     [LC]
>     Yes. And a white board too!
>     Unfortunately, I cannot attend physically this time.
>     Can we schedule a face-to-face ad-hoc meeting?
>     [/LC]
>
>
>>     Michael
>>
>>     *From:*John Strassner [mailto:strazpdj@gmail.com]
>>     *Sent:* 22 March 2016 21:32
>>     *To:* Michael Behringer (mbehring) 
>>     ; John Strassner 
>>     
>>     *Cc:* Duzongpeng 
>>     ; anima@ietf.org
>>     ; Laurent Ciavaglia
>>     
>>     ; Jéferson Campos Nobre
>>      ; Sheng Jiang
>>      
>>     *Subject:* Re: [Anima] Does Autonomic network need some
>>     intervention beyond intent
>>
>>     In general, I agree with Michael. I'd like to emphasize and
>>     expand on a couple of key points:
>>
>>     First, ANIMA isn’t the only thing “influencing†or “steering†a
>>     network, at least not initially. There is still traditional
>>     config, plus all the existing and emerging SDN models. My view
>>     has always been that ANIMA sets a “default†policy in a network,
>>     and all other means are more specific and can override intent.
>>     This is described in RFC7575. Main point here: ANIMA Intent isn’t
>>     the only tool in a network.
>>
>>     
>>
>>     Absolutely agree. Note that even in the Congress case I mentioned
>>
>>     earlier, it is NOT pushing config, and it is NOT viewed as the
>>
>>     "only" policy
>>
>>     
>>
>>     Intent could perfectly well cover a high level policy such as
>>     “all nodes of type x do this; type y does thatâ€. (Which I think
>>     is what you mention below.) When we started with Intent, the
>>     vision was to keep it very high level, more along the line “do
>>     the right thing†J
>>
>>     
>>
>>     Agreed. Although to me, this means that we need to consider the
>>     types of actors that are using intent. For example, most of my
>>     CCIE and JCIE friends have no interest in using an intent
>>     language for pushing configuration. On the other hand, when I
>>     talk to operators and application developers, they would love to
>>     be able to use a high-level intent and have the intent engine
>>     take care of the details for them.
>>
>>     
>>
>>     It has become clear that in today’s networking, people will want
>>     to push more “parameters†than “policyâ€, and I think we’ll just
>>     have to acknowledge that.
>>
>>     
>>
>>     Sorry Michael, I don't follow here. What do you mean
>>
>>     by "parameters"?
>>
>>     
>>
>>     Personally I think it’ll pan out like this: Intent will have
>>     sections per autonomic function. An Intent parser will validate
>>     overall correctness (signature, etc), then segment the Intent
>>     into its bits: One part will be the autonomic fabric itself, then
>>     there will be bits for each autonomic function.
>>
>>     
>>
>>     Do you mean that the intent writer will **know the difference**
>>
>>     between the fabric and each autonomic function? I hope not,
>>
>>     but perhaps I'm misunderstanding you...
>>
>>     
>>
>>     So the Intent parser on a node will grab intent, see there is
>>     some Intent for autonomic function A, will see whether this is
>>     running locally, and if so, it’ll just pass that piece of Intent
>>     to that function.
>>
>>     
>>
>>     Agreed.
>>
>>     
>>
>>     In this way, a function owner defines both the intent itself, and
>>     how it is consumed by his/her function.
>>
>>     
>>
>>     I'm not sure what you mean by "function owner". (Sorry to be
>>
>>     picky, I'm likely under-caffeinated.)
>>
>>     The above seemed like an autonomous decision, made by the parser.
>>
>>     I don't think that the intent writer should be cognizant of how
>>     many or
>>
>>     what types of autonomic functions are present.
>>
>>     
>>
>>     The downside is that there is no “control†and people can do what
>>     they want, in the worst case even push config, which is clearly
>>     NOT what Intent is meant to do.
>>
>>     
>>
>>     And hence, the "HAL 2000" scenario. :-)
>>
>>     
>>
>>     I’d hope that people don’t start writing Intent like “node 17:
>>     here is your CLI; node 23: here is your CLIâ€; node 37: here is
>>     your CLIâ€. But what I describe above COULD be abused in this way.
>>     RFC7575 says this is NOT Intent.
>>
>>     
>>
>>     Absolutely agree!
>>
>>     
>>
>>      regards,
>>
>>     John
>>
>>     On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring)
>>     > wrote:
>>
>>         Hi Zongpeng,
>>
>>         Yes, this is a recurring question and topic. Let me give my
>>         view on two levels:
>>
>>         First, ANIMA isn’t the only thing “influencing†or “steeringâ€
>>         a network, at least not initially. There is still traditional
>>         config, plus all the existing and emerging SDN models. My
>>         view has always been that ANIMA sets a “default†policy in a
>>         network, and all other means are more specific and can
>>         override intent. This is described in RFC7575. Main point
>>         here: ANIMA Intent isn’t the only tool in a network.
>>
>>         Intent could perfectly well cover a high level policy such as
>>         “all nodes of type x do this; type y does thatâ€. (Which I
>>         think is what you mention below.) When we started with
>>         Intent, the vision was to keep it very high level, more along
>>         the line “do the right thing†JIt has become clear that in
>>         today’s networking, people will want to push more
>>         “parameters†than “policyâ€, and I think we’ll just have to
>>         acknowledge that.
>>
>>         Personally I think it’ll pan out like this: Intent will have
>>         sections per autonomic function. An Intent parser will
>>         validate overall correctness (signature, etc), then segment
>>         the Intent into its bits: One part will be the autonomic
>>         fabric itself, then there will be bits for each autonomic
>>         function.
>>
>>         So the Intent parser on a node will grab intent, see there is
>>         some Intent for autonomic function A, will see whether this
>>         is running locally, and if so, it’ll just pass that piece of
>>         Intent to that function.
>>
>>         In this way, a function owner defines both the intent itself,
>>         and how it is consumed by his/her function.
>>
>>         The downside is that there is no “control†and people can do
>>         what they want, in the worst case even push config, which is
>>         clearly NOT what Intent is meant to do. I’d hope that people
>>         don’t start writing Intent like “node 17: here is your CLI;
>>         node 23: here is your CLIâ€; node 37: here is your CLIâ€. But
>>         what I describe above COULD be abused in this way. RFC7575
>>         says this is NOT Intent.
>>
>>         In other words: Let’s not be too prescriptive, and hope that
>>         folks will do more or less reasonable things.
>>
>>         I’ve been meaning to comment on the Intent drafts and
>>         possibly contribute, I’m just struggling with priorities at
>>         the moment (as you can see by my relative silence over the
>>         past weeks).
>>
>>         Michael
>>
>>         *From:*Anima
>>         [mailto:anima-bounces@ietf.org
>>         ] *On Behalf Of *Duzongpeng
>>         *Sent:* 16 March 2016 02:38
>>         *To:* anima@ietf.org 
>>         *Cc:* Laurent Ciavaglia
>>         <laurent.ciavaglia@nokia.com
>>         >; Jéferson Campos Nobre
>>         <jcnobre@inf.ufrgs.br
>>         >; Sheng Jiang
>>         <jiangsheng@huawei.com
>>         >
>>         *Subject:* [Anima] Does Autonomic network need some
>>         intervention beyond intent
>>
>>         Hello, everyone
>>
>>                  I understand that Autonomic network should be able
>>         to work with as less configuration as possible. In the ideal
>>         case, an Autonomic network should be able to work well while
>>         only intent is needed from human operators.
>>
>>                  However, I wonder whether Autonomic network may need
>>         some intervention from human operators through some low-level
>>         policies beyond intents.
>>
>>                  For example, in the
>>         https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,
>>         it is suggested that the prefix lengths for the CSG, ASG, RSG
>>         (different roles in IP RAN) can be assigned as an "intent".
>>         These configuration parameters need to be distributed in the
>>         autonomic domain to influence the detail configurations on
>>         each autonomic node.
>>
>>                  As intent is described as an abstract, declarative,
>>         high-level policy used to operate an autonomic domain, such
>>         as an enterprise network. Perhaps we can rename the “intentâ€
>>         in the “prefix-management ID†to some other words.
>>
>>         Is “ASA parameters" ok here?
>>
>>         I think some characteristics for this kind of low-level
>>         policies are
>>
>>         1.Network level parameters, need to be distributed by GRASP
>>         and ACP
>>
>>         2.Relatively static compared to intent, perhaps only needed
>>         to be configured once
>>
>>         3.Unavoidable in current network environment, mostly for
>>         establishing the network infrastructure
>>
>>         4.Rarely need coordinations with others parameters, usually
>>         is closed related to one specific autonomic function
>>
>>         Welcome for comment. Your suggestions will be appreciated.
>>
>>         Best Regards
>>
>>         Zongpeng Du
>>
>>
>>         _______________________________________________
>>         Anima mailing list
>>         Anima@ietf.org 
>>         https://www.ietf.org/mailman/listinfo/anima
>>
>>
>>
>>
>>     -- 
>>
>>     regards,
>>
>>     John
>>
>
>     -- 
>
>     Laurent Ciavaglia
>
>     Senior Research Manager
>
>     Bell Labs, Nokia
>
>     +33 160 402 636
>
>     Route de Villejust | 91620 Nozay | France
>
>     LinkedIn: laurentciavaglia
>     
>
>
>
>
> -- 
> regards,
> John

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------020101060608030109080505
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Hi John,

      

      Thanks for your comments.

      Please see below [lc2][/lc2].

    

    
On 24/03/2016 04:22, EXT John Strassner
      wrote:

    

    

      
      

        
Hi Laurent,

        
Thanks for your note. I'd be happy to join
          in a virtual meeting. Please see <jcs>..</jcs> for
          my observations.

        
On 23/03/2016
            11:57, EXT Michael Behringer (mbehring) wrote:

        
John, 

        
[stripped]

        

        
 More
            fundamentally, I think it is misleading to imagine that
            intent policies will be contained in one "file" (it is
            possible to do so but impractical/cumbersome to manage). cf
            also the next comment.

            [/LC]

          <jcs>

          See my previous reply. I am more worried about the user model
          - constructing a single file means that the constructor has to
          know implementation details. I think that this is unreasonable
          for the intent writer. However, I see no reason why the intent
          compiler could not produce a single file, or sets of files
          that group similar intent (e.g., invoking multiple autonomic
          functions in the same ASA) together.

          < /jcs>

      

    

    [lc2]: Fully agree. I was thinking about the user model. The output
    of the intent compiler could be a single file, I am fine with
    that.[/lc2]

    

    

      

        
 

        
My
            suggestion: Intent is flooded to all nodes in a domain.

        
[LC]

            It is the "resolution" of the intent clauses that determines
            the domain/set of target "nodes".

            So in that order, yes the intent will be flooded to all the
            relevant nodes in the "intent policy domain". This domain
            may be all nodes in an ANIMA network, or a sub-set, only one
            node, or zero. This is contextual.

            There will be many intent policies for a
            given ANIMA network.

            [/LC]

        
<jcs>

        
Let's use a concept of:

        
 

        
   +--------+     +----------+    
            +-----------+

        
   | Intent |---->|  Intent  |---->|
            Autonomic |

        
   | Writer |     | Compiler |     | 
            Network  |

        
   +--------+     +----------+    
            +-----------+

        
 

        
I **think** what Laurent is saying is that
          once the Intent Compiler resolves the intent of the
          user/intent writer, the target of the intent policy is then
          identified. I agree with this. Even for the broad case of
          "intent fabric", I don't think that an intent writer need be
          conscious of this (vs. an autonomic function). Why should he
          or she care?

        
 

        
Laurent, please correct me if I am wrong
          here.

        
</jcs>

      

    

    [lc2]: You read me right.[/lc2]
    

      

        
 

        
 

        
The ASA
            owners define the content of what they want to push in their
            bit of

            the Intent. If we insist of full flooding, we encourage the
            right behaviour. And

            if someone really puts into his Intent “node A: do this;
            node B: do the other; node C: do the third†it is awkward
            enough for people to take a step back and think whether
            they’re doing the right thing J

        
[LC]

            I think we have a disconnect on what an intent look like and
            how it can be used.

            In my "mental model", there is not one "big" intent with
            sub-parts for each ASA, the ANI, etc... 

            Each ASA designer (owner) will define the content /
            specificity of its ASA functionality or links to the
            existing elements of the ANIMA intent policy model, but it
            does not mean that we must aggregate all parts into a single
            "file".

            What would be the reason to do that?

            

            In addition, there will not only be intent policies per
            functions, or for the ANI.

            There will be intent policies conveying the Service level,
            the Business level... these will not resolve/map 1-to-1 in a
            single ASA "policy".

            [/LC]

        
<jcs>

        
I don't think that you and Michael are
          really disagreeing (other than one implementation (the file),
          which could be one option of many).

        
 

        
Indeed, in FOCALE and other work I have
          done, once intent was compiled, we did group things together
          for a number of reasons (e.g., dependencies). I think that the
          Autonomic Functions should advertise their capabilities, or
          else, how does the Intent Compiler have a clue as to what gets
          shipped to what (not to mention, how does it translate intent
          into a form that is consumable by ASAs). This is even more
          important for Services - one could imagine that a single
          Service intent would need multiple Autonomic Functions to be
          implemented.

        
</jcs>

      

    

    [lc2]:I fully agree, weused a same approach in UNIVERSELF for the
    policy continuum from the operator high-level policy input to the
    service- and network-levels policy translation.

    and indeed, for the service-level this is a key aspect (mapping the
    autonomic functions contributing/involved/impacting the
    services).[/lc2]

    

    

      

        
 

        


          jcs>

            Do you mean that the intent writer will **know the
            difference**

            between the fabric and each autonomic function? I hope not,

            but perhaps I'm misunderstanding you...

            < /jcs> 

        
Not sure
            I understand your concern. Yes, in my model there is a
            section for the fabric, and sections for the various
            functions. Seems a natural structure?!? 

        
[LC]

            It is OK to have different intent-policies for the fabric
            and for the functions, but why put them all into one
            file/policy?

            [/LC]

        


          

        
<jcs>

        
See my previous reply. In short, I was
          worried about this requiring the intent writer to be aware of
          which autonomic functions were present, and the nature of the
          autonomic fabric. I don't care that much about the structure
          of the input. 

        
</jcs>

      

    

    [lc2]: I agree, it is still early to discuss the structure/format of
    the intent, but my point was to start thinking about what a common
    intent format would look like for the ASAs, to ensure a certain
    level of interoperability and automation in the processing of the
    intent. This is linked to what you say below/in another message on
    the ASA capabilities being advertised/registered.[/lc2]

    

      

        
 

        


          “Function ownerâ€: Yes, I’m
            suggesting to structure Intent around Autonomic Functions.
            This is not optimal from a scientific point of view, but
            aligns very well with the way how things are organised.
            Probably we need to discuss that in BA. But refer to my YANG
            model sent a few days ago. 

        
[LC]

            I fully agree on the focus per autonomic function, but it
            does not imply to have one single intent policy with all
            parts inside.

            [/LC]

        
<jcs>

        
My point is that the intent writer should
          not have to have detailed knowledge about how the autonomic
          network works, or what its constituent autonomic functions
          are. In my previous post, I talked about an "autonomic
          catalog" that the intent writer could use. This would hide the
          fact that a particular service in the autonomic function had a
          1:1 or 1:n correspondence to a set of autonomic functions.

        
</jcs>

      

    

    [lc2]: I agree with the catalog approach. This relates to what we
    have tried to push in different drafts with the life-cycle of ASA,
    where ASA need to register or be discovered, for governance,
    coordination or information/knowledge purposes.[/lc2]

    

    

      

        
 

        


          We should schedule some
            discussion time outside the ANIMA meeting window...

        
[LC]

            Yes. And a white board too!

            Unfortunately, I cannot attend physically this time.

            Can we schedule a face-to-face ad-hoc meeting?

            [/LC]

        
<jcs>

        
Count me in.

        
</jcs>

      

    

    [lc]: ok. I will see how to set this up, and will keep the
    WG/interested participants posted.[/lc2]

    thank, best regards, Laurent.

    

    

      

        
 

        
regards,

        
John

      

      


        
On Wed, Mar 23, 2016 at 4:54 AM,
          Laurent Ciavaglia <laurent.ciavaglia@nokia.com>
          wrote:

          

            
 Michael,

                

                There are some aspects I am not sure to fully understand
                / agree with (marked below with [LC]).

                

                Best regards, Laurent.

                

              

              
On 23/03/2016 11:57, EXT Michael Behringer (mbehring)
                wrote:

              

              

                

                  
John,

                      

                  [stripped]

                  

                  
The

                        “parameter†versus “policy†discussion: My view
                        is that we shouldn’t try to define exactly *what*
                        intent allows and doesn’t allow, since we’ve
                        heard from many sides that the boundary is not
                        clear, and it may be impossible to have a really
                        clear, usable definition. So my point was: Let’s
                        focus on the distribution model: Intent is
                        flooded to all nodes. Let’s NOT do selective
                        flooding, unicast Intent to specific nodes, etc,
                        because that would blur the boundary even worse.
                      

                

              

              

              [LC]

              Before stating that intent policies should be flooded to
              all nodes, can we first compare/evaluate the different
              options?

              -broadcast is simple...but how important is it compared to
              other criteria...?

              I see at least two issues:

              1. flooding a "big" intent file (cf your current tentative
              example) may be bandwidth hungry, if one considers how
              frequently such an intent file could be distributed (ASA
              stop/start, device join/leave, policy change/update...)
              compared to distributing only what is required / where it
              is required.

              2. security: flooding the entire file reveal information
              to target (possibly undesirable/rogue) entities that do
              not need it (and thus may mis-use it).

              

              More fundamentally, I think it is misleading to imagine
              that intent policies will be contained in one "file" (it
              is possible to do so but impractical/cumbersome to
              manage). cf also the next comment.

              [/LC]

              

              

                

                  

                  
My

                        suggestion: Intent is flooded to all nodes in a
                        domain.

                

              

              [LC]

              It is the "resolution" of the intent clauses that
              determines the domain/set of target "nodes".

              So in that order, yes the intent will be flooded to all
              the relevant nodes in the "intent policy domain". This
              domain may be all nodes in an ANIMA network, or a sub-set,
              only one node, or zero. This is contextual.

              There will be many intent policies for a
              given ANIMA network.

              [/LC]

              

              

                

                  

                        The ASA owners define the content of what they
                        want to push in their bit of the Intent. If we
                        insist of full flooding, we encourage the right
                        behaviour. And if someone really puts into his
                        Intent “node A: do this; node B: do the other;
                        node C: do the third†it is awkward enough for
                        people to take a step back and think whether
                        they’re doing the right thing  J

                

              

              [LC]

              I think we have a disconnect on what an intent look like
              and how it can be used.

              In my "mental model", there is not one "big" intent with
              sub-parts for each ASA, the ANI, etc... 

              Each ASA designer (owner) will define the content /
              specificity of its ASA functionality or links to the
              existing elements of the ANIMA intent policy model, but it
              does not mean that we must aggregate all parts into a
              single "file".

              What would be the reason to do that?

              

              In addition, there will not only be intent policies per
              functions, or for the ANI.

              There will be intent policies conveying the Service level,
              the Business level... these will not resolve/map 1-to-1 in
              a single ASA "policy".

              [/LC]

              

              

                

                  
<jcs>

                  
Do

                        you mean that the intent writer will **know the
                        difference**

                  
between

                        the fabric and each autonomic function? I hope
                        not,

                  
but

                        perhaps I'm misunderstanding you...

                  
</jcs>

                  
 

                  
Not

                        sure I understand your concern. Yes, in my model
                        there is a section for the fabric, and sections
                        for the various functions. Seems a natural
                        structure?!? 

                

              

              [LC]

              It is OK to have different intent-policies for the fabric
              and for the functions, but why put them all into one
              file/policy?

              [/LC]

              

                

                  

                  
 

                  
“Function

                        ownerâ€: Yes, I’m suggesting to structure Intent
                        around Autonomic Functions. This is not optimal
                        from a scientific point of view, but aligns very
                        well with the way how things are organised.
                        Probably we need to discuss that in BA. But
                        refer to my YANG model sent a few days ago. 

                

              

              [LC]

              I fully agree on the focus per autonomic function, but it
              does not imply to have one single intent policy with all
              parts inside.

              [/LC]

              

              

                

                  

                  
 

                  
We

                        should schedule some discussion time outside the
                        ANIMA meeting window...

                  
 

                

              

              [LC]

              Yes. And a white board too!

              Unfortunately, I cannot attend physically this time.

              Can we schedule a face-to-face ad-hoc meeting?

              [/LC]

              

              

              

                

                  
Michael

                  
 

                  
 

                  

                    

                      

                        
From: John Strassner [mailto:strazpdj@gmail.com]
                              

                              Sent:
                              22 March 2016 21:32

                              To:
                              Michael Behringer (mbehring) <mbehring@cisco.com>;
                              John Strassner <strazpdj@gmail.com>

                              Cc:
                              Duzongpeng <duzongpeng@huawei.com>;
                              anima@ietf.org;
                              Laurent Ciavaglia <laurent.ciavaglia@nokia.com>;
                              Jéferson Campos Nobre <jcnobre@inf.ufrgs.br>;
                              Sheng Jiang <jiangsheng@huawei.com>

                              Subject:
                              Re: [Anima] Does Autonomic network need
                              some intervention beyond intent

                      

                    

                    
 

                    

                      

                        
In
                              general, I agree with Michael. I'd like to
                              emphasize and expand on a couple of key
                              points:

                      

                      

                        
 

                      

                      

                        
 

                      

                      

                        
First,

                              ANIMA isn’t the only thing “influencingâ€
                              or “steering†a network, at least not
                              initially. There is still traditional
                              config, plus all the existing and emerging
                              SDN models. My view has always been that
                              ANIMA sets a “default†policy in a
                              network, and all other means are more
                              specific and can override intent. This is
                              described in RFC7575. Main point here:
                              ANIMA Intent isn’t the only tool in a
                              network. 

                        
 

                      

                      

                        
<jcs>

                      

                      

                        
Absolutely
                              agree. Note that even in the Congress case
                              I mentioned

                      

                      

                        
earlier,
                              it is NOT pushing config, and it is NOT
                              viewed as the

                      

                      

                        
"only"
                              policy

                      

                      

                        

                          
</jcs>

                        

                      

                      

                        
 

                      

                      

                        
 

                      

                      

                        
Intent

                              could perfectly well cover a high level
                              policy such as “all nodes of type x do
                              this; type y does thatâ€. (Which I think is
                              what you mention below.) When we started
                              with Intent, the vision was to keep it
                              very high level, more along the line “do
                              the right thing†J 

                        
 

                      

                      

                        

                          
<jcs>

                        

                        

                          
Agreed. Although
                                to me, this means that we need to
                                consider the types of actors that are
                                using intent. For example, most of my
                                CCIE and JCIE friends have no interest
                                in using an intent language for pushing
                                configuration. On the other hand, when I
                                talk to operators and application
                                developers, they would love to be able
                                to use a high-level intent and have the
                                intent engine take care of the details
                                for them.

                        

                        

                          
</jcs>

                        

                        

                          
 

                        

                      

                      

                        

                          
It
                                has become clear that in today’s
                                networking, people will want to push
                                more “parameters†than “policyâ€, and I
                                think we’ll just have to acknowledge
                                that.

                        

                        

                          
 

                        

                        

                          
<jcs>

                        

                        

                          
Sorry Michael, I
                                don't follow here. What do you mean

                        

                        

                          
by "parameters"?

                        

                        

                          
</jcs>

                        

                        

                          
 

                        

                        

                          

                            
 

                          

                          

                            
Personally

                                  I think it’ll pan out like this:
                                  Intent will have sections per
                                  autonomic function. An Intent parser
                                  will validate overall correctness
                                  (signature, etc), then segment the
                                  Intent into its bits: One part will be
                                  the autonomic fabric itself, then
                                  there will be bits for each autonomic
                                  function. 

                            
 

                          

                          

                            
<jcs>

                          

                          

                            
Do you mean
                                  that the intent writer will **know the
                                  difference**

                          

                          

                            
between the
                                  fabric and each autonomic function? I
                                  hope not,

                          

                          

                            
but perhaps I'm
                                  misunderstanding you...

                          

                          

                            
</jcs>

                          

                          

                            
 

                          

                          

                            
 

                          

                          

                            
So

                                  the Intent parser on a node will grab
                                  intent, see there is some Intent for
                                  autonomic function A, will see whether
                                  this is running locally, and if so,
                                  it’ll just pass that piece of Intent
                                  to that function. 

                          

                          

                            

                              
 

                            

                            

                              
<jcs>

                            

                            

                              
Agreed.

                            

                            

                              
</jcs>

                            

                            

                              
 

                            

                            

                              

                                
 

                              

                              

                                
In

                                      this way, a function owner defines
                                      both the intent itself, and how it
                                      is consumed by his/her function. 

                                
 

                              

                              

                                
<jcs>

                              

                              

                                
I'm not
                                      sure what you mean by "function
                                      owner". (Sorry to be

                              

                              

                                
picky, I'm
                                      likely under-caffeinated.)

                              

                              

                                
 

                              

                              

                                
The above
                                      seemed like an autonomous
                                      decision, made by the parser.

                              

                              

                                
I don't
                                      think that the intent writer
                                      should be cognizant of how many or

                              

                              

                                
what types
                                      of autonomic functions are
                                      present.

                              

                              

                                
</jcs>

                              

                              

                                
 

                              

                              

                                
 

                              

                              

                                

                                  
The

                                        downside is that there is no
                                        “control†and people can do what
                                        they want, in the worst case
                                        even push config, which is
                                        clearly NOT what Intent is meant
                                        to do. 

                                  
 

                                

                                

                                  
<jcs>

                                

                                

                                  
And
                                        hence, the "HAL 2000" scenario.
                                        :-)

                                

                                

                                  
</jcs>

                                

                                

                                  
 

                                

                                

                                  

                                    
 

                                  

                                  

                                    
I’d

                                          hope that people don’t start
                                          writing Intent like “node 17:
                                          here is your CLI; node 23:
                                          here is your CLIâ€; node 37:
                                          here is your CLIâ€. But what I
                                          describe above COULD be abused
                                          in this way. RFC7575 says this
                                          is NOT Intent. 

                                  

                                  

                                    
 

                                  

                                  

                                    
<jcs>

                                  

                                  

                                    
Absolutely

                                          agree!

                                  

                                  

                                    
</jcs>

                                  

                                  

                                    
 

                                  

                                

                              

                            

                          

                        

                      

                      

                        
 regards,

                        
John

                        
 

                        
 

                      

                    

                    

                      
 

                      

                        
On
                              Wed, Mar 16, 2016 at 1:16 AM, Michael
                              Behringer (mbehring) <mbehring@cisco.com>

                              wrote:

                        

                          

                            

                              
Hi

                                    Zongpeng, 

                              
 

                              
Yes,

                                    this is a recurring question and
                                    topic. Let me give my view on two
                                    levels: 

                              
 

                              
First,

                                    ANIMA isn’t the only thing
                                    “influencing†or “steering†a
                                    network, at least not initially.
                                    There is still traditional config,
                                    plus all the existing and emerging
                                    SDN models. My view has always been
                                    that ANIMA sets a “default†policy
                                    in a network, and all other means
                                    are more specific and can override
                                    intent. This is described in
                                    RFC7575. Main point here: ANIMA
                                    Intent isn’t the only tool in a
                                    network. 

                              
 

                              
Intent

                                    could perfectly well cover a high
                                    level policy such as “all nodes of
                                    type x do this; type y does thatâ€.
                                    (Which I think is what you mention
                                    below.) When we started with Intent,
                                    the vision was to keep it very high
                                    level, more along the line “do the
                                    right thing†J 
                                    It has become clear that in today’s
                                    networking, people will want to push
                                    more “parameters†than “policyâ€, and
                                    I think we’ll just have to
                                    acknowledge that. 

                              
 

                              
Personally

                                    I think it’ll pan out like this:
                                    Intent will have sections per
                                    autonomic function. An Intent parser
                                    will validate overall correctness
                                    (signature, etc), then segment the
                                    Intent into its bits: One part will
                                    be the autonomic fabric itself, then
                                    there will be bits for each
                                    autonomic function. 

                              
 

                              
So

                                    the Intent parser on a node will
                                    grab intent, see there is some
                                    Intent for autonomic function A,
                                    will see whether this is running
                                    locally, and if so, it’ll just pass
                                    that piece of Intent to that
                                    function. 

                              
 

                              
In

                                    this way, a function owner defines
                                    both the intent itself, and how it
                                    is consumed by his/her function. 

                              
 

                              
The

                                    downside is that there is no
                                    “control†and people can do what
                                    they want, in the worst case even
                                    push config, which is clearly NOT
                                    what Intent is meant to do. I’d hope
                                    that people don’t start writing
                                    Intent like “node 17: here is your
                                    CLI; node 23: here is your CLIâ€;
                                    node 37: here is your CLIâ€. But what
                                    I describe above COULD be abused in
                                    this way. RFC7575 says this is NOT
                                    Intent.  

                              
 

                              
In

                                    other words: Let’s not be too
                                    prescriptive, and hope that folks
                                    will do more or less reasonable
                                    things. 

                              
 

                              
I’ve

                                    been meaning to comment on the
                                    Intent drafts and possibly
                                    contribute, I’m just struggling with
                                    priorities at the moment (as you can
                                    see by my relative silence over the
                                    past weeks). 

                              
 

                              
Michael

                              
 

                              
 

                              
 

                              

                                

                                  

                                    
From: Anima [mailto:anima-bounces@ietf.org] On

                                              Behalf Of Duzongpeng

                                          Sent:
                                          16 March 2016 02:38

                                          To:
                                        anima@ietf.org

                                          Cc:
                                          Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; Jéferson
                                          Campos Nobre <jcnobre@inf.ufrgs.br>; Sheng Jiang
                                          <jiangsheng@huawei.com>

                                          Subject:
                                          [Anima] Does Autonomic network
                                          need some intervention beyond
                                          intent

                                  

                                

                                
 

                                
Hello, everyone

                                
 

                                
         I understand
                                      that Autonomic network should be
                                      able to work with as less
                                      configuration as possible. In the
                                      ideal case, an Autonomic network
                                      should be able to work well while
                                      only intent is needed from human
                                      operators.

                                
 

                                
         However, I
                                      wonder whether Autonomic network
                                      may need some intervention from
                                      human operators through some
                                      low-level policies beyond intents.

                                
 

                                
         For example,
                                      in the https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is suggested
                                      that the prefix lengths for the
                                      CSG, ASG, RSG (different roles in
                                      IP RAN) can be assigned as an
                                      "intent". These configuration
                                      parameters need to be distributed
                                      in the autonomic domain to
                                      influence the detail
                                      configurations on each autonomic
                                      node.

                                
 

                                
         As intent is
                                      described as an abstract,
                                      declarative, high-level policy
                                      used to operate an autonomic
                                      domain, such as an enterprise
                                      network. Perhaps we can rename the
                                      “intent†in the “prefix-management
                                      ID†to some other words.

                                
 

                                
Is “ASA parameters"
                                      ok here? 

                                
 

                                
I think some
                                      characteristics for this kind of
                                      low-level policies are

                                
1.       Network level
                                    parameters, need to be distributed
                                    by GRASP and ACP

                                
2.       Relatively static
                                    compared to intent, perhaps only
                                    needed to be configured once

                                
3.       Unavoidable in current
                                    network environment, mostly for
                                    establishing the network
                                    infrastructure

                                
4.       Rarely need
                                    coordinations with others
                                    parameters, usually is closed
                                    related to one specific autonomic
                                    function

                                
         

                                
Welcome for comment.
                                      Your suggestions will be
                                      appreciated.

                                
 

                                
 

                                
Best Regards

                                
Zongpeng Du

                              

                            

                          

                          


_______________________________________________

                                Anima mailing list

                                Anima@ietf.org

                                https://www.ietf.org/mailman/listinfo/anima

                        

                      

                      


                            

                            

                            -- 

                      

                        

                          
regards,

                        

                        

                          
John

                        

                      

                    

                  

                

              

              

              
-- 

                

                  
Laurent Ciavaglia

                  
Senior
                      Research Manager

                  
Bell
                      Labs, Nokia

                  
 

                  
+33 160
                      402 636 

                  
Route
                      de Villejust | 91620 Nozay | France

                  
LinkedIn:
                    laurentciavaglia

                  
 

                

              

            

          

        

        

        

        

        -- 

        

          
regards,

          
John

        

      

    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------020101060608030109080505--


From nobody Thu Mar 24 16:46:09 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E30D212D875 for ; Thu, 24 Mar 2016 16:46:07 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.939
X-Spam-Level: 
X-Spam-Status: No, score=-5.939 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_FONT_FACE_BAD=0.981, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=unavailable autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3i4roNz3wrVs for ; Thu, 24 Mar 2016 16:46:03 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C3FAF12D0F9 for ; Thu, 24 Mar 2016 16:37:19 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id B8AFBA5DA2089; Thu, 24 Mar 2016 23:37:12 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2ONbHax017947 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 24 Mar 2016 23:37:17 GMT
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2ONbFBL007019 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 25 Mar 2016 00:37:15 +0100
Received: from [135.224.194.183] (135.239.27.40) by FR711WXCHHUB02.zeu.alcatel-lucent.com (135.239.2.112) with Microsoft SMTP Server (TLS) id 14.3.195.1; Fri, 25 Mar 2016 00:37:15 +0100
To: EXT John Strassner , Laurent Ciavaglia 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <56F28403.6060401@alcatel-lucent.com> 
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56F47A2A.6050807@alcatel-lucent.com>
Date: Fri, 25 Mar 2016 00:37:14 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: multipart/alternative; boundary="------------060102010106090907050407"
X-Originating-IP: [135.239.27.40]
Archived-At: 
Cc: Duzongpeng , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , "EXT Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 24 Mar 2016 23:46:08 -0000

--------------060102010106090907050407
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit

Hi John,

Thanks for your comments.
Please see below [lc2][/lc2].

On 24/03/2016 04:22, EXT John Strassner wrote:
>
> Hi Laurent,
>
> Thanks for your note. I'd be happy to join in a virtual meeting. 
> Please see .. for my observations.
>
> On 23/03/2016 11:57, EXT Michael Behringer (mbehring) wrote:
>
> John,
>
> /[stripped]/
>
>
> More fundamentally, I think it is misleading to imagine that intent 
> policies will be contained in one "file" (it is possible to do so but 
> impractical/cumbersome to manage). cf also the next comment.
> [/LC]
> 
> See my previous reply. I am more worried about the user model - 
> constructing a single file means that the constructor has to know 
> implementation details. I think that this is unreasonable for the 
> intent writer. However, I see no reason why the intent compiler could 
> not produce a single file, or sets of files that group similar intent 
> (e.g., invoking multiple autonomic functions in the same ASA) together.
> < /jcs>
[lc2]: Fully agree. I was thinking about the user model. The output of 
the intent compiler could be a single file, I am fine with that.[/lc2]

> My suggestion: Intent is flooded to all nodes in a domain.
>
> [LC]
> It is the "resolution" of the intent clauses that determines the 
> domain/set of target "nodes".
> So in that order, yes the intent will be flooded to all the relevant 
> nodes in the "intent policy domain". This domain may be all nodes in 
> an ANIMA network, or a sub-set, only one node, or zero. This is 
> contextual.
> There will be *_many_* intent policies for a given ANIMA network.
> [/LC]
>
> 
>
> Let's use a concept of:
>
>    +--------+     +----------+ +-----------+
>
>    | Intent |---->|  Intent  |---->| Autonomic |
>
>    | Writer |     | Compiler |     | Network  |
>
>    +--------+     +----------+ +-----------+
>
> I **think** what Laurent is saying is that once the Intent Compiler 
> resolves the intent of the user/intent writer, the target of the 
> intent policy is then identified. I agree with this. Even for the 
> broad case of "intent fabric", I don't think that an intent writer 
> need be conscious of this (vs. an autonomic function). Why should he 
> or she care?
>
> Laurent, please correct me if I am wrong here.
>
> 
>
[lc2]: You read me right.[/lc2]
>
> The ASA owners define the content of what they want to push in their 
> bit of
> the Intent. If we insist of full flooding, we encourage the right 
> behaviour. And
> if someone really puts into his Intent “node A: do this; node B: do 
> the other; node C: do the third†it is awkward enough for people to 
> take a step back and think whether they’re doing the right thing J
>
> [LC]
> I think we have a disconnect on what an intent look like and how it 
> can be used.
> In my "mental model", there is not one "big" intent with sub-parts for 
> each ASA, the ANI, etc...
> Each ASA designer (owner) will define the content / specificity of its 
> ASA functionality or links to the existing elements of the ANIMA 
> intent policy model, but it does not mean that we must aggregate all 
> parts into a single "file".
> What would be the reason to do that?
>
> In addition, there will not only be intent policies per functions, or 
> for the ANI.
> There will be intent policies conveying the Service level, the 
> Business level... these will not resolve/map 1-to-1 in a single ASA 
> "policy".
> [/LC]
>
> 
>
> I don't think that you and Michael are really disagreeing (other than 
> one implementation (the file), which could be one option of many).
>
> Indeed, in FOCALE and other work I have done, once intent was 
> compiled, we did group things together for a number of reasons (e.g., 
> dependencies). I think that the Autonomic Functions should advertise 
> their capabilities, or else, how does the Intent Compiler have a clue 
> as to what gets shipped to what (not to mention, how does it translate 
> intent into a form that is consumable by ASAs). This is even more 
> important for Services - one could imagine that a single Service 
> intent would need multiple Autonomic Functions to be implemented.
>
> 
>
[lc2]:I fully agree, weused a same approach in UNIVERSELF for the policy 
continuum from the operator high-level policy input to the service- and 
network-levels policy translation.
and indeed, for the service-level this is a key aspect (mapping the 
autonomic functions contributing/involved/impacting the services).[/lc2]

>
> jcs>
> Do you mean that the intent writer will **know the difference**
> between the fabric and each autonomic function? I hope not,
> but perhaps I'm misunderstanding you...
> < /jcs>
>
> Not sure I understand your concern. Yes, in my model there is a 
> section for the fabric, and sections for the various functions. Seems 
> a natural structure?!?
>
> [LC]
> It is OK to have different intent-policies for the fabric and for the 
> functions, but why put them all into one file/policy?
> [/LC]
> _
> _
>
> 
>
> See my previous reply. In short, I was worried about this requiring 
> the intent writer to be aware of which autonomic functions were 
> present, and the nature of the autonomic fabric. I don't care that 
> much about the structure of the input.
>
> 
>
[lc2]: I agree, it is still early to discuss the structure/format of the 
intent, but my point was to start thinking about what a common intent 
format would look like for the ASAs, to ensure a certain level of 
interoperability and automation in the processing of the intent. This is 
linked to what you say below/in another message on the ASA capabilities 
being advertised/registered.[/lc2]
>
>
> “Function ownerâ€: Yes, I’m suggesting to structure Intent around 
> Autonomic Functions. This is not optimal from a scientific point of 
> view, but aligns very well with the way how things are organised. 
> Probably we need to discuss that in BA. But refer to my YANG model 
> sent a few days ago.
>
> [LC]
> I fully agree on the focus per autonomic function, but it does not 
> imply to have one single intent policy with all parts inside.
> [/LC]
>
> 
>
> My point is that the intent writer should not have to have detailed 
> knowledge about how the autonomic network works, or what its 
> constituent autonomic functions are. In my previous post, I talked 
> about an "autonomic catalog" that the intent writer could use. This 
> would hide the fact that a particular service in the autonomic 
> function had a 1:1 or 1:n correspondence to a set of autonomic functions.
>
> 
>
[lc2]: I agree with the catalog approach. This relates to what we have 
tried to push in different drafts with the life-cycle of ASA, where ASA 
need to register or be discovered, for governance, coordination or 
information/knowledge purposes.[/lc2]

>
> We should schedule some discussion time outside the ANIMA meeting 
> window...
>
> [LC]
> Yes. And a white board too!
> Unfortunately, I cannot attend physically this time.
> Can we schedule a face-to-face ad-hoc meeting?
> [/LC]
>
> 
>
> Count me in.
>
> 
>
[lc]: ok. I will see how to set this up, and will keep the WG/interested 
participants posted.[/lc2]
Thanks, best regards, Laurent.

> regards,
>
> John
>
>
> On Wed, Mar 23, 2016 at 4:54 AM, Laurent Ciavaglia 
> > wrote:
>
>     Michael,
>
>     There are some aspects I am not sure to fully understand / agree
>     with (marked below with [LC]).
>
>     Best regards, Laurent.
>
>
>     On 23/03/2016 11:57, EXT Michael Behringer (mbehring) wrote:
>>
>>     John,
>>
>>     /[stripped]/
>>
>>     The “parameter†versus “policy†discussion: My view is that we
>>     shouldn’t try to define exactly **what** intent allows and
>>     doesn’t allow, since we’ve heard from many sides that the
>>     boundary is not clear, and it may be impossible to have a really
>>     clear, usable definition. So my point was: Let’s focus on the
>>     distribution model: Intent is flooded to all nodes. Let’s NOT do
>>     selective flooding, unicast Intent to specific nodes, etc,
>>     because that would blur the boundary even worse.
>>
>
>     [LC]
>     Before stating that intent policies should be flooded to all
>     nodes, can we first compare/evaluate the different options?
>     -broadcast is simple...but how important is it compared to other
>     criteria...?
>     I see at least two issues:
>     1. flooding a "big" intent file (cf your current tentative
>     example) may be bandwidth hungry, if one considers how frequently
>     such an intent file could be distributed (ASA stop/start, device
>     join/leave, policy change/update...) compared to distributing only
>     what is required / where it is required.
>     2. security: flooding the entire file reveal information to target
>     (possibly undesirable/rogue) entities that do not need it (and
>     thus may mis-use it).
>
>     More fundamentally, I think it is misleading to imagine that
>     intent policies will be contained in one "file" (it is possible to
>     do so but impractical/cumbersome to manage). cf also the next comment.
>     [/LC]
>
>>     My suggestion: Intent is flooded to all nodes in a domain.
>>
>     [LC]
>     It is the "resolution" of the intent clauses that determines the
>     domain/set of target "nodes".
>     So in that order, yes the intent will be flooded to all the
>     relevant nodes in the "intent policy domain". This domain may be
>     all nodes in an ANIMA network, or a sub-set, only one node, or
>     zero. This is contextual.
>     There will be _*many*_ intent policies for a given ANIMA network.
>     [/LC]
>
>>     The ASA owners define the content of what they want to push in
>>     their bit of the Intent. If we insist of full flooding, we
>>     encourage the right behaviour. And if someone really puts into
>>     his Intent “node A: do this; node B: do the other; node C: do the
>>     third†it is awkward enough for people to take a step back and
>>     think whether they’re doing the right thing J
>>
>     [LC]
>     I think we have a disconnect on what an intent look like and how
>     it can be used.
>     In my "mental model", there is not one "big" intent with sub-parts
>     for each ASA, the ANI, etc...
>     Each ASA designer (owner) will define the content / specificity of
>     its ASA functionality or links to the existing elements of the
>     ANIMA intent policy model, but it does not mean that we must
>     aggregate all parts into a single "file".
>     What would be the reason to do that?
>
>     In addition, there will not only be intent policies per functions,
>     or for the ANI.
>     There will be intent policies conveying the Service level, the
>     Business level... these will not resolve/map 1-to-1 in a single
>     ASA "policy".
>     [/LC]
>
>>     
>>
>>     Do you mean that the intent writer will **know the difference**
>>
>>     between the fabric and each autonomic function? I hope not,
>>
>>     but perhaps I'm misunderstanding you...
>>
>>     
>>
>>     Not sure I understand your concern. Yes, in my model there is a
>>     section for the fabric, and sections for the various functions.
>>     Seems a natural structure?!?
>>
>     [LC]
>     It is OK to have different intent-policies for the fabric and for
>     the functions, but why put them all into one file/policy?
>     [/LC]
>>
>>     “Function ownerâ€: Yes, I’m suggesting to structure Intent around
>>     Autonomic Functions. This is not optimal from a scientific point
>>     of view, but aligns very well with the way how things are
>>     organised. Probably we need to discuss that in BA. But refer to
>>     my YANG model sent a few days ago.
>>
>     [LC]
>     I fully agree on the focus per autonomic function, but it does not
>     imply to have one single intent policy with all parts inside.
>     [/LC]
>
>>     We should schedule some discussion time outside the ANIMA meeting
>>     window...
>>
>     [LC]
>     Yes. And a white board too!
>     Unfortunately, I cannot attend physically this time.
>     Can we schedule a face-to-face ad-hoc meeting?
>     [/LC]
>
>
>>     Michael
>>
>>     *From:*John Strassner [mailto:strazpdj@gmail.com]
>>     *Sent:* 22 March 2016 21:32
>>     *To:* Michael Behringer (mbehring) 
>>     ; John Strassner 
>>     
>>     *Cc:* Duzongpeng 
>>     ; anima@ietf.org
>>     ; Laurent Ciavaglia
>>     
>>     ; Jéferson Campos Nobre
>>      ; Sheng Jiang
>>      
>>     *Subject:* Re: [Anima] Does Autonomic network need some
>>     intervention beyond intent
>>
>>     In general, I agree with Michael. I'd like to emphasize and
>>     expand on a couple of key points:
>>
>>     First, ANIMA isn’t the only thing “influencing†or “steering†a
>>     network, at least not initially. There is still traditional
>>     config, plus all the existing and emerging SDN models. My view
>>     has always been that ANIMA sets a “default†policy in a network,
>>     and all other means are more specific and can override intent.
>>     This is described in RFC7575. Main point here: ANIMA Intent isn’t
>>     the only tool in a network.
>>
>>     
>>
>>     Absolutely agree. Note that even in the Congress case I mentioned
>>
>>     earlier, it is NOT pushing config, and it is NOT viewed as the
>>
>>     "only" policy
>>
>>     
>>
>>     Intent could perfectly well cover a high level policy such as
>>     “all nodes of type x do this; type y does thatâ€. (Which I think
>>     is what you mention below.) When we started with Intent, the
>>     vision was to keep it very high level, more along the line “do
>>     the right thing†J
>>
>>     
>>
>>     Agreed. Although to me, this means that we need to consider the
>>     types of actors that are using intent. For example, most of my
>>     CCIE and JCIE friends have no interest in using an intent
>>     language for pushing configuration. On the other hand, when I
>>     talk to operators and application developers, they would love to
>>     be able to use a high-level intent and have the intent engine
>>     take care of the details for them.
>>
>>     
>>
>>     It has become clear that in today’s networking, people will want
>>     to push more “parameters†than “policyâ€, and I think we’ll just
>>     have to acknowledge that.
>>
>>     
>>
>>     Sorry Michael, I don't follow here. What do you mean
>>
>>     by "parameters"?
>>
>>     
>>
>>     Personally I think it’ll pan out like this: Intent will have
>>     sections per autonomic function. An Intent parser will validate
>>     overall correctness (signature, etc), then segment the Intent
>>     into its bits: One part will be the autonomic fabric itself, then
>>     there will be bits for each autonomic function.
>>
>>     
>>
>>     Do you mean that the intent writer will **know the difference**
>>
>>     between the fabric and each autonomic function? I hope not,
>>
>>     but perhaps I'm misunderstanding you...
>>
>>     
>>
>>     So the Intent parser on a node will grab intent, see there is
>>     some Intent for autonomic function A, will see whether this is
>>     running locally, and if so, it’ll just pass that piece of Intent
>>     to that function.
>>
>>     
>>
>>     Agreed.
>>
>>     
>>
>>     In this way, a function owner defines both the intent itself, and
>>     how it is consumed by his/her function.
>>
>>     
>>
>>     I'm not sure what you mean by "function owner". (Sorry to be
>>
>>     picky, I'm likely under-caffeinated.)
>>
>>     The above seemed like an autonomous decision, made by the parser.
>>
>>     I don't think that the intent writer should be cognizant of how
>>     many or
>>
>>     what types of autonomic functions are present.
>>
>>     
>>
>>     The downside is that there is no “control†and people can do what
>>     they want, in the worst case even push config, which is clearly
>>     NOT what Intent is meant to do.
>>
>>     
>>
>>     And hence, the "HAL 2000" scenario. :-)
>>
>>     
>>
>>     I’d hope that people don’t start writing Intent like “node 17:
>>     here is your CLI; node 23: here is your CLIâ€; node 37: here is
>>     your CLIâ€. But what I describe above COULD be abused in this way.
>>     RFC7575 says this is NOT Intent.
>>
>>     
>>
>>     Absolutely agree!
>>
>>     
>>
>>      regards,
>>
>>     John
>>
>>     On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring)
>>     > wrote:
>>
>>         Hi Zongpeng,
>>
>>         Yes, this is a recurring question and topic. Let me give my
>>         view on two levels:
>>
>>         First, ANIMA isn’t the only thing “influencing†or “steeringâ€
>>         a network, at least not initially. There is still traditional
>>         config, plus all the existing and emerging SDN models. My
>>         view has always been that ANIMA sets a “default†policy in a
>>         network, and all other means are more specific and can
>>         override intent. This is described in RFC7575. Main point
>>         here: ANIMA Intent isn’t the only tool in a network.
>>
>>         Intent could perfectly well cover a high level policy such as
>>         “all nodes of type x do this; type y does thatâ€. (Which I
>>         think is what you mention below.) When we started with
>>         Intent, the vision was to keep it very high level, more along
>>         the line “do the right thing†JIt has become clear that in
>>         today’s networking, people will want to push more
>>         “parameters†than “policyâ€, and I think we’ll just have to
>>         acknowledge that.
>>
>>         Personally I think it’ll pan out like this: Intent will have
>>         sections per autonomic function. An Intent parser will
>>         validate overall correctness (signature, etc), then segment
>>         the Intent into its bits: One part will be the autonomic
>>         fabric itself, then there will be bits for each autonomic
>>         function.
>>
>>         So the Intent parser on a node will grab intent, see there is
>>         some Intent for autonomic function A, will see whether this
>>         is running locally, and if so, it’ll just pass that piece of
>>         Intent to that function.
>>
>>         In this way, a function owner defines both the intent itself,
>>         and how it is consumed by his/her function.
>>
>>         The downside is that there is no “control†and people can do
>>         what they want, in the worst case even push config, which is
>>         clearly NOT what Intent is meant to do. I’d hope that people
>>         don’t start writing Intent like “node 17: here is your CLI;
>>         node 23: here is your CLIâ€; node 37: here is your CLIâ€. But
>>         what I describe above COULD be abused in this way. RFC7575
>>         says this is NOT Intent.
>>
>>         In other words: Let’s not be too prescriptive, and hope that
>>         folks will do more or less reasonable things.
>>
>>         I’ve been meaning to comment on the Intent drafts and
>>         possibly contribute, I’m just struggling with priorities at
>>         the moment (as you can see by my relative silence over the
>>         past weeks).
>>
>>         Michael
>>
>>         *From:*Anima
>>         [mailto:anima-bounces@ietf.org
>>         ] *On Behalf Of *Duzongpeng
>>         *Sent:* 16 March 2016 02:38
>>         *To:* anima@ietf.org 
>>         *Cc:* Laurent Ciavaglia
>>         <laurent.ciavaglia@nokia.com
>>         >; Jéferson Campos Nobre
>>         <jcnobre@inf.ufrgs.br
>>         >; Sheng Jiang
>>         <jiangsheng@huawei.com
>>         >
>>         *Subject:* [Anima] Does Autonomic network need some
>>         intervention beyond intent
>>
>>         Hello, everyone
>>
>>                  I understand that Autonomic network should be able
>>         to work with as less configuration as possible. In the ideal
>>         case, an Autonomic network should be able to work well while
>>         only intent is needed from human operators.
>>
>>                  However, I wonder whether Autonomic network may need
>>         some intervention from human operators through some low-level
>>         policies beyond intents.
>>
>>                  For example, in the
>>         https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,
>>         it is suggested that the prefix lengths for the CSG, ASG, RSG
>>         (different roles in IP RAN) can be assigned as an "intent".
>>         These configuration parameters need to be distributed in the
>>         autonomic domain to influence the detail configurations on
>>         each autonomic node.
>>
>>                  As intent is described as an abstract, declarative,
>>         high-level policy used to operate an autonomic domain, such
>>         as an enterprise network. Perhaps we can rename the “intentâ€
>>         in the “prefix-management ID†to some other words.
>>
>>         Is “ASA parameters" ok here?
>>
>>         I think some characteristics for this kind of low-level
>>         policies are
>>
>>         1.Network level parameters, need to be distributed by GRASP
>>         and ACP
>>
>>         2.Relatively static compared to intent, perhaps only needed
>>         to be configured once
>>
>>         3.Unavoidable in current network environment, mostly for
>>         establishing the network infrastructure
>>
>>         4.Rarely need coordinations with others parameters, usually
>>         is closed related to one specific autonomic function
>>
>>         Welcome for comment. Your suggestions will be appreciated.
>>
>>         Best Regards
>>
>>         Zongpeng Du
>>
>>
>>         _______________________________________________
>>         Anima mailing list
>>         Anima@ietf.org 
>>         https://www.ietf.org/mailman/listinfo/anima
>>
>>
>>
>>
>>     -- 
>>
>>     regards,
>>
>>     John
>>
>
>     -- 
>
>     Laurent Ciavaglia
>
>     Senior Research Manager
>
>     Bell Labs, Nokia
>
>     +33 160 402 636
>
>     Route de Villejust | 91620 Nozay | France
>
>     LinkedIn: laurentciavaglia
>     
>
>
>
>
> -- 
> regards,
> John

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------060102010106090907050407
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Hi John,

      

      Thanks for your comments.

      Please see below [lc2][/lc2].

    

    
On 24/03/2016 04:22, EXT John Strassner
      wrote:

    

    

      
      

        
Hi Laurent,

        
Thanks for your note. I'd be happy to join
          in a virtual meeting. Please see <jcs>..</jcs> for
          my observations.

        
On 23/03/2016
            11:57, EXT Michael Behringer (mbehring) wrote:

        
John, 

        
[stripped]

        

        
 More
            fundamentally, I think it is misleading to imagine that
            intent policies will be contained in one "file" (it is
            possible to do so but impractical/cumbersome to manage). cf
            also the next comment.

            [/LC]

          <jcs>

          See my previous reply. I am more worried about the user model
          - constructing a single file means that the constructor has to
          know implementation details. I think that this is unreasonable
          for the intent writer. However, I see no reason why the intent
          compiler could not produce a single file, or sets of files
          that group similar intent (e.g., invoking multiple autonomic
          functions in the same ASA) together.

          < /jcs>

      

    

    [lc2]: Fully agree. I was thinking about the user model. The output
    of the intent compiler could be a single file, I am fine with
    that.[/lc2]

    

    

      

        
 

        
My
            suggestion: Intent is flooded to all nodes in a domain.

        
[LC]

            It is the "resolution" of the intent clauses that determines
            the domain/set of target "nodes".

            So in that order, yes the intent will be flooded to all the
            relevant nodes in the "intent policy domain". This domain
            may be all nodes in an ANIMA network, or a sub-set, only one
            node, or zero. This is contextual.

            There will be many intent policies for a
            given ANIMA network.

            [/LC]

        
<jcs>

        
Let's use a concept of:

        
 

        
   +--------+     +----------+    
            +-----------+

        
   | Intent |---->|  Intent  |---->|
            Autonomic |

        
   | Writer |     | Compiler |     | 
            Network  |

        
   +--------+     +----------+    
            +-----------+

        
 

        
I **think** what Laurent is saying is that
          once the Intent Compiler resolves the intent of the
          user/intent writer, the target of the intent policy is then
          identified. I agree with this. Even for the broad case of
          "intent fabric", I don't think that an intent writer need be
          conscious of this (vs. an autonomic function). Why should he
          or she care?

        
 

        
Laurent, please correct me if I am wrong
          here.

        
</jcs>

      

    

    [lc2]: You read me right.[/lc2]
    

      

        
 

        
 

        
The ASA
            owners define the content of what they want to push in their
            bit of

            the Intent. If we insist of full flooding, we encourage the
            right behaviour. And

            if someone really puts into his Intent “node A: do this;
            node B: do the other; node C: do the third†it is awkward
            enough for people to take a step back and think whether
            they’re doing the right thing J

        
[LC]

            I think we have a disconnect on what an intent look like and
            how it can be used.

            In my "mental model", there is not one "big" intent with
            sub-parts for each ASA, the ANI, etc... 

            Each ASA designer (owner) will define the content /
            specificity of its ASA functionality or links to the
            existing elements of the ANIMA intent policy model, but it
            does not mean that we must aggregate all parts into a single
            "file".

            What would be the reason to do that?

            

            In addition, there will not only be intent policies per
            functions, or for the ANI.

            There will be intent policies conveying the Service level,
            the Business level... these will not resolve/map 1-to-1 in a
            single ASA "policy".

            [/LC]

        
<jcs>

        
I don't think that you and Michael are
          really disagreeing (other than one implementation (the file),
          which could be one option of many).

        
 

        
Indeed, in FOCALE and other work I have
          done, once intent was compiled, we did group things together
          for a number of reasons (e.g., dependencies). I think that the
          Autonomic Functions should advertise their capabilities, or
          else, how does the Intent Compiler have a clue as to what gets
          shipped to what (not to mention, how does it translate intent
          into a form that is consumable by ASAs). This is even more
          important for Services - one could imagine that a single
          Service intent would need multiple Autonomic Functions to be
          implemented.

        
</jcs>

      

    

    [lc2]:I fully agree, weused a same approach in UNIVERSELF for the
    policy continuum from the operator high-level policy input to the
    service- and network-levels policy translation.

    and indeed, for the service-level this is a key aspect (mapping the
    autonomic functions contributing/involved/impacting the
    services).[/lc2]

    

    

      

        
 

        


          jcs>

            Do you mean that the intent writer will **know the
            difference**

            between the fabric and each autonomic function? I hope not,

            but perhaps I'm misunderstanding you...

            < /jcs> 

        
Not sure
            I understand your concern. Yes, in my model there is a
            section for the fabric, and sections for the various
            functions. Seems a natural structure?!? 

        
[LC]

            It is OK to have different intent-policies for the fabric
            and for the functions, but why put them all into one
            file/policy?

            [/LC]

        


          

        
<jcs>

        
See my previous reply. In short, I was
          worried about this requiring the intent writer to be aware of
          which autonomic functions were present, and the nature of the
          autonomic fabric. I don't care that much about the structure
          of the input. 

        
</jcs>

      

    

    [lc2]: I agree, it is still early to discuss the structure/format of
    the intent, but my point was to start thinking about what a common
    intent format would look like for the ASAs, to ensure a certain
    level of interoperability and automation in the processing of the
    intent. This is linked to what you say below/in another message on
    the ASA capabilities being advertised/registered.[/lc2]

    

      

        
 

        


          “Function ownerâ€: Yes, I’m
            suggesting to structure Intent around Autonomic Functions.
            This is not optimal from a scientific point of view, but
            aligns very well with the way how things are organised.
            Probably we need to discuss that in BA. But refer to my YANG
            model sent a few days ago. 

        
[LC]

            I fully agree on the focus per autonomic function, but it
            does not imply to have one single intent policy with all
            parts inside.

            [/LC]

        
<jcs>

        
My point is that the intent writer should
          not have to have detailed knowledge about how the autonomic
          network works, or what its constituent autonomic functions
          are. In my previous post, I talked about an "autonomic
          catalog" that the intent writer could use. This would hide the
          fact that a particular service in the autonomic function had a
          1:1 or 1:n correspondence to a set of autonomic functions.

        
</jcs>

      

    

    [lc2]: I agree with the catalog approach. This relates to what we
    have tried to push in different drafts with the life-cycle of ASA,
    where ASA need to register or be discovered, for governance,
    coordination or information/knowledge purposes.[/lc2]

    

    

      

        
 

        


          We should schedule some
            discussion time outside the ANIMA meeting window...

        
[LC]

            Yes. And a white board too!

            Unfortunately, I cannot attend physically this time.

            Can we schedule a face-to-face ad-hoc meeting?

            [/LC]

        
<jcs>

        
Count me in.

        
</jcs>

      

    

    [lc]: ok. I will see how to set this up, and will keep the
    WG/interested participants posted.[/lc2]

    Thanks, best regards, Laurent.

    

    

      

        
 

        
regards,

        
John

      

      


        
On Wed, Mar 23, 2016 at 4:54 AM,
          Laurent Ciavaglia <laurent.ciavaglia@nokia.com>
          wrote:

          

            
 Michael,

                

                There are some aspects I am not sure to fully understand
                / agree with (marked below with [LC]).

                

                Best regards, Laurent.

                

              

              
On 23/03/2016 11:57, EXT Michael Behringer (mbehring)
                wrote:

              

              

                

                  
John,

                      

                  [stripped]

                  

                  
The

                        “parameter†versus “policy†discussion: My view
                        is that we shouldn’t try to define exactly *what*
                        intent allows and doesn’t allow, since we’ve
                        heard from many sides that the boundary is not
                        clear, and it may be impossible to have a really
                        clear, usable definition. So my point was: Let’s
                        focus on the distribution model: Intent is
                        flooded to all nodes. Let’s NOT do selective
                        flooding, unicast Intent to specific nodes, etc,
                        because that would blur the boundary even worse.
                      

                

              

              

              [LC]

              Before stating that intent policies should be flooded to
              all nodes, can we first compare/evaluate the different
              options?

              -broadcast is simple...but how important is it compared to
              other criteria...?

              I see at least two issues:

              1. flooding a "big" intent file (cf your current tentative
              example) may be bandwidth hungry, if one considers how
              frequently such an intent file could be distributed (ASA
              stop/start, device join/leave, policy change/update...)
              compared to distributing only what is required / where it
              is required.

              2. security: flooding the entire file reveal information
              to target (possibly undesirable/rogue) entities that do
              not need it (and thus may mis-use it).

              

              More fundamentally, I think it is misleading to imagine
              that intent policies will be contained in one "file" (it
              is possible to do so but impractical/cumbersome to
              manage). cf also the next comment.

              [/LC]

              

              

                

                  

                  
My

                        suggestion: Intent is flooded to all nodes in a
                        domain.

                

              

              [LC]

              It is the "resolution" of the intent clauses that
              determines the domain/set of target "nodes".

              So in that order, yes the intent will be flooded to all
              the relevant nodes in the "intent policy domain". This
              domain may be all nodes in an ANIMA network, or a sub-set,
              only one node, or zero. This is contextual.

              There will be many intent policies for a
              given ANIMA network.

              [/LC]

              

              

                

                  

                        The ASA owners define the content of what they
                        want to push in their bit of the Intent. If we
                        insist of full flooding, we encourage the right
                        behaviour. And if someone really puts into his
                        Intent “node A: do this; node B: do the other;
                        node C: do the third†it is awkward enough for
                        people to take a step back and think whether
                        they’re doing the right thing  J

                

              

              [LC]

              I think we have a disconnect on what an intent look like
              and how it can be used.

              In my "mental model", there is not one "big" intent with
              sub-parts for each ASA, the ANI, etc... 

              Each ASA designer (owner) will define the content /
              specificity of its ASA functionality or links to the
              existing elements of the ANIMA intent policy model, but it
              does not mean that we must aggregate all parts into a
              single "file".

              What would be the reason to do that?

              

              In addition, there will not only be intent policies per
              functions, or for the ANI.

              There will be intent policies conveying the Service level,
              the Business level... these will not resolve/map 1-to-1 in
              a single ASA "policy".

              [/LC]

              

              

                

                  
<jcs>

                  
Do

                        you mean that the intent writer will **know the
                        difference**

                  
between

                        the fabric and each autonomic function? I hope
                        not,

                  
but

                        perhaps I'm misunderstanding you...

                  
</jcs>

                  
 

                  
Not

                        sure I understand your concern. Yes, in my model
                        there is a section for the fabric, and sections
                        for the various functions. Seems a natural
                        structure?!? 

                

              

              [LC]

              It is OK to have different intent-policies for the fabric
              and for the functions, but why put them all into one
              file/policy?

              [/LC]

              

                

                  

                  
 

                  
“Function

                        ownerâ€: Yes, I’m suggesting to structure Intent
                        around Autonomic Functions. This is not optimal
                        from a scientific point of view, but aligns very
                        well with the way how things are organised.
                        Probably we need to discuss that in BA. But
                        refer to my YANG model sent a few days ago. 

                

              

              [LC]

              I fully agree on the focus per autonomic function, but it
              does not imply to have one single intent policy with all
              parts inside.

              [/LC]

              

              

                

                  

                  
 

                  
We

                        should schedule some discussion time outside the
                        ANIMA meeting window...

                  
 

                

              

              [LC]

              Yes. And a white board too!

              Unfortunately, I cannot attend physically this time.

              Can we schedule a face-to-face ad-hoc meeting?

              [/LC]

              

              

              

                

                  
Michael

                  
 

                  
 

                  

                    

                      

                        
From: John Strassner [mailto:strazpdj@gmail.com]
                              

                              Sent:
                              22 March 2016 21:32

                              To:
                              Michael Behringer (mbehring) <mbehring@cisco.com>;
                              John Strassner <strazpdj@gmail.com>

                              Cc:
                              Duzongpeng <duzongpeng@huawei.com>;
                              anima@ietf.org;
                              Laurent Ciavaglia <laurent.ciavaglia@nokia.com>;
                              Jéferson Campos Nobre <jcnobre@inf.ufrgs.br>;
                              Sheng Jiang <jiangsheng@huawei.com>

                              Subject:
                              Re: [Anima] Does Autonomic network need
                              some intervention beyond intent

                      

                    

                    
 

                    

                      

                        
In
                              general, I agree with Michael. I'd like to
                              emphasize and expand on a couple of key
                              points:

                      

                      

                        
 

                      

                      

                        
 

                      

                      

                        
First,

                              ANIMA isn’t the only thing “influencingâ€
                              or “steering†a network, at least not
                              initially. There is still traditional
                              config, plus all the existing and emerging
                              SDN models. My view has always been that
                              ANIMA sets a “default†policy in a
                              network, and all other means are more
                              specific and can override intent. This is
                              described in RFC7575. Main point here:
                              ANIMA Intent isn’t the only tool in a
                              network. 

                        
 

                      

                      

                        
<jcs>

                      

                      

                        
Absolutely
                              agree. Note that even in the Congress case
                              I mentioned

                      

                      

                        
earlier,
                              it is NOT pushing config, and it is NOT
                              viewed as the

                      

                      

                        
"only"
                              policy

                      

                      

                        

                          
</jcs>

                        

                      

                      

                        
 

                      

                      

                        
 

                      

                      

                        
Intent

                              could perfectly well cover a high level
                              policy such as “all nodes of type x do
                              this; type y does thatâ€. (Which I think is
                              what you mention below.) When we started
                              with Intent, the vision was to keep it
                              very high level, more along the line “do
                              the right thing†J 

                        
 

                      

                      

                        

                          
<jcs>

                        

                        

                          
Agreed. Although
                                to me, this means that we need to
                                consider the types of actors that are
                                using intent. For example, most of my
                                CCIE and JCIE friends have no interest
                                in using an intent language for pushing
                                configuration. On the other hand, when I
                                talk to operators and application
                                developers, they would love to be able
                                to use a high-level intent and have the
                                intent engine take care of the details
                                for them.

                        

                        

                          
</jcs>

                        

                        

                          
 

                        

                      

                      

                        

                          
It
                                has become clear that in today’s
                                networking, people will want to push
                                more “parameters†than “policyâ€, and I
                                think we’ll just have to acknowledge
                                that.

                        

                        

                          
 

                        

                        

                          
<jcs>

                        

                        

                          
Sorry Michael, I
                                don't follow here. What do you mean

                        

                        

                          
by "parameters"?

                        

                        

                          
</jcs>

                        

                        

                          
 

                        

                        

                          

                            
 

                          

                          

                            
Personally

                                  I think it’ll pan out like this:
                                  Intent will have sections per
                                  autonomic function. An Intent parser
                                  will validate overall correctness
                                  (signature, etc), then segment the
                                  Intent into its bits: One part will be
                                  the autonomic fabric itself, then
                                  there will be bits for each autonomic
                                  function. 

                            
 

                          

                          

                            
<jcs>

                          

                          

                            
Do you mean
                                  that the intent writer will **know the
                                  difference**

                          

                          

                            
between the
                                  fabric and each autonomic function? I
                                  hope not,

                          

                          

                            
but perhaps I'm
                                  misunderstanding you...

                          

                          

                            
</jcs>

                          

                          

                            
 

                          

                          

                            
 

                          

                          

                            
So

                                  the Intent parser on a node will grab
                                  intent, see there is some Intent for
                                  autonomic function A, will see whether
                                  this is running locally, and if so,
                                  it’ll just pass that piece of Intent
                                  to that function. 

                          

                          

                            

                              
 

                            

                            

                              
<jcs>

                            

                            

                              
Agreed.

                            

                            

                              
</jcs>

                            

                            

                              
 

                            

                            

                              

                                
 

                              

                              

                                
In

                                      this way, a function owner defines
                                      both the intent itself, and how it
                                      is consumed by his/her function. 

                                
 

                              

                              

                                
<jcs>

                              

                              

                                
I'm not
                                      sure what you mean by "function
                                      owner". (Sorry to be

                              

                              

                                
picky, I'm
                                      likely under-caffeinated.)

                              

                              

                                
 

                              

                              

                                
The above
                                      seemed like an autonomous
                                      decision, made by the parser.

                              

                              

                                
I don't
                                      think that the intent writer
                                      should be cognizant of how many or

                              

                              

                                
what types
                                      of autonomic functions are
                                      present.

                              

                              

                                
</jcs>

                              

                              

                                
 

                              

                              

                                
 

                              

                              

                                

                                  
The

                                        downside is that there is no
                                        “control†and people can do what
                                        they want, in the worst case
                                        even push config, which is
                                        clearly NOT what Intent is meant
                                        to do. 

                                  
 

                                

                                

                                  
<jcs>

                                

                                

                                  
And
                                        hence, the "HAL 2000" scenario.
                                        :-)

                                

                                

                                  
</jcs>

                                

                                

                                  
 

                                

                                

                                  

                                    
 

                                  

                                  

                                    
I’d

                                          hope that people don’t start
                                          writing Intent like “node 17:
                                          here is your CLI; node 23:
                                          here is your CLIâ€; node 37:
                                          here is your CLIâ€. But what I
                                          describe above COULD be abused
                                          in this way. RFC7575 says this
                                          is NOT Intent. 

                                  

                                  

                                    
 

                                  

                                  

                                    
<jcs>

                                  

                                  

                                    
Absolutely

                                          agree!

                                  

                                  

                                    
</jcs>

                                  

                                  

                                    
 

                                  

                                

                              

                            

                          

                        

                      

                      

                        
 regards,

                        
John

                        
 

                        
 

                      

                    

                    

                      
 

                      

                        
On
                              Wed, Mar 16, 2016 at 1:16 AM, Michael
                              Behringer (mbehring) <mbehring@cisco.com>

                              wrote:

                        

                          

                            

                              
Hi

                                    Zongpeng, 

                              
 

                              
Yes,

                                    this is a recurring question and
                                    topic. Let me give my view on two
                                    levels: 

                              
 

                              
First,

                                    ANIMA isn’t the only thing
                                    “influencing†or “steering†a
                                    network, at least not initially.
                                    There is still traditional config,
                                    plus all the existing and emerging
                                    SDN models. My view has always been
                                    that ANIMA sets a “default†policy
                                    in a network, and all other means
                                    are more specific and can override
                                    intent. This is described in
                                    RFC7575. Main point here: ANIMA
                                    Intent isn’t the only tool in a
                                    network. 

                              
 

                              
Intent

                                    could perfectly well cover a high
                                    level policy such as “all nodes of
                                    type x do this; type y does thatâ€.
                                    (Which I think is what you mention
                                    below.) When we started with Intent,
                                    the vision was to keep it very high
                                    level, more along the line “do the
                                    right thing†J 
                                    It has become clear that in today’s
                                    networking, people will want to push
                                    more “parameters†than “policyâ€, and
                                    I think we’ll just have to
                                    acknowledge that. 

                              
 

                              
Personally

                                    I think it’ll pan out like this:
                                    Intent will have sections per
                                    autonomic function. An Intent parser
                                    will validate overall correctness
                                    (signature, etc), then segment the
                                    Intent into its bits: One part will
                                    be the autonomic fabric itself, then
                                    there will be bits for each
                                    autonomic function. 

                              
 

                              
So

                                    the Intent parser on a node will
                                    grab intent, see there is some
                                    Intent for autonomic function A,
                                    will see whether this is running
                                    locally, and if so, it’ll just pass
                                    that piece of Intent to that
                                    function. 

                              
 

                              
In

                                    this way, a function owner defines
                                    both the intent itself, and how it
                                    is consumed by his/her function. 

                              
 

                              
The

                                    downside is that there is no
                                    “control†and people can do what
                                    they want, in the worst case even
                                    push config, which is clearly NOT
                                    what Intent is meant to do. I’d hope
                                    that people don’t start writing
                                    Intent like “node 17: here is your
                                    CLI; node 23: here is your CLIâ€;
                                    node 37: here is your CLIâ€. But what
                                    I describe above COULD be abused in
                                    this way. RFC7575 says this is NOT
                                    Intent.  

                              
 

                              
In

                                    other words: Let’s not be too
                                    prescriptive, and hope that folks
                                    will do more or less reasonable
                                    things. 

                              
 

                              
I’ve

                                    been meaning to comment on the
                                    Intent drafts and possibly
                                    contribute, I’m just struggling with
                                    priorities at the moment (as you can
                                    see by my relative silence over the
                                    past weeks). 

                              
 

                              
Michael

                              
 

                              
 

                              
 

                              

                                

                                  

                                    
From: Anima [mailto:anima-bounces@ietf.org] On

                                              Behalf Of Duzongpeng

                                          Sent:
                                          16 March 2016 02:38

                                          To:
                                        anima@ietf.org

                                          Cc:
                                          Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; Jéferson
                                          Campos Nobre <jcnobre@inf.ufrgs.br>; Sheng Jiang
                                          <jiangsheng@huawei.com>

                                          Subject:
                                          [Anima] Does Autonomic network
                                          need some intervention beyond
                                          intent

                                  

                                

                                
 

                                
Hello, everyone

                                
 

                                
         I understand
                                      that Autonomic network should be
                                      able to work with as less
                                      configuration as possible. In the
                                      ideal case, an Autonomic network
                                      should be able to work well while
                                      only intent is needed from human
                                      operators.

                                
 

                                
         However, I
                                      wonder whether Autonomic network
                                      may need some intervention from
                                      human operators through some
                                      low-level policies beyond intents.

                                
 

                                
         For example,
                                      in the https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is suggested
                                      that the prefix lengths for the
                                      CSG, ASG, RSG (different roles in
                                      IP RAN) can be assigned as an
                                      "intent". These configuration
                                      parameters need to be distributed
                                      in the autonomic domain to
                                      influence the detail
                                      configurations on each autonomic
                                      node.

                                
 

                                
         As intent is
                                      described as an abstract,
                                      declarative, high-level policy
                                      used to operate an autonomic
                                      domain, such as an enterprise
                                      network. Perhaps we can rename the
                                      “intent†in the “prefix-management
                                      ID†to some other words.

                                
 

                                
Is “ASA parameters"
                                      ok here? 

                                
 

                                
I think some
                                      characteristics for this kind of
                                      low-level policies are

                                
1.       Network level
                                    parameters, need to be distributed
                                    by GRASP and ACP

                                
2.       Relatively static
                                    compared to intent, perhaps only
                                    needed to be configured once

                                
3.       Unavoidable in current
                                    network environment, mostly for
                                    establishing the network
                                    infrastructure

                                
4.       Rarely need
                                    coordinations with others
                                    parameters, usually is closed
                                    related to one specific autonomic
                                    function

                                
         

                                
Welcome for comment.
                                      Your suggestions will be
                                      appreciated.

                                
 

                                
 

                                
Best Regards

                                
Zongpeng Du

                              

                            

                          

                          


_______________________________________________

                                Anima mailing list

                                Anima@ietf.org

                                https://www.ietf.org/mailman/listinfo/anima

                        

                      

                      


                            

                            

                            -- 

                      

                        

                          
regards,

                        

                        

                          
John

                        

                      

                    

                  

                

              

              

              
-- 

                

                  
Laurent Ciavaglia

                  
Senior
                      Research Manager

                  
Bell
                      Labs, Nokia

                  
 

                  
+33 160
                      402 636 

                  
Route
                      de Villejust | 91620 Nozay | France

                  
LinkedIn:
                    laurentciavaglia

                  
 

                

              

            

          

        

        

        

        

        -- 

        

          
regards,

          
John

        

      

    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------060102010106090907050407--


From nobody Thu Mar 24 18:08:38 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0B03D12D84F for ; Wed, 23 Mar 2016 12:48:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.311
X-Spam-Level: 
X-Spam-Status: No, score=-4.311 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=ufrgs.br
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vNWC70msidBb for ; Wed, 23 Mar 2016 12:48:55 -0700 (PDT)
Received: from delivery2.ufrgs.br (delivery2.ufrgs.br [143.54.2.212]) by ietfa.amsl.com (Postfix) with ESMTP id 4316812D7DE for ; Wed, 23 Mar 2016 12:48:51 -0700 (PDT)
Received: from delivery2.ufrgs.br (localhost [127.0.0.1]) by delivery2.ufrgs.br (Postfix) with ESMTP id 295E020A3F6 for ; Wed, 23 Mar 2016 16:48:49 -0300 (BRT)
DKIM-Signature: v=1; a=rsa-sha256; c=simple/simple; d=ufrgs.br; s=mail; t=1458762530; bh=rEuF0OXW3GVFxSv7fKPJPmep+q4L7GH5HurRCwUUfMk=; h=MIME-Version:References:In-Reply-To:From:Date:Message-ID:Subject: To:Cc:Content-Type; b=wOs1PiIjOLMQcuu4tim3nUpSm9Gl31ddMsO9vgk4m+XQfwsSGFbudh5vhqptNeR4B mBXh/UwrTEvTqx0QvByGGFS4D426s/ekYlZS7ZIXoVOz2TNnBiLqzJjyjw6PAe9PT3 OdRyWrEinm9AIcfE1vJMuHfxwgglnODhcquhDu+M=
Received: from msa1.ufrgs.br (msa1.ufrgs.br [143.54.2.208]) by delivery2.ufrgs.br (Postfix) with ESMTP id 6DEE5300814 for ; Wed, 23 Mar 2016 16:48:49 -0300 (BRT)
Received: by mail-io0-f174.google.com with SMTP id m184so61577194iof.1 for ; Wed, 23 Mar 2016 12:48:48 -0700 (PDT)
X-Gm-Message-State: AD7BkJKF/Se5EhWdnoAXeCScjVXYC7es/kHD7jR70TD4DFCIVijQVrcW/FKnrldD1DFCDiuhTpa07DP7/Cu1Jw==
X-Received: by 10.107.26.201 with SMTP id a192mr5814571ioa.13.1458762527163; Wed, 23 Mar 2016 12:48:47 -0700 (PDT)
MIME-Version: 1.0
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com> <56F0E9CF.7070602@tzi.org> <56F1969D.1060804@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5E464@nkgeml514-mbx.china.huawei.com> <56F259F6.5000807@alcatel-lucent.com> <56F2EE26.9030107@gmail.com>
In-Reply-To: <56F2EE26.9030107@gmail.com>
From: =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= 
Date: Wed, 23 Mar 2016 19:48:37 +0000
X-Gmail-Original-Message-ID: 
Message-ID: 
To: Brian E Carpenter ,  Laurent Ciavaglia , "Liubing (Leo)" , Carsten Bormann 
Content-Type: multipart/alternative; boundary=001a113ff0bc3e91d6052ebc9e4f
X-AV-Checked: ClamAV using ClamSMTP
Archived-At: 
X-Mailman-Approved-At: Thu, 24 Mar 2016 18:08:37 -0700
Cc: "anima@ietf.org" , Michael Behringer 
Subject: Re: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 23 Mar 2016 19:48:58 -0000

--001a113ff0bc3e91d6052ebc9e4f
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Em qua, 23 de mar de 2016 =C3=A0s 16:27, Brian E Carpenter <
brian.e.carpenter@gmail.com> escreveu:

> On 23/03/2016 21:55, Laurent Ciavaglia wrote:
> ...
> >> b)  Sophisticated operation on Intent, e.g. partial update, deleting,
> query etc.
>
> I'm pretty sure we will need partial update and deletion. I don't expect =
we
> will need query.
>

Probably, query would be needed for (self?) monitoring and intent
verification and consistency checks.


>
> ...
> > As for RESTful "operations", this is a possible approach but we should
> probably first discuss the requirements before choosing a type of
> communication scheme (cf characteristics of autonomic networks).
>
> I agree. RESTful has become a sort of religion recently but it's really
> just a
> way of almost obtaining transactional integrity across proxies and caches=
.
> It may not be what we need (for example, we might need semantics more lik=
e
> DNCP).
> It may be that simple flooding (already supported by GRASP) will be
> sufficient,
> if the update and deletion operations are atomic.
>
>
+ 1


>    Brian
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>

--001a113ff0bc3e91d6052ebc9e4f
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Em qua, 23 de =
mar de 2016 =C3=A0s 16:27, Brian E Carpenter <brian.e.carpenter@gmail.com> escreveu:
On 23/03/2016 21:55, Laurent Ciavaglia wrot=
e:

...

>> b)=C2=A0 Sophisticated operation on Intent, e.g. partial update, d=
eleting, query etc.



I'm pretty sure we will need partial update and deletion. I don't e=
xpect we

will need query.

Probably, query would =
be needed for (self?) monitoring and intent verification and consistency ch=
ecks.
=C2=A0



...

> As for RESTful "operations", this is a possible approach but=
 we should probably first discuss the requirements before choosing a type o=
f communication scheme (cf characteristics of autonomic networks).



I agree. RESTful has become a sort of religion recently but it's really=
 just a

way of almost obtaining transactional integrity across proxies and caches.<=
br>
It may not be what we need (for example, we might need semantics more like =
DNCP).

It may be that simple flooding (already supported by GRASP) will be suffici=
ent,

if the update and deletion operations are atomic.



+ 1
=C2=A0

=C2=A0 =C2=A0Brian



_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima




--001a113ff0bc3e91d6052ebc9e4f--


From nobody Thu Mar 24 20:31:39 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0B67D12D0BE for ; Thu, 24 Mar 2016 20:31:38 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.23
X-Spam-Level: 
X-Spam-Status: No, score=-4.23 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hjcgMS0PNBX8 for ; Thu, 24 Mar 2016 20:31:35 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D904412D0A0 for ; Thu, 24 Mar 2016 20:31:33 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml703-cah.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CLF46403; Fri, 25 Mar 2016 03:31:31 +0000 (GMT)
Received: from NKGEML408-HUB.china.huawei.com (10.98.56.39) by lhreml703-cah.china.huawei.com (10.201.5.104) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 25 Mar 2016 03:31:30 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml408-hub.china.huawei.com ([10.98.56.39]) with mapi id 14.03.0235.001; Fri, 25 Mar 2016 11:31:19 +0800
From: Duzongpeng 
To: John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgEXJoaAAK+NZRA=
Date: Fri, 25 Mar 2016 03:31:19 +0000
Message-ID: 
References:  
In-Reply-To: 
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.149.226]
Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFC8F0nkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A090204.56F4B114.0021, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 459087b6c3337b62f73452de91af5c9c
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 03:31:38 -0000

--_000_BAFEC9523F57BC48A51C20226A5589575FDFC8F0nkgeml514mbxchi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFC8F0nkgeml514mbxchi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFC8F0nkgeml514mbxchi_--


From nobody Fri Mar 25 00:54:48 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 48A1B12D55E for ; Fri, 25 Mar 2016 00:54:47 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0oqYIhP_m9km for ; Fri, 25 Mar 2016 00:54:37 -0700 (PDT)
Received: from rcdn-iport-2.cisco.com (rcdn-iport-2.cisco.com [173.37.86.73]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3BC3A12D09F for ; Fri, 25 Mar 2016 00:54:37 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=3446; q=dns/txt; s=iport; t=1458892477; x=1460102077; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=nH9TnGRdioP5oi0R0dTLOQJ9suAPQNsrn/YoTg4VPMs=; b=e2dg352WSVA5c075ya2PzZjNYMmo0KtFw3KM1VfLHjGCHAHUobHwsx+f i0HdW6LG+X+W2i3enmtmePhh7xTJ8khqYgciug5ECb61ATZGteAzGqNIF clndrKpiPh8XAsHbRQkr8GK9jQfUhUE4EDEwapoDuuQSZbSOWus2tfsa2 g=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D3AQAc7vRW/4QNJK1egzOBUAavBotOA?= =?us-ascii?q?Q2BcIYNAhyBFjgUAQEBAQEBAWQnhEEBAQEDASMRRQUHBAIBCBEEAQEBAgIjAwI?= =?us-ascii?q?CAh8RFAEICAIEAQ0FCIgKAwoIsCqLVg2EdwEBAQEBAQEBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?RV8hSKERII+gWYugmqCVgWXLzEBjA6Bbo8ShzWHVAEeAQFCgjCBNWwBiFd+AQE?= =?us-ascii?q?B?=
X-IronPort-AV: E=Sophos;i="5.24,389,1454976000"; d="scan'208";a="89376086"
Received: from alln-core-10.cisco.com ([173.36.13.132]) by rcdn-iport-2.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 25 Mar 2016 07:54:35 +0000
Received: from XCH-RCD-006.cisco.com (xch-rcd-006.cisco.com [173.37.102.16]) by alln-core-10.cisco.com (8.14.5/8.14.5) with ESMTP id u2P7sZJP004464 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 25 Mar 2016 07:54:35 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-006.cisco.com (173.37.102.16) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 25 Mar 2016 02:54:34 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 25 Mar 2016 02:54:34 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , Laurent Ciavaglia , "EXT Joel M. Halpern" 
Thread-Topic: Re-flooding [was: Does Autonomic network need some intervention beyond intent]
Thread-Index: AQHRhgV/xjAWxQrPfUKm/+t8QaUqkZ9pySWA
Date: Fri, 25 Mar 2016 07:54:34 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com> <56F409D0.9030900@joelhalpern.com> <56F41351.8080309@alcatel-lucent.com> <56F4437C.7000803@gmail.com>
In-Reply-To: <56F4437C.7000803@gmail.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Cc: Duzongpeng , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Re-flooding [was: Does Autonomic network need some intervention beyond intent]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 07:54:47 -0000
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From nobody Fri Mar 25 01:00:16 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9D3FE12D1B5 for ; Fri, 25 Mar 2016 01:00:14 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.52
X-Spam-Level: 
X-Spam-Status: No, score=-14.52 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_KAM_HTML_FONT_INVALID=0.01, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cs_XCmmuSfm8 for ; Fri, 25 Mar 2016 01:00:12 -0700 (PDT)
Received: from alln-iport-8.cisco.com (alln-iport-8.cisco.com [173.37.142.95]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6A7C812D193 for ; Fri, 25 Mar 2016 01:00:12 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=17894; q=dns/txt; s=iport; t=1458892812; x=1460102412; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=LyC50hvkb9lybbAx/aV8MiAd9tezuEGUwuIvZ0nKVZg=; b=ik7u7sqXwZypQbBTVScjLzTYiPekxL/xyH+YaIWzeHgqI09FhTYSucHM fTKB3XIDKhejRt+UdguftDkLtdCQKhf583RceM0ddaJpXehsv25JzOdSS rBjVUDBoVxh+UibvgL2NxCBXgi3OFlKQZp/cpsrdBP8PfJPkB2ys5QDdp g=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0ALAgBH7/RW/4ENJK1egmdMU30GrwaGY?= =?us-ascii?q?IRuAQ2BcIYNAhyBFzgUAQEBAQEBAWQnhEEBAQEEIwpMEAIBCBEEAQEoAwICAh8?= =?us-ascii?q?RFAkIAgQBDQUIiAoDErAqi1YNhHcBAQEBAQEBAQEBAQEBAQEBAQEBAQEVhh6ER?= =?us-ascii?q?II+gWYuIIJKglYFkwaEKTEBjA6Bbo8ShzWHVAEeAQFCgjCBNWwBiFd+AQEB?=
X-IronPort-AV: E=Sophos;i="5.24,389,1454976000";  d="scan'208,217";a="253353731"
Received: from alln-core-9.cisco.com ([173.36.13.129]) by alln-iport-8.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 25 Mar 2016 08:00:11 +0000
Received: from XCH-ALN-007.cisco.com (xch-aln-007.cisco.com [173.36.7.17]) by alln-core-9.cisco.com (8.14.5/8.14.5) with ESMTP id u2P80B2a006706 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 25 Mar 2016 08:00:11 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-007.cisco.com (173.36.7.17) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 25 Mar 2016 03:00:10 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 25 Mar 2016 03:00:10 -0500
From: "Michael Behringer (mbehring)" 
To: Laurent Ciavaglia , EXT Brian E Carpenter , "EXT Joel M. Halpern" 
Thread-Topic: [Anima] Re-flooding [was: Does Autonomic network need some intervention beyond intent]
Thread-Index: AQHRhgV/xjAWxQrPfUKm/+t8QaUqkZ9pbe0AgABdX1A=
Date: Fri, 25 Mar 2016 08:00:10 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com> <56F409D0.9030900@joelhalpern.com> <56F41351.8080309@alcatel-lucent.com> <56F4437C.7000803@gmail.com> <56F45A42.4090207@alcatel-lucent.com>
In-Reply-To: <56F45A42.4090207@alcatel-lucent.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: multipart/alternative; boundary="_000_ed40ea6cc124413a860bc745c4964f56XCHRCD006ciscocom_"
MIME-Version: 1.0
Archived-At: 
Cc: Duzongpeng , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Re-flooding [was: Does Autonomic network need some intervention beyond intent]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 08:00:14 -0000

--_000_ed40ea6cc124413a860bc745c4964f56XCHRCD006ciscocom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_ed40ea6cc124413a860bc745c4964f56XCHRCD006ciscocom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_ed40ea6cc124413a860bc745c4964f56XCHRCD006ciscocom_--


From nobody Fri Mar 25 01:31:16 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 15D2D12D6EA for ; Fri, 25 Mar 2016 01:31:14 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.23
X-Spam-Level: 
X-Spam-Status: No, score=-4.23 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fD1w4g5JS1MY for ; Fri, 25 Mar 2016 01:31:09 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1F90512D603 for ; Fri, 25 Mar 2016 01:31:06 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml706-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGO34367; Fri, 25 Mar 2016 08:31:03 +0000 (GMT)
Received: from NKGEML410-HUB.china.huawei.com (10.98.56.41) by lhreml706-cah.china.huawei.com (10.201.5.182) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 25 Mar 2016 08:31:01 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml410-hub.china.huawei.com ([10.98.56.41]) with mapi id 14.03.0235.001; Fri, 25 Mar 2016 16:30:52 +0800
From: Duzongpeng 
To: John Strassner , "Michael Behringer (mbehring)" 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwATdGeIAAjeb4MA==
Date: Fri, 25 Mar 2016 08:30:52 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> 
In-Reply-To: 
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.149.226]
Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFC9BBnkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020203.56F4F748.008E, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 75564e734a57cfeb0e99f48b48eaa534
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 08:31:14 -0000

--_000_BAFEC9523F57BC48A51C20226A5589575FDFC9BBnkgeml514mbxchi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFC9BBnkgeml514mbxchi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFC9BBnkgeml514mbxchi_--


From nobody Fri Mar 25 01:55:43 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B57C812DBF4 for ; Fri, 25 Mar 2016 01:55:42 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id zGyOtvncLhBg for ; Fri, 25 Mar 2016 01:55:41 -0700 (PDT)
Received: from alln-iport-5.cisco.com (alln-iport-5.cisco.com [173.37.142.92]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 158FF12DBEB for ; Fri, 25 Mar 2016 01:55:41 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=2398; q=dns/txt; s=iport; t=1458896141; x=1460105741; h=from:to:subject:date:message-id: content-transfer-encoding:mime-version; bh=LCBBdMctIEIJTsQL8uNhjIDvx1tL3EBvC97n8P6dzkw=; b=NXFoLxnIeLnRyC7ykKv1JvmgEV10md0c8Ii/8aYVYYCFGslZHVkdGmaj Z6JpUB6aQa32bLizMuTCm8bzWrMjFHD1RuRZcocHhMbsMc5gbEehq9E2V KMnudKLAsD3VO3xf16NY5H6CHqrn+Adk2DkdUDt05oqStgMVb+2eXa+ud c=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D0AQAJ/PRW/5tdJa1egzOBVrpUAQ2Bc?= =?us-ascii?q?IdCOBQBAQEBAQEBZBwLhEgnE1EBPkImAQQKEYgfoFOgNwEBCAIBHYYejlYFkmG?= =?us-ascii?q?EfwGNfI8SjwkBHgEBQoNliQclGH4BAQE?=
X-IronPort-AV: E=Sophos;i="5.24,390,1454976000"; d="scan'208";a="251668989"
Received: from rcdn-core-4.cisco.com ([173.37.93.155]) by alln-iport-5.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 25 Mar 2016 08:55:39 +0000
Received: from XCH-RCD-010.cisco.com (xch-rcd-010.cisco.com [173.37.102.20]) by rcdn-core-4.cisco.com (8.14.5/8.14.5) with ESMTP id u2P8tdog007510 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL) for ; Fri, 25 Mar 2016 08:55:40 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-010.cisco.com (173.37.102.20) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 25 Mar 2016 03:55:39 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 25 Mar 2016 03:55:39 -0500
From: "Michael Behringer (mbehring)" 
To: "anima@ietf.org" 
Thread-Topic: Content Distribution
Thread-Index: AdGGcsTKwEw5a+bhS4CjQV2oOW2Ymg==
Date: Fri, 25 Mar 2016 08:55:39 +0000
Message-ID: <90c950d5911040b28da16e33e2ca4d05@XCH-RCD-006.cisco.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Subject: [Anima] Content Distribution
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 08:55:42 -0000

Just writing down how I see Intent distribution as a whole process, for dis=
cussion. And, I want to abstract from distributing "Intent" to distributing=
 "content", where "Intent" is one type only.=20

Starting point: a node is booted; say it has no Intent initially;=20
   $content_type =3D "Intent"
   $my_version =3D 0;
   it consults the AN adjacency table (see section 5 of the reference model=
 draft).
   it has n neighbours in the same domain --> n ACP connections.=20

Run every t sec, or when I just received new content:=20
- For each neighbour in the adjacency table with an ACP connection: =20
   - send $content_type; $my_version
   - receive $content_type; $remote _version
   - if  $my_version =3D $remote _version
           exit;
   - if  $my_version < $remote_ version
           pull $content (unicast from that neighbour)
   - if  $my _version > $remote_version
           push $content (unicast to that neighbour)

All of the above happens in the ANI; ASAs are not involved in this operatio=
n.=20
Probably GRASP does the version comparison; not sure about the file transfe=
r.

The $content_type allows us to use the same mechanism to flood other things=
; this gives us a lot of flexibility to distribute other content types late=
r on; and, if we find out that we *really* need to separate Intent for diff=
erent autonomic functions, we could then say $content_type =3D "Intent for =
function a" for example (of course this is schematic only; we may want sub-=
fields).=20

The Intent is stored as a file. it is that file that gets re-distributed. T=
his file is locally validated (signature) and parsed, and then passed to th=
e ASAs. And a portion of Intent will affect the ANI itself.=20

How far off am I with this thinking? I think this is very simple and robust=
.=20

draft-liu-anima-grasp-distribution-01 has I think the same fundamental thin=
king, but specified a lot of things that are not needed, IMO:=20
- I do not think we need selective distribution (2.2)
- I do not think we need incremental distribution (2.3)
- 3.1 is a bit redundant, if we assume distribution in the ACP. I would rat=
her state
  as above "to all ACP neighbors", then we have the boundary detection impl=
icit.=20
- I think the above algorithm avoids loops automatically.=20
- I don't think we need selective flooding (3.4)

Michael


From nobody Fri Mar 25 01:57:04 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2EC6F12DBF4 for ; Fri, 25 Mar 2016 01:57:03 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.23
X-Spam-Level: 
X-Spam-Status: No, score=-4.23 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yjmwiOt84-tR for ; Fri, 25 Mar 2016 01:56:59 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E8B4C12D100 for ; Fri, 25 Mar 2016 01:56:57 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml702-cah.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CLF66397; Fri, 25 Mar 2016 08:56:55 +0000 (GMT)
Received: from NKGEML401-HUB.china.huawei.com (10.98.56.32) by lhreml702-cah.china.huawei.com (10.201.5.99) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 25 Mar 2016 08:56:53 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml401-hub.china.huawei.com ([10.98.56.32]) with mapi id 14.03.0235.001; Fri, 25 Mar 2016 16:56:42 +0800
From: Duzongpeng 
To: John Strassner , "Michael Behringer (mbehring)" 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwATdGeIAAju2B4A==
Date: Fri, 25 Mar 2016 08:56:42 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com> 
In-Reply-To: 
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.149.226]
Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFCA07nkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020204.56F4FD58.0027, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 459087b6c3337b62f73452de91af5c9c
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 08:57:03 -0000

--_000_BAFEC9523F57BC48A51C20226A5589575FDFCA07nkgeml514mbxchi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFCA07nkgeml514mbxchi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFCA07nkgeml514mbxchi_--


From nobody Fri Mar 25 02:06:06 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CB38412D100 for ; Fri, 25 Mar 2016 02:06:04 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=unavailable autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id FCq_dScdxhbu for ; Fri, 25 Mar 2016 02:06:03 -0700 (PDT)
Received: from rcdn-iport-6.cisco.com (rcdn-iport-6.cisco.com [173.37.86.77]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E7BDA12DBEB for ; Fri, 25 Mar 2016 01:59:32 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1688; q=dns/txt; s=iport; t=1458896372; x=1460105972; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=y2xbHzcBf82YDcHtd9Uh4nLqpMZnVczajfLwpVzczao=; b=Ux5YO0GMttzUL6sirXIQjqTBooUZtiNCQWHQI+Z6T6Y7LnloosFSKyct D+cAvVgN4fRyFMRNVW+5aJ1k/7zwEzGiVXPLw1mbXleMx/qroo4izcYMF X+ll+2N3F+n7xcHeZ/qUuEuE/dTsXDNNdyBWacXlTiChFM7Ervh+VzRKD Y=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D3AQA2/fRW/4wNJK1egzOBUAavBok/g?= =?us-ascii?q?g8BDYFwhg0CHIEXOBQBAQEBAQEBZCeEQQEBAQMBIxFFBQcEAgEIEQQBAQMCIwM?= =?us-ascii?q?CAgIfERQBCAgCBAENBQiICgMKCLAoi1cNhHcBAQEBAQEBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEVfIUihESBSHaBXwEBgx2CVgWXLzEBjA6Bbo8ShzWHVAEeAQFCgjCBNWwBiCE?= =?us-ascii?q?2fgEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,390,1454976000"; d="scan'208";a="86510889"
Received: from alln-core-7.cisco.com ([173.36.13.140]) by rcdn-iport-6.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 25 Mar 2016 08:59:32 +0000
Received: from XCH-RCD-007.cisco.com (xch-rcd-007.cisco.com [173.37.102.17]) by alln-core-7.cisco.com (8.14.5/8.14.5) with ESMTP id u2P8xWSt026943 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 25 Mar 2016 08:59:32 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-007.cisco.com (173.37.102.17) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 25 Mar 2016 03:59:31 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 25 Mar 2016 03:59:31 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , Laurent Ciavaglia , "Liubing (Leo)" , Carsten Bormann 
Thread-Topic: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
Thread-Index: AQHRhLanV1CJbWvMW0GAElzHzEqps59nDdYAgACwpwCAAAXigIAAUmqAgAAkklCAARYoAIAAjZNA
Date: Fri, 25 Mar 2016 08:59:31 +0000
Message-ID: 
References: <0a0b7fb39cb746faa6e846b35fe5c91c@XCH-RCD-006.cisco.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5C9CD@nkgeml514-mbx.china.huawei.com> <56EFD090.1070501@alcatel-lucent.com> <7471201223ba4962a66dd5058190eced@XCH-RCD-006.cisco.com> <56EFE0AE.5040007@alcatel-lucent.com> <56F04B4B.1060703@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5DDDB@nkgeml514-mbx.china.huawei.com> <56F0E9CF.7070602@tzi.org> <56F1969D.1060804@gmail.com> <8AE0F17B87264D4CAC7DE0AA6C406F45C2D5E464@nkgeml514-mbx.china.huawei.com> <56F259F6.5000807@alcatel-lucent.com> <56F2EE26.9030107@gmail.com>  <56F33837.4020001@gmail.com> <498df852d71145ad87eb42f68513fb65@XCH-RCD-006.cisco.com> <56F4403A.1030007@gmail.com>
In-Reply-To: <56F4403A.1030007@gmail.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Cc: "anima@ietf.org" 
Subject: Re: [Anima] How we manage Intent?-//RE: CBOR for Intent?-//RE: Intent yang model - some thoughts
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 09:06:05 -0000
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From nobody Fri Mar 25 02:12:59 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D2EF712DBEE for ; Fri, 25 Mar 2016 02:12:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.722
X-Spam-Level: 
X-Spam-Status: No, score=-2.722 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 56op4WOVF1O5 for ; Fri, 25 Mar 2016 02:12:57 -0700 (PDT)
Received: from mailb2.tigertech.net (mailb2.tigertech.net [208.80.4.154]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1922E12D5DE for ; Fri, 25 Mar 2016 02:12:57 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by mailb2.tigertech.net (Postfix) with ESMTP id D93261C0498; Fri, 25 Mar 2016 02:12:56 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1458897176; bh=J+wB3SAiWunYQtKg1G614qsZZQt+fdzmuE7e065yc50=; h=Subject:To:References:From:Date:In-Reply-To:From; b=ZAKj2fmBq4rCI62Xv0L94QtjJsKZqQN3y+mA64qaNhf9XwOIxC6MhvSUY8SyvAJgD rZZxV9w7lukpSkY18q9SdcQlDkHYGEy4b4OMtIDQ34j2PThMGCN4fC/yHvytyP1BTs /+uoPWC7s7Ha/HdmIdqOU2swEvrsri+3qiy5k2C8=
X-Virus-Scanned: Debian amavisd-new at b2.tigertech.net
Received: from Joels-MacBook-Pro.local (unknown [88.131.67.22]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mailb2.tigertech.net (Postfix) with ESMTPSA id 140991C0477; Fri, 25 Mar 2016 02:12:55 -0700 (PDT)
To: "Michael Behringer (mbehring)" , "anima@ietf.org" 
References: <90c950d5911040b28da16e33e2ca4d05@XCH-RCD-006.cisco.com>
From: "Joel M. Halpern" 
Message-ID: <56F50115.7040607@joelhalpern.com>
Date: Fri, 25 Mar 2016 05:12:53 -0400
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <90c950d5911040b28da16e33e2ca4d05@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: Re: [Anima] Content Distribution
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 09:12:59 -0000

If we want to go with that approach (and it is one reasonable approach), 
I would presume it would be worth considering a minor optimization:

store the version you received from the neighbor,
any time you receive something new, see if any neighbor is interested in 
it,
if they are interested, and it is newer than what they have, push it to 
them.

This improves convergence.  It has the cost that an entity may receive 
the same intent from several neighbors.  At which point I think one 
could make the time based check a much lower rate backup mechanism.

Yours,
Joel

On 3/25/16 4:55 AM, Michael Behringer (mbehring) wrote:
> Just writing down how I see Intent distribution as a whole process, for discussion. And, I want to abstract from distributing "Intent" to distributing "content", where "Intent" is one type only.
>
> Starting point: a node is booted; say it has no Intent initially;
>     $content_type = "Intent"
>     $my_version = 0;
>     it consults the AN adjacency table (see section 5 of the reference model draft).
>     it has n neighbours in the same domain --> n ACP connections.
>
> Run every t sec, or when I just received new content:
> - For each neighbour in the adjacency table with an ACP connection:
>     - send $content_type; $my_version
>     - receive $content_type; $remote _version
>     - if  $my_version = $remote _version
>             exit;
>     - if  $my_version < $remote_ version
>             pull $content (unicast from that neighbour)
>     - if  $my _version > $remote_version
>             push $content (unicast to that neighbour)
>
> All of the above happens in the ANI; ASAs are not involved in this operation.
> Probably GRASP does the version comparison; not sure about the file transfer.
>
> The $content_type allows us to use the same mechanism to flood other things; this gives us a lot of flexibility to distribute other content types later on; and, if we find out that we *really* need to separate Intent for different autonomic functions, we could then say $content_type = "Intent for function a" for example (of course this is schematic only; we may want sub-fields).
>
> The Intent is stored as a file. it is that file that gets re-distributed. This file is locally validated (signature) and parsed, and then passed to the ASAs. And a portion of Intent will affect the ANI itself.
>
> How far off am I with this thinking? I think this is very simple and robust.
>
> draft-liu-anima-grasp-distribution-01 has I think the same fundamental thinking, but specified a lot of things that are not needed, IMO:
> - I do not think we need selective distribution (2.2)
> - I do not think we need incremental distribution (2.3)
> - 3.1 is a bit redundant, if we assume distribution in the ACP. I would rather state
>    as above "to all ACP neighbors", then we have the boundary detection implicit.
> - I think the above algorithm avoids loops automatically.
> - I don't think we need selective flooding (3.4)
>
> Michael
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>


From nobody Fri Mar 25 02:23:00 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8A0C812DBED for ; Fri, 25 Mar 2016 02:22:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3mx8R1DR56KT for ; Fri, 25 Mar 2016 02:22:56 -0700 (PDT)
Received: from alln-iport-5.cisco.com (alln-iport-5.cisco.com [173.37.142.92]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 831C912D695 for ; Fri, 25 Mar 2016 02:22:56 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=4142; q=dns/txt; s=iport; t=1458897776; x=1460107376; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=K1iThabl8fy9ihsGwuM0/yQ8SQyp3eQ8P+iqJPQX/sU=; b=NstwvEnmhh1J24IArVka7BfjG8YtOkj3s2n6rXn7lO+BO3GNlZQvrF7e xXmRTLtDqGMiM3VRL70zAndgBvWAsbkkcJQLgz+Tb7vl0kDyt1Pn8KvU3 LIgBM1k63KiEmbnmrmyuKXbnKnYVwv55gGSESvCv886xJtgeap/Pu+vEQ w=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D3AQAPA/VW/49dJa1egzNTfQa6VAENg?= =?us-ascii?q?XAXCoVsAoE0OBQBAQEBAQEBZCeEQQEBAQMBAQEBJBM0EAcEAgEIEQQBAQEeCQc?= =?us-ascii?q?nCxQJCAEBBAEJCQiIFwgOwG0BAQEBAQEBAQEBAQEBAQEBAQEBAQERBIYehESEA?= =?us-ascii?q?4YPBZJhhH8BjXyPEo8JAR4BAUKDZWyIGyUYfgEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,390,1454976000"; d="scan'208";a="251675155"
Received: from rcdn-core-7.cisco.com ([173.37.93.143]) by alln-iport-5.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 25 Mar 2016 09:22:55 +0000
Received: from XCH-RCD-006.cisco.com (xch-rcd-006.cisco.com [173.37.102.16]) by rcdn-core-7.cisco.com (8.14.5/8.14.5) with ESMTP id u2P9MtX3013337 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 25 Mar 2016 09:22:55 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-006.cisco.com (173.37.102.16) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 25 Mar 2016 04:22:54 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 25 Mar 2016 04:22:54 -0500
From: "Michael Behringer (mbehring)" 
To: "Joel M. Halpern" , "anima@ietf.org" 
Thread-Topic: [Anima] Content Distribution
Thread-Index: AdGGcsTKwEw5a+bhS4CjQV2oOW2YmgALabmAAApGS8A=
Date: Fri, 25 Mar 2016 09:22:54 +0000
Message-ID: <3279f8be902d41649e9198a0454b91ba@XCH-RCD-006.cisco.com>
References: <90c950d5911040b28da16e33e2ca4d05@XCH-RCD-006.cisco.com> <56F50115.7040607@joelhalpern.com>
In-Reply-To: <56F50115.7040607@joelhalpern.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Subject: Re: [Anima] Content Distribution
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 09:22:58 -0000

> -----Original Message-----
> From: Joel M. Halpern [mailto:jmh@joelhalpern.com]
> Sent: 25 March 2016 10:13
> To: Michael Behringer (mbehring) ; anima@ietf.org
> Subject: Re: [Anima] Content Distribution
>=20
> If we want to go with that approach (and it is one reasonable approach), =
I
> would presume it would be worth considering a minor optimization:
>=20
> store the version you received from the neighbor, any time you receive
> something new, see if any neighbor is interested in it, if they are inter=
ested,
> and it is newer than what they have, push it to them.

Thanks Joel. I think the proposed algorithm actually does that. There is th=
e sub-clause "or when I just received new Intent", which does mostly what y=
ou describe, except in this algorithm you'd still exchange the version numb=
er before pushing the file; that could be removed; yes, optimisations are p=
ossible of course.=20

My main point is: This algorithm would take care of flooding through the do=
main, loop avoidance, etc, while being very simple.=20
=20
Thanks for the fast feedback! :-)=20
Michael

> This improves convergence.  It has the cost that an entity may receive th=
e
> same intent from several neighbors.  At which point I think one could mak=
e
> the time based check a much lower rate backup mechanism.
>=20
> Yours,
> Joel
>=20
> On 3/25/16 4:55 AM, Michael Behringer (mbehring) wrote:
> > Just writing down how I see Intent distribution as a whole process, for
> discussion. And, I want to abstract from distributing "Intent" to distrib=
uting
> "content", where "Intent" is one type only.
> >
> > Starting point: a node is booted; say it has no Intent initially;
> >     $content_type =3D "Intent"
> >     $my_version =3D 0;
> >     it consults the AN adjacency table (see section 5 of the reference =
model
> draft).
> >     it has n neighbours in the same domain --> n ACP connections.
> >
> > Run every t sec, or when I just received new content:
> > - For each neighbour in the adjacency table with an ACP connection:
> >     - send $content_type; $my_version
> >     - receive $content_type; $remote _version
> >     - if  $my_version =3D $remote _version
> >             exit;
> >     - if  $my_version < $remote_ version
> >             pull $content (unicast from that neighbour)
> >     - if  $my _version > $remote_version
> >             push $content (unicast to that neighbour)
> >
> > All of the above happens in the ANI; ASAs are not involved in this
> operation.
> > Probably GRASP does the version comparison; not sure about the file
> transfer.
> >
> > The $content_type allows us to use the same mechanism to flood other
> things; this gives us a lot of flexibility to distribute other content ty=
pes later
> on; and, if we find out that we *really* need to separate Intent for diff=
erent
> autonomic functions, we could then say $content_type =3D "Intent for func=
tion
> a" for example (of course this is schematic only; we may want sub-fields)=
.
> >
> > The Intent is stored as a file. it is that file that gets re-distribute=
d. This file is
> locally validated (signature) and parsed, and then passed to the ASAs. An=
d a
> portion of Intent will affect the ANI itself.
> >
> > How far off am I with this thinking? I think this is very simple and ro=
bust.
> >
> > draft-liu-anima-grasp-distribution-01 has I think the same fundamental
> thinking, but specified a lot of things that are not needed, IMO:
> > - I do not think we need selective distribution (2.2)
> > - I do not think we need incremental distribution (2.3)
> > - 3.1 is a bit redundant, if we assume distribution in the ACP. I would=
 rather
> state
> >    as above "to all ACP neighbors", then we have the boundary detection
> implicit.
> > - I think the above algorithm avoids loops automatically.
> > - I don't think we need selective flooding (3.4)
> >
> > Michael
> >
> > _______________________________________________
> > Anima mailing list
> > Anima@ietf.org
> > https://www.ietf.org/mailman/listinfo/anima
> >


From nobody Fri Mar 25 02:31:30 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A8B3B12D130 for ; Fri, 25 Mar 2016 02:31:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.23
X-Spam-Level: 
X-Spam-Status: No, score=-4.23 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id H52TreNVQG6Q for ; Fri, 25 Mar 2016 02:31:24 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 055D912D5E8 for ; Fri, 25 Mar 2016 02:31:19 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml701-cah.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CLF69489; Fri, 25 Mar 2016 09:31:16 +0000 (GMT)
Received: from nkgeml405-hub.china.huawei.com (10.98.56.36) by lhreml701-cah.china.huawei.com (10.201.5.93) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 25 Mar 2016 09:31:14 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml405-hub.china.huawei.com ([10.98.56.36]) with mapi id 14.03.0235.001; Fri, 25 Mar 2016 17:31:02 +0800
From: Duzongpeng 
To: John Strassner , "Michael Behringer (mbehring)" 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwATdGeIAAHjH6AAAg/AuAAFA8wmA=
Date: Fri, 25 Mar 2016 09:31:01 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> 
In-Reply-To: 
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.149.226]
Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFCA51nkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020203.56F50566.004D, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 459087b6c3337b62f73452de91af5c9c
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 09:31:29 -0000

--_000_BAFEC9523F57BC48A51C20226A5589575FDFCA51nkgeml514mbxchi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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=

--_000_BAFEC9523F57BC48A51C20226A5589575FDFCA51nkgeml514mbxchi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64

PGh0bWwgeG1sbnM6dj0idXJuOnNjaGVtYXMtbWljcm9zb2Z0LWNvbTp2bWwiIHhtbG5zOm89InVy
bjpzY2hlbWFzLW1pY3Jvc29mdC1jb206b2ZmaWNlOm9mZmljZSIgeG1sbnM6dz0idXJuOnNjaGVt
YXMtbWljcm9zb2Z0LWNvbTpvZmZpY2U6d29yZCIgeG1sbnM6bT0iaHR0cDovL3NjaGVtYXMubWlj
cm9zb2Z0LmNvbS9vZmZpY2UvMjAwNC8xMi9vbW1sIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcv
VFIvUkVDLWh0bWw0MCI+DQo8aGVhZD4NCjxtZXRhIGh0dHAtZXF1aXY9IkNvbnRlbnQtVHlwZSIg
Y29udGVudD0idGV4dC9odG1sOyBjaGFyc2V0PXV0Zi04Ij4NCjxtZXRhIG5hbWU9IkdlbmVyYXRv
ciIgY29udGVudD0iTWljcm9zb2Z0IFdvcmQgMTIgKGZpbHRlcmVkIG1lZGl1bSkiPg0KPHN0eWxl
PjwhLS0NCi8qIEZvbnQgRGVmaW5pdGlvbnMgKi8NCkBmb250LWZhY2UNCgl7Zm9udC1mYW1pbHk6
V2luZ2RpbmdzOw0KCXBhbm9zZS0xOjUgMCAwIDAgMCAwIDAgMCAwIDA7fQ0KQGZvbnQtZmFjZQ0K
CXtmb250LWZhbWlseTrlrovkvZM7DQoJcGFub3NlLTE6MiAxIDYgMCAzIDEgMSAxIDEgMTt9DQpA
Zm9udC1mYWNlDQoJe2ZvbnQtZmFtaWx5OiJDYW1icmlhIE1hdGgiOw0KCXBhbm9zZS0xOjIgNCA1
IDMgNSA0IDYgMyAyIDQ7fQ0KQGZvbnQtZmFjZQ0KCXtmb250LWZhbWlseTpDYWxpYnJpOw0KCXBh
bm9zZS0xOjIgMTUgNSAyIDIgMiA0IDMgMiA0O30NCkBmb250LWZhY2UNCgl7Zm9udC1mYW1pbHk6
IlxA5a6L5L2TIjsNCglwYW5vc2UtMToyIDEgNiAwIDMgMSAxIDEgMSAxO30NCkBmb250LWZhY2UN
Cgl7Zm9udC1mYW1pbHk6VGFob21hOw0KCXBhbm9zZS0xOjIgMTEgNiA0IDMgNSA0IDQgMiA0O30N
Ci8qIFN0eWxlIERlZmluaXRpb25zICovDQpwLk1zb05vcm1hbCwgbGkuTXNvTm9ybWFsLCBkaXYu
TXNvTm9ybWFsDQoJe21hcmdpbjowY207DQoJbWFyZ2luLWJvdHRvbTouMDAwMXB0Ow0KCWZvbnQt
c2l6ZToxMi4wcHQ7DQoJZm9udC1mYW1pbHk65a6L5L2TO30NCmE6bGluaywgc3Bhbi5Nc29IeXBl
cmxpbmsNCgl7bXNvLXN0eWxlLXByaW9yaXR5Ojk5Ow0KCWNvbG9yOmJsdWU7DQoJdGV4dC1kZWNv
cmF0aW9uOnVuZGVybGluZTt9DQphOnZpc2l0ZWQsIHNwYW4uTXNvSHlwZXJsaW5rRm9sbG93ZWQN
Cgl7bXNvLXN0eWxlLXByaW9yaXR5Ojk5Ow0KCWNvbG9yOnB1cnBsZTsNCgl0ZXh0LWRlY29yYXRp
b246dW5kZXJsaW5lO30NCnANCgl7bXNvLXN0eWxlLXByaW9yaXR5Ojk5Ow0KCW1zby1tYXJnaW4t
dG9wLWFsdDphdXRvOw0KCW1hcmdpbi1yaWdodDowY207DQoJbXNvLW1hcmdpbi1ib3R0b20tYWx0
OmF1dG87DQoJbWFyZ2luLWxlZnQ6MGNtOw0KCWZvbnQtc2l6ZToxMi4wcHQ7DQoJZm9udC1mYW1p
bHk65a6L5L2TO30NCnAuTXNvQWNldGF0ZSwgbGkuTXNvQWNldGF0ZSwgZGl2Lk1zb0FjZXRhdGUN
Cgl7bXNvLXN0eWxlLXByaW9yaXR5Ojk5Ow0KCW1zby1zdHlsZS1saW5rOiLmibnms6jmoYbmlofm
nKwgQ2hhciI7DQoJbWFyZ2luOjBjbTsNCgltYXJnaW4tYm90dG9tOi4wMDAxcHQ7DQoJZm9udC1z
aXplOjkuMHB0Ow0KCWZvbnQtZmFtaWx5OuWui+S9kzt9DQpzcGFuLkVtYWlsU3R5bGUxOA0KCXtt
c28tc3R5bGUtdHlwZTpwZXJzb25hbC1yZXBseTsNCglmb250LWZhbWlseToiQ2FsaWJyaSIsInNh
bnMtc2VyaWYiOw0KCWNvbG9yOiMxRjQ5N0Q7fQ0Kc3Bhbi5DaGFyDQoJe21zby1zdHlsZS1uYW1l
OiLmibnms6jmoYbmlofmnKwgQ2hhciI7DQoJbXNvLXN0eWxlLXByaW9yaXR5Ojk5Ow0KCW1zby1z
dHlsZS1saW5rOuaJueazqOahhuaWh+acrDsNCglmb250LWZhbWlseTrlrovkvZM7fQ0KLk1zb0No
cERlZmF1bHQNCgl7bXNvLXN0eWxlLXR5cGU6ZXhwb3J0LW9ubHk7fQ0KLk1zb1BhcERlZmF1bHQN
Cgl7bXNvLXN0eWxlLXR5cGU6ZXhwb3J0LW9ubHk7DQoJdGV4dC1hbGlnbjpqdXN0aWZ5Ow0KCXRl
eHQtanVzdGlmeTppbnRlci1pZGVvZ3JhcGg7fQ0KQHBhZ2UgV29yZFNlY3Rpb24xDQoJe3NpemU6
NjEyLjBwdCA3OTIuMHB0Ow0KCW1hcmdpbjo3Mi4wcHQgOTAuMHB0IDcyLjBwdCA5MC4wcHQ7fQ0K
ZGl2LldvcmRTZWN0aW9uMQ0KCXtwYWdlOldvcmRTZWN0aW9uMTt9DQotLT48L3N0eWxlPjwhLS1b
aWYgZ3RlIG1zbyA5XT48eG1sPg0KPG86c2hhcGVkZWZhdWx0cyB2OmV4dD0iZWRpdCIgc3BpZG1h
eD0iMTAyNiIgLz4NCjwveG1sPjwhW2VuZGlmXS0tPjwhLS1baWYgZ3RlIG1zbyA5XT48eG1sPg0K
PG86c2hhcGVsYXlvdXQgdjpleHQ9ImVkaXQiPg0KPG86aWRtYXAgdjpleHQ9ImVkaXQiIGRhdGE9
IjEiIC8+DQo8L286c2hhcGVsYXlvdXQ+PC94bWw+PCFbZW5kaWZdLS0+DQo8L2hlYWQ+DQo8Ym9k
eSBsYW5nPSJaSC1DTiIgbGluaz0iYmx1ZSIgdmxpbms9InB1cnBsZSI+DQo8ZGl2IGNsYXNzPSJX
b3JkU2VjdGlvbjEiPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0
eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1
b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5IaSBKb2huPG86cD48L286cD48L3Nw
YW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJm
b250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fu
cy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+
DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6
ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlm
JnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPlNvbWUgcGVyc29uYWwgdW5kZXJzdGFuZGluZyBoZXJlLiBJ
ZiBhbnkgbWlzdW5kZXJzdGFuZGluZywgcGxlYXNlIGNvcnJlY3QgbWUuIFRoYW5rcy48bzpwPjwv
bzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5nPSJFTi1VUyIg
c3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90Oywm
cXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPjxvOnA+Jm5ic3A7PC9vOnA+PC9z
cGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJ0ZXh0LWFsaWduOmp1c3RpZnk7
dGV4dC1qdXN0aWZ5OmludGVyLWlkZW9ncmFwaCI+PHNwYW4gbGFuZz0iRU4tVVMiPiZsdDtqY3My
Jmd0Ozxicj4NCkkgd291bGQgc3VnZ2VzdCBhIHNsaWdodCBtb2RpZmljYXRpb24sIHRvIGVuYWJs
ZSB1c2VycyBvZjxicj4NCmludGVudCAoYW5kIGV2ZW4gaW50ZW50IHdyaXRlcnMpIHRvIE5PVCBo
YXZlIHRvIGhhdmU8YnI+DQpkZXRhaWxlZCBpbXBsZW1lbnRhdGlvbiBpbmZvcm1hdGlvbi48YnI+
DQo8YnI+DQpXaGF0IGlmIGFsbCBvZiB0aGUgQXV0b25vbWljIEZ1bmN0aW9ucywgYWZ0ZXIgdGhl
eSBqb2luZWQ8YnI+DQp0aGUgbmV0d29yaywgYWR2ZXJ0aXNlZCBib3RoIHRoZWlyIGNhcGFiaWxp
dGllcyBhcyB3ZWxsIGFzPGJyPg0KdGhlaXIgbmVlZHM/IFdlIGNhbiBwdXQgYXNpZGUgdGhlIGxh
dHRlciAobmVlZHMpIGFzIHRoaXM8YnI+DQppcyBtb3JlIGNvbXBsZXgsIGJ1dCBpZiB0aGUgY2Fw
YWJpbGl0aWVzIGFyZSBhZHZlcnRpc2VkLDxicj4NCnRoZW4gdGhlIGF1dG9ub21pYyBtYW5hZ2Vt
ZW50IHN5c3RlbSBjYW4gYXNzZW1ibGU8YnI+DQp0aG9zZSBuZWVkcyBpbnRvIGEgY2F0YWxvZy4g
VGhlbiwgdGhlIGludGVudCB3cml0ZXIgY2FuPGJyPg0KbG9vayBhdCB0aGUgY2F0YWxvZyBhbmQg
d3JpdGUgYSBkZWNsYXJhdGl2ZSByZXF1ZXN0IHRvPGJyPg0KdXNlIHRoZSBmdW5jdGlvbmFsaXR5
IG9mIHRoZSBjYXRhbG9nLjxicj4NCjxicj4NCkkgd2lsbCBsb29rIGF0IHlvdXIgWUFORyBtb2Rl
bCBpbiBhIGJpdCAoc29ycnksIEknbSB3b2VmdWxseTxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxw
IGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJ0ZXh0LWFsaWduOmp1c3RpZnk7dGV4dC1qdXN0aWZ5
OmludGVyLWlkZW9ncmFwaCI+PHNwYW4gbGFuZz0iRU4tVVMiPmJlaGluZC4uLikuPGJyPg0KJmx0
OyAvamNzMiZndDs8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48
c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1
b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPjxv
OnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIGxh
bmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxp
YnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+SSBhZ3JlZSB0
aGF0IGF1dG9ub21pYyBub2RlIHNob3VsZCBoYXZlIHRoYXQgc2VsZi1hZHZlcnRpc2VtZW50IGFi
aWxpdHksIGFuZCBhdXRvbm9taWMgbWFuYWdlbWVudCBzeXN0ZW0gc2hvdWxkIGhhdmUgdGhpcyBj
YXRhbG9nLjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFu
IGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtD
YWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+PG86cD4m
bmJzcDs8L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gbGFuZz0i
RU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkm
cXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5CdXQgSSBkbyBub3Qg
c3VwcG9ydCB0aGF0IHRoZSBpbnRlbnQgd3JpdGVyIHNob3VsZCBsb29rIGV2ZXJ5IGRldGFpbCBp
biB0aGUgY2F0YWxvZy48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFs
Ij48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6
JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0Qi
PjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFu
IGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtD
YWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+VGhlIGlu
dGVudCB3cml0ZXIgbWF5IG1haW5seSBjYXJlIGFib3V0IHdoYXQgdGhleSB3YW50LCBidXQgaG93
IHRvIGRvIGl0IGNhbiBiZSBsZWZ0IHRvIHRoZSBhdXRvbm9taWMgbmV0d29yayB0byBkZWFsIHdp
dGguDQo8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBs
YW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2Fs
aWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPlBlcmhhcHMg
c2V2ZXJhbCBzb2x1dGlvbnMgZXhpc3QsIGFuZCBhdXRvbm9taWMgbWFuYWdlbWVudCBzeXN0ZW0g
Y2hvb3NlIG9uZSBhY2NvcmRpbmcgdG8gdGhlIGFiaWxpdGllcyBjb2xsZWN0ZWQuPG86cD48L286
cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0
eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1
b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bh
bj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZv
bnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5z
LXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPkluIHRoZSBpZGVhbCwgdGhlIG5ldHdvcmsgb3Bl
cmF0b3IgbWF5IGNvbW11bmljYXRlIHdpdGggdGhlIGF1dG9ub21pYyBuZXR3b3JrIHVzaW5nIG5l
dHdvcmstbGV2ZWwgbGV2ZWwgaW50ZW50IGFuZCBuZXR3b3JrLWxldmVsIHJlcG9ydC48bzpwPjwv
bzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5nPSJFTi1VUyIg
c3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90Oywm
cXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPkkgbWVhbiBpbiBjdXJyZW50IHN0
YXR1cywgc29tZSBvZiB0aGUgQXV0b25vbWljIEZ1bmN0aW9ucyBzaG91bGQgYmUgZXhwb3NlZCB0
byB0aGUgbmV0d29yayBvcGVyYXRvciwgYW5kIHNvbWUgb2YgdGhlIEF1dG9ub21pYyBGdW5jdGlv
bnMgY2FuIGJlDQogaGlkZGVuIHRvIHRoZSBuZXR3b3JrIG9wZXJhdG9yPG86cD48L286cD48L3Nw
YW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJm
b250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fu
cy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5QZXJoYXBzIGluIGZ1dHVyZSwgd2hlbiB0aGUg
bmV0d29yayBub2RlcyBiZWNvbWUgbW9yZSBpbnRlbGxpZ2VudCwgdGhlIGhpZGRlbiBwYXJ0IHdp
bGwgaW5jcmVhc2UuPG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+
PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZx
dW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj48
bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBs
YW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2Fs
aWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPjxvOnA+Jm5i
c3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIGxhbmc9IkVO
LVVTIiBzdHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1
b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+QmVzdCByZWdhcmRzPG86
cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gbGFuZz0iRU4t
VVMiIHN0eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVv
dDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5ab25ncGVuZyBEdTxvOnA+
PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIGxhbmc9IkVOLVVT
IiBzdHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7
LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+PG86cD4mbmJzcDs8L286cD48
L3NwYW4+PC9wPg0KPGRpdiBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLXRvcDpzb2xpZCAjQjVD
NERGIDEuMHB0O3BhZGRpbmc6My4wcHQgMGNtIDBjbSAwY20iPg0KPHAgY2xhc3M9Ik1zb05vcm1h
bCI+PGI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAuMHB0O2ZvbnQtZmFt
aWx5OiZxdW90O1RhaG9tYSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij5Gcm9tOjwvc3Bh
bj48L2I+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAuMHB0O2ZvbnQtZmFt
aWx5OiZxdW90O1RhaG9tYSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij4gQW5pbWEgW21h
aWx0bzphbmltYS1ib3VuY2VzQGlldGYub3JnXQ0KPGI+T24gQmVoYWxmIE9mIDwvYj5Kb2huIFN0
cmFzc25lcjxicj4NCjxiPlNlbnQ6PC9iPiBUaHVyc2RheSwgTWFyY2ggMjQsIDIwMTYgMTA6NDEg
QU08YnI+DQo8Yj5Ubzo8L2I+IE1pY2hhZWwgQmVocmluZ2VyIChtYmVocmluZyk7IEpvaG4gU3Ry
YXNzbmVyPGJyPg0KPGI+Q2M6PC9iPiBEdXpvbmdwZW5nOyBMYXVyZW50IENpYXZhZ2xpYTsgU2hl
bmcgSmlhbmc7IGFuaW1hQGlldGYub3JnOyBKw6lmZXJzb24gQ2FtcG9zIE5vYnJlPGJyPg0KPGI+
U3ViamVjdDo8L2I+IFJlOiBbQW5pbWFdIERvZXMgQXV0b25vbWljIG5ldHdvcmsgbmVlZCBzb21l
IGludGVydmVudGlvbiBiZXlvbmQgaW50ZW50PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+
DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5nPSJFTi1VUyI+PG86cD4mbmJzcDs8L286
cD48L3NwYW4+PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFy
Z2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0i
RU4tVVMiPkhpIE1pY2hhZWwsPG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs
dDphdXRvIj48c3BhbiBsYW5nPSJFTi1VUyI+Jm5ic3A7PG86cD48L286cD48L3NwYW4+PC9wPg0K
PHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1t
YXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1VUyI+dGhhbmtzIGZvciB5b3Vy
IHJlcGx5LiBJIGRvIHRoaW5rIHRoYXQgd2UgbGFyZ2VseSBhZ3JlZSwgYW5kIHdvdWxkIGVuam95
IGNoYXR0aW5nIGR1cmluZyBvciAoaG9wZWZ1bGx5IGJlZm9yZSkgQkEuIFBsZWFzZSBzZWUgJmx0
O2pjczImZ3Q7Li4mbHQ7L2pjczImZ3Q7LiBBbHNvIHRyeWluZyB0bw0KIHVuY2x1dHRlciwgc28g
SSBhbSBlbGlkaW5nIHBvaW50cyB0aGF0IHdlIGFncmVlIG9uLjxvOnA+PC9vOnA+PC9zcGFuPjwv
cD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bztt
c28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tVVMiPiZuYnNwOzxvOnA+
PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2lu
LXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4t
VVMiPiZuYnNwOzxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0
eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+
PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJjb2xvcjojMUY0OTdEIj5XaGlsZSAqQ0lFcyBkb27i
gJl0IGxpa2UgbWFnaWMsIHRoZXkgZG8gdW5kZXJzdGFuZCB0aGUgY29uY2VwdCBvZiBhIGRlZmF1
bHQgYmVoYXZpb3VyIHdoaWNoIHRoZXkgY2FuIGluZmx1ZW5jZSBpZiBuZWVkZWQuIEFueSByZWFz
b25hYmxlIG5ldHdvcmsNCiBtYW5hZ2VtZW50IGFwcHJvYWNoIHRyaWVzIHRvIGRlZmluZSDigJxk
ZWZhdWx04oCdIGJlaGF2aW91ciBhbmQg4oCcZXhjZXB0aW9uc+KAnS4gVGhlIEFOIHN0b3J5IGZp
bGxzIHRoZSDigJxkZWZhdWx04oCdIHBhcnQgdmVyeSBuaWNlbHksIGFuZCAqQ0lFcyBnZXQgdGhh
dC48L3NwYW4+PHNwYW4gbGFuZz0iRU4tVVMiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNs
YXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2lu
LWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tVVMiPiZsdDtqY3MyJmd0OzxvOnA+PC9v
OnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRv
cC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tVVMi
PkFncmVlZCE8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHls
ZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxz
cGFuIGxhbmc9IkVOLVVTIj4mbHQ7L2pjczImZ3Q7PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAg
Y2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJn
aW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1VUyI+Jm5ic3A7PG86cD48L286cD48
L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFs
dDphdXRvO21hcmdpbi1ib3R0b206MTIuMHB0Ij48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImNv
bG9yOiMxRjQ5N0QiPlRoZSDigJxwYXJhbWV0ZXLigJ0gdmVyc3VzIOKAnHBvbGljeeKAnSBkaXNj
dXNzaW9uOiBNeSB2aWV3IGlzIHRoYXQgd2Ugc2hvdWxkbuKAmXQgdHJ5IHRvIGRlZmluZSBleGFj
dGx5ICo8Yj53aGF0PC9iPiogaW50ZW50IGFsbG93cyBhbmQgZG9lc27igJl0IGFsbG93LA0KIHNp
bmNlIHdl4oCZdmUgaGVhcmQgZnJvbSBtYW55IHNpZGVzIHRoYXQgdGhlIGJvdW5kYXJ5IGlzIG5v
dCBjbGVhciwgYW5kIGl0IG1heSBiZSBpbXBvc3NpYmxlIHRvIGhhdmUgYSByZWFsbHkgY2xlYXIs
IHVzYWJsZSBkZWZpbml0aW9uLiBTbyBteSBwb2ludCB3YXM6IExldOKAmXMgZm9jdXMgb24gdGhl
IGRpc3RyaWJ1dGlvbiBtb2RlbDogSW50ZW50IGlzIGZsb29kZWQgdG8gYWxsIG5vZGVzLiBMZXTi
gJlzIE5PVCBkbyBzZWxlY3RpdmUgZmxvb2RpbmcsIHVuaWNhc3QNCiBJbnRlbnQgdG8gc3BlY2lm
aWMgbm9kZXMsIGV0YywgYmVjYXVzZSB0aGF0IHdvdWxkIGJsdXIgdGhlIGJvdW5kYXJ5IGV2ZW4g
d29yc2UuPGJyPg0KPC9zcGFuPjxzcGFuIGxhbmc9IkVOLVVTIj4mbHQ7amNzMiZndDs8YnI+DQpJ
IGxpa2UgZmxvb2RpbmcgaW50ZW50IHRvIGFsbCBub2Rlcy4gSW4gRk9DQUxFICh2MS0zLCBhbmQg
cmVtZW1iZXIsIHYxIHdhcyAyMDA2LCBhbG1vc3QgMTAgeWVhcnMgYWdvKSwgd2UgdXNlZCBhIG1l
c3NhZ2UgYnVzIGFuZCBwdWItc3ViLiBJbiB2Miwgd2UgZmxvb2RlZDsgaW4gdjMsIHdlIHdlbnQg
YmFjayB0byBwdWItc3ViIGJlY2F1c2Ugd2Ugd2FudGVkIHRvIGRvIHNvbWUgb3B0aW1pemF0aW9u
IGFuZCBhdWRpdGluZy4gQnV0IEkgd291bGQgYmUNCiBoYXBweSB0byB0cnkgc29tZSB0eXBlIG9m
IGNvbnRyb2xsZWQgZmxvb2RpbmcgYWdhaW4uPGJyPg0KJmx0Oy9qY3MyJmd0OzxvOnA+PC9vOnA+
PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1h
bHQ6YXV0bzttYXJnaW4tYm90dG9tOjEyLjBwdCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJj
b2xvcjojMUY0OTdEIj5NeSBzdWdnZXN0aW9uOiBJbnRlbnQgaXMgZmxvb2RlZCB0byBhbGwgbm9k
ZXMgaW4gYSBkb21haW4uIFRoZSBBU0Egb3duZXJzIGRlZmluZSB0aGUgY29udGVudCBvZiB3aGF0
IHRoZXkgd2FudCB0byBwdXNoIGluIHRoZWlyIGJpdCBvZiB0aGUgSW50ZW50Lg0KIElmIHdlIGlu
c2lzdCBvZiBmdWxsIGZsb29kaW5nLCB3ZSBlbmNvdXJhZ2UgdGhlIHJpZ2h0IGJlaGF2aW91ci4g
QW5kIGlmIHNvbWVvbmUgcmVhbGx5IHB1dHMgaW50byBoaXMgSW50ZW50IOKAnG5vZGUgQTogZG8g
dGhpczsgbm9kZSBCOiBkbyB0aGUgb3RoZXI7IG5vZGUgQzogZG8gdGhlIHRoaXJk4oCdIGl0IGlz
IGF3a3dhcmQgZW5vdWdoIGZvciBwZW9wbGUgdG8gdGFrZSBhIHN0ZXAgYmFjayBhbmQgdGhpbmsg
d2hldGhlciB0aGV54oCZcmUgZG9pbmcgdGhlDQogcmlnaHQgdGhpbmcgPC9zcGFuPjxzcGFuIGxh
bmc9IkVOLVVTIiBzdHlsZT0iZm9udC1mYW1pbHk6V2luZ2RpbmdzO2NvbG9yOiMxRjQ5N0QiPko8
L3NwYW4+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJjb2xvcjojMUY0OTdEIj4NCjxicj4NCjwv
c3Bhbj48c3BhbiBsYW5nPSJFTi1VUyI+Jmx0O2pjczImZ3Q7PGJyPg0KSSBhZ3JlZSBpbiBwcmlu
Y2lwbGUgd2l0aCB3aGF0IHlvdSBzYWlkLiBQbHVzLCBzZWUgYWJvdmUuPGJyPg0KJmx0OyAvamNz
MiZndDs8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0i
bXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bWFyZ2luLWJvdHRvbToxMi4wcHQiPjxzcGFuIGxhbmc9
IkVOLVVTIiBzdHlsZT0iY29sb3I6IzFGNDk3RCI+Jmx0O2pjcyZndDs8YnI+DQpEbyB5b3UgbWVh
biB0aGF0IHRoZSBpbnRlbnQgd3JpdGVyIHdpbGwgKiprbm93IHRoZSBkaWZmZXJlbmNlKio8YnI+
DQpiZXR3ZWVuIHRoZSBmYWJyaWMgYW5kIGVhY2ggYXV0b25vbWljIGZ1bmN0aW9uPyBJIGhvcGUg
bm90LDxicj4NCmJ1dCBwZXJoYXBzIEknbSBtaXN1bmRlcnN0YW5kaW5nIHlvdS4uLjxicj4NCiZs
dDsgL2pjcyZndDs8YnI+DQpOb3Qgc3VyZSBJIHVuZGVyc3RhbmQgeW91ciBjb25jZXJuLiBZZXMs
IGluIG15IG1vZGVsIHRoZXJlIGlzIGEgc2VjdGlvbiBmb3IgdGhlIGZhYnJpYywgYW5kIHNlY3Rp
b25zIGZvciB0aGUgdmFyaW91cyBmdW5jdGlvbnMuIFNlZW1zIGEgbmF0dXJhbCBzdHJ1Y3R1cmU/
IT8NCjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1VUyI+Jmx0O2pjczImZ3Q7PGJyPg0KVGhlIGludGVu
dCAqKndyaXRlcioqIHdvdWxkIGhhdmUgdG8ga25vdyBpbXBsZW1lbnRhdGlvbjxicj4NCmRldGFp
bHMgaW4gb3JkZXIgdG8gcGxhY2UgZXZlcnl0aGluZyBpbiBhIHNpbmdsZSBmaWxlLiBJIHdvdWxk
PGJyPg0KcmF0aGVyIG5vdCByZXF1aXJlIHRoaXMsIGFuZCBpbnN0ZWFkLCBoYXZlIHRoZSBpbnRl
bnQgZW5naW5lPGJyPg0KKGUuZy4sIGNvbXBpbGVyIGFuZCBhc3NvY2lhdGVkIGxvZ2ljKSBkbyB0
aGlzLjxicj4NCiZsdDsgL2pjczImZ3Q7PG86cD48L286cD48L3NwYW4+PC9wPg0KPGRpdj4NCjxw
IGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJ0ZXh0LWFsaWduOmp1c3RpZnk7dGV4dC1qdXN0aWZ5
OmludGVyLWlkZW9ncmFwaCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJjb2xvcjojMUY0OTdE
Ij7igJxGdW5jdGlvbiBvd25lcuKAnTogWWVzLCBJ4oCZbSBzdWdnZXN0aW5nIHRvIHN0cnVjdHVy
ZSBJbnRlbnQgYXJvdW5kIEF1dG9ub21pYyBGdW5jdGlvbnMuIFRoaXMgaXMgbm90IG9wdGltYWwg
ZnJvbSBhIHNjaWVudGlmaWMgcG9pbnQgb2YgdmlldywNCiBidXQgYWxpZ25zIHZlcnkgd2VsbCB3
aXRoIHRoZSB3YXkgaG93IHRoaW5ncyBhcmUgb3JnYW5pc2VkLiBQcm9iYWJseSB3ZSBuZWVkIHRv
IGRpc2N1c3MgdGhhdCBpbiBCQS4gQnV0IHJlZmVyIHRvIG15IFlBTkcgbW9kZWwgc2VudCBhIGZl
dyBkYXlzIGFnby4NCjxicj4NCjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1VUyI+Jmx0O2pjczImZ3Q7
PGJyPg0KSSB3b3VsZCBzdWdnZXN0IGEgc2xpZ2h0IG1vZGlmaWNhdGlvbiwgdG8gZW5hYmxlIHVz
ZXJzIG9mPGJyPg0KaW50ZW50IChhbmQgZXZlbiBpbnRlbnQgd3JpdGVycykgdG8gTk9UIGhhdmUg
dG8gaGF2ZTxicj4NCmRldGFpbGVkIGltcGxlbWVudGF0aW9uIGluZm9ybWF0aW9uLjxicj4NCjxi
cj4NCldoYXQgaWYgYWxsIG9mIHRoZSBBdXRvbm9taWMgRnVuY3Rpb25zLCBhZnRlciB0aGV5IGpv
aW5lZDxicj4NCnRoZSBuZXR3b3JrLCBhZHZlcnRpc2VkIGJvdGggdGhlaXIgY2FwYWJpbGl0aWVz
IGFzIHdlbGwgYXM8YnI+DQp0aGVpciBuZWVkcz8gV2UgY2FuIHB1dCBhc2lkZSB0aGUgbGF0dGVy
IChuZWVkcykgYXMgdGhpczxicj4NCmlzIG1vcmUgY29tcGxleCwgYnV0IGlmIHRoZSBjYXBhYmls
aXRpZXMgYXJlIGFkdmVydGlzZWQsPGJyPg0KdGhlbiB0aGUgYXV0b25vbWljIG1hbmFnZW1lbnQg
c3lzdGVtIGNhbiBhc3NlbWJsZTxicj4NCnRob3NlIG5lZWRzIGludG8gYSBjYXRhbG9nLiBUaGVu
LCB0aGUgaW50ZW50IHdyaXRlciBjYW48YnI+DQpsb29rIGF0IHRoZSBjYXRhbG9nIGFuZCB3cml0
ZSBhIGRlY2xhcmF0aXZlIHJlcXVlc3QgdG88YnI+DQp1c2UgdGhlIGZ1bmN0aW9uYWxpdHkgb2Yg
dGhlIGNhdGFsb2cuPGJyPg0KPGJyPg0KSSB3aWxsIGxvb2sgYXQgeW91ciBZQU5HIG1vZGVsIGlu
IGEgYml0IChzb3JyeSwgSSdtIHdvZWZ1bGx5PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+
DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9InRleHQtYWxpZ246anVzdGlmeTt0
ZXh0LWp1c3RpZnk6aW50ZXItaWRlb2dyYXBoIj48c3BhbiBsYW5nPSJFTi1VUyI+YmVoaW5kLi4u
KS48YnI+DQombHQ7IC9qY3MyJmd0OzxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPHAg
Y2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJn
aW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1VUyI+Jm5ic3A7PG86cD48L286cD48
L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFs
dDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1VUyI+Jm5i
c3A7PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1z
by1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBs
YW5nPSJFTi1VUyI+cmVnYXJkcyw8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNv
Tm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20t
YWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLVVTIj5Kb2huPG86cD48L286cD48L3NwYW4+PC9wPg0K
PC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9InRleHQtYWxpZ246anVz
dGlmeTt0ZXh0LWp1c3RpZnk6aW50ZXItaWRlb2dyYXBoIj48c3BhbiBsYW5nPSJFTi1VUyI+PG86
cD4mbmJzcDs8L286cD48L3NwYW4+PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0
eWxlPSJ0ZXh0LWFsaWduOmp1c3RpZnk7dGV4dC1qdXN0aWZ5OmludGVyLWlkZW9ncmFwaCI+PHNw
YW4gbGFuZz0iRU4tVVMiPk9uIFdlZCwgTWFyIDIzLCAyMDE2IGF0IDM6NTcgQU0sIE1pY2hhZWwg
QmVocmluZ2VyIChtYmVocmluZykgJmx0OzxhIGhyZWY9Im1haWx0bzptYmVocmluZ0BjaXNjby5j
b20iIHRhcmdldD0iX2JsYW5rIj5tYmVocmluZ0BjaXNjby5jb208L2E+Jmd0OyB3cm90ZTo8bzpw
PjwvbzpwPjwvc3Bhbj48L3A+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0
eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+
PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZx
dW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5K
b2huLA0KPC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8
cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1h
cmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1zaXpl
OjExLjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYm
cXVvdDs7Y29sb3I6IzFGNDk3RCI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpw
PjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdp
bi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVO
LUdCIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1
b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+SSB0aGluayB3ZeKAmXJl
IG9uIHRoZSBzYW1lIHBhZ2UuIENvdXBsZSBvZiBwb2ludHMsIHRvcCBwb3N0aW5nIGhlcmUsIG90
aGVyd2lzZSBpdCBnZXRzDQogdG9vIG1lc3N5Ljwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86
cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJn
aW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJF
Ti1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZx
dW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48
c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs
dDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1m
YW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMx
RjQ5N0QiPldoaWxlICpDSUVzIGRvbuKAmXQgbGlrZSBtYWdpYywgdGhleSBkbyB1bmRlcnN0YW5k
IHRoZSBjb25jZXB0IG9mIGEgZGVmYXVsdCBiZWhhdmlvdXINCiB3aGljaCB0aGV5IGNhbiBpbmZs
dWVuY2UgaWYgbmVlZGVkLiBBbnkgcmVhc29uYWJsZSBuZXR3b3JrIG1hbmFnZW1lbnQgYXBwcm9h
Y2ggdHJpZXMgdG8gZGVmaW5lIOKAnGRlZmF1bHTigJ0gYmVoYXZpb3VyIGFuZCDigJxleGNlcHRp
b25z4oCdLiBUaGUgQU4gc3RvcnkgZmlsbHMgdGhlIOKAnGRlZmF1bHTigJ0gcGFydCB2ZXJ5IG5p
Y2VseSwgYW5kICpDSUVzIGdldCB0aGF0Lg0KPC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpw
PjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdp
bi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVO
LUdCIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1
b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+Jm5ic3A7PC9zcGFuPjxz
cGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9y
bWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0
OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZh
bWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFG
NDk3RCI+VGhlIOKAnHBhcmFtZXRlcuKAnSB2ZXJzdXMg4oCccG9saWN54oCdIGRpc2N1c3Npb246
IE15IHZpZXcgaXMgdGhhdCB3ZSBzaG91bGRu4oCZdCB0cnkgdG8gZGVmaW5lDQogZXhhY3RseSAq
PGI+d2hhdDwvYj4qIGludGVudCBhbGxvd3MgYW5kIGRvZXNu4oCZdCBhbGxvdywgc2luY2Ugd2Xi
gJl2ZSBoZWFyZCBmcm9tIG1hbnkgc2lkZXMgdGhhdCB0aGUgYm91bmRhcnkgaXMgbm90IGNsZWFy
LCBhbmQgaXQgbWF5IGJlIGltcG9zc2libGUgdG8gaGF2ZSBhIHJlYWxseSBjbGVhciwgdXNhYmxl
IGRlZmluaXRpb24uIFNvIG15IHBvaW50IHdhczogTGV04oCZcyBmb2N1cyBvbiB0aGUgZGlzdHJp
YnV0aW9uIG1vZGVsOiBJbnRlbnQgaXMgZmxvb2RlZA0KIHRvIGFsbCBub2Rlcy4gTGV04oCZcyBO
T1QgZG8gc2VsZWN0aXZlIGZsb29kaW5nLCB1bmljYXN0IEludGVudCB0byBzcGVjaWZpYyBub2Rl
cywgZXRjLCBiZWNhdXNlIHRoYXQgd291bGQgYmx1ciB0aGUgYm91bmRhcnkgZXZlbiB3b3JzZS4N
Cjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xh
c3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4t
Ym90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4w
cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7
O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286
cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9w
LWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIg
c3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90Oywm
cXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPk15IHN1Z2dlc3Rpb246IEludGVu
dCBpcyBmbG9vZGVkIHRvIGFsbCBub2RlcyBpbiBhIGRvbWFpbi4gVGhlIEFTQSBvd25lcnMgZGVm
aW5lIHRoZQ0KIGNvbnRlbnQgb2Ygd2hhdCB0aGV5IHdhbnQgdG8gcHVzaCBpbiB0aGVpciBiaXQg
b2YgdGhlIEludGVudC4gSWYgd2UgaW5zaXN0IG9mIGZ1bGwgZmxvb2RpbmcsIHdlIGVuY291cmFn
ZSB0aGUgcmlnaHQgYmVoYXZpb3VyLiBBbmQgaWYgc29tZW9uZSByZWFsbHkgcHV0cyBpbnRvIGhp
cyBJbnRlbnQg4oCcbm9kZSBBOiBkbyB0aGlzOyBub2RlIEI6IGRvIHRoZSBvdGhlcjsgbm9kZSBD
OiBkbyB0aGUgdGhpcmTigJ0gaXQgaXMgYXdrd2FyZCBlbm91Z2ggZm9yDQogcGVvcGxlIHRvIHRh
a2UgYSBzdGVwIGJhY2sgYW5kIHRoaW5rIHdoZXRoZXIgdGhleeKAmXJlIGRvaW5nIHRoZSByaWdo
dCB0aGluZyA8L3NwYW4+DQo8c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4w
cHQ7Zm9udC1mYW1pbHk6V2luZ2RpbmdzO2NvbG9yOiMxRjQ5N0QiPko8L3NwYW4+PHNwYW4gbGFu
Zz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGli
cmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj4NCjwvc3Bhbj48
c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs
dDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1m
YW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMx
RjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+
PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRv
O21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZv
bnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5z
LXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPiZsdDtqY3MmZ3Q7PC9zcGFuPjxzcGFuIGxhbmc9
IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHls
ZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxz
cGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVv
dDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+RG8g
eW91IG1lYW4gdGhhdCB0aGUgaW50ZW50IHdyaXRlciB3aWxsICoqa25vdyB0aGUgZGlmZmVyZW5j
ZSoqPC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBj
bGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdp
bi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1zaXplOjEx
LjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVv
dDs7Y29sb3I6IzFGNDk3RCI+YmV0d2VlbiB0aGUgZmFicmljIGFuZCBlYWNoIGF1dG9ub21pYyBm
dW5jdGlvbj8gSSBob3BlIG5vdCw8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+
PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1h
bHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0
eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1
b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5idXQgcGVyaGFwcyBJJ20gbWlzdW5k
ZXJzdGFuZGluZyB5b3UuLi48L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9z
cGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6
YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxl
PSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7
c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj4mbHQ7L2pjcyZndDs8L3NwYW4+PHNwYW4g
bGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi
IHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0
byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5
OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdE
Ij4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4N
CjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28t
bWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNp
emU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJp
ZiZxdW90Oztjb2xvcjojMUY0OTdEIj5Ob3Qgc3VyZSBJIHVuZGVyc3RhbmQgeW91ciBjb25jZXJu
LiBZZXMsIGluIG15IG1vZGVsIHRoZXJlIGlzIGEgc2VjdGlvbiBmb3IgdGhlIGZhYnJpYywNCiBh
bmQgc2VjdGlvbnMgZm9yIHRoZSB2YXJpb3VzIGZ1bmN0aW9ucy4gU2VlbXMgYSBuYXR1cmFsIHN0
cnVjdHVyZT8hPyA8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwv
cD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bztt
c28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250
LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1z
ZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0Ii
PjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28t
bWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFu
Zz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGli
cmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj7igJxGdW5jdGlv
biBvd25lcuKAnTogWWVzLCBJ4oCZbSBzdWdnZXN0aW5nIHRvIHN0cnVjdHVyZSBJbnRlbnQgYXJv
dW5kIEF1dG9ub21pYyBGdW5jdGlvbnMuDQogVGhpcyBpcyBub3Qgb3B0aW1hbCBmcm9tIGEgc2Np
ZW50aWZpYyBwb2ludCBvZiB2aWV3LCBidXQgYWxpZ25zIHZlcnkgd2VsbCB3aXRoIHRoZSB3YXkg
aG93IHRoaW5ncyBhcmUgb3JnYW5pc2VkLiBQcm9iYWJseSB3ZSBuZWVkIHRvIGRpc2N1c3MgdGhh
dCBpbiBCQS4gQnV0IHJlZmVyIHRvIG15IFlBTkcgbW9kZWwgc2VudCBhIGZldyBkYXlzIGFnby4N
Cjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xh
c3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4t
Ym90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4w
cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7
O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286
cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9w
LWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIg
c3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90Oywm
cXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPldlIHNob3VsZCBzY2hlZHVsZSBz
b21lIGRpc2N1c3Npb24gdGltZSBvdXRzaWRlIHRoZSBBTklNQSBtZWV0aW5nIHdpbmRvdy4uLjwv
c3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9
Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90
dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7
Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2Nv
bG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48
L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFs
dDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5
bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVv
dDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPk1pY2hhZWw8L3NwYW4+PHNwYW4gbGFu
Zz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0
eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+
PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZx
dW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj4m
bmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxw
IGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFy
Z2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6
MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZx
dW90Oztjb2xvcjojMUY0OTdEIj4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+
PC9vOnA+PC9zcGFuPjwvcD4NCjxkaXYgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNv
bGlkIHdpbmRvd3RleHQgMS41cHQ7cGFkZGluZzowY20gMGNtIDBjbSA0LjBwdDtib3JkZXItY29s
b3I6Y3VycmVudENvbG9yIGN1cnJlbnRDb2xvciBjdXJyZW50Q29sb3IgYmx1ZSI+DQo8ZGl2Pg0K
PGRpdiBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVyLXRvcDpzb2xpZCB3aW5kb3d0ZXh0IDEuMHB0
O3BhZGRpbmc6My4wcHQgMGNtIDBjbSAwY207Ym9yZGVyLWNvbG9yOmN1cnJlbnRDb2xvciBjdXJy
ZW50Q29sb3IiPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFs
dDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48Yj48c3BhbiBsYW5nPSJFTi1VUyIg
c3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90Oywm
cXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij5Gcm9tOjwvc3Bhbj48L2I+PHNwYW4gbGFuZz0iRU4tVVMi
IHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDss
JnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+DQogSm9obiBTdHJhc3NuZXIgW21haWx0bzo8YSBocmVm
PSJtYWlsdG86c3RyYXpwZGpAZ21haWwuY29tIiB0YXJnZXQ9Il9ibGFuayI+c3RyYXpwZGpAZ21h
aWwuY29tPC9hPl0NCjxicj4NCjxiPlNlbnQ6PC9iPiAyMiBNYXJjaCAyMDE2IDIxOjMyPGJyPg0K
PGI+VG86PC9iPiBNaWNoYWVsIEJlaHJpbmdlciAobWJlaHJpbmcpICZsdDs8YSBocmVmPSJtYWls
dG86bWJlaHJpbmdAY2lzY28uY29tIiB0YXJnZXQ9Il9ibGFuayI+bWJlaHJpbmdAY2lzY28uY29t
PC9hPiZndDs7IEpvaG4gU3RyYXNzbmVyICZsdDs8YSBocmVmPSJtYWlsdG86c3RyYXpwZGpAZ21h
aWwuY29tIiB0YXJnZXQ9Il9ibGFuayI+c3RyYXpwZGpAZ21haWwuY29tPC9hPiZndDs8YnI+DQo8
Yj5DYzo8L2I+IER1em9uZ3BlbmcgJmx0OzxhIGhyZWY9Im1haWx0bzpkdXpvbmdwZW5nQGh1YXdl
aS5jb20iIHRhcmdldD0iX2JsYW5rIj5kdXpvbmdwZW5nQGh1YXdlaS5jb208L2E+Jmd0OzsNCjxh
IGhyZWY9Im1haWx0bzphbmltYUBpZXRmLm9yZyIgdGFyZ2V0PSJfYmxhbmsiPmFuaW1hQGlldGYu
b3JnPC9hPjsgTGF1cmVudCBDaWF2YWdsaWEgJmx0OzxhIGhyZWY9Im1haWx0bzpsYXVyZW50LmNp
YXZhZ2xpYUBub2tpYS5jb20iIHRhcmdldD0iX2JsYW5rIj5sYXVyZW50LmNpYXZhZ2xpYUBub2tp
YS5jb208L2E+Jmd0OzsgSsOpZmVyc29uIENhbXBvcyBOb2JyZSAmbHQ7PGEgaHJlZj0ibWFpbHRv
Ompjbm9icmVAaW5mLnVmcmdzLmJyIiB0YXJnZXQ9Il9ibGFuayI+amNub2JyZUBpbmYudWZyZ3Mu
YnI8L2E+Jmd0OzsNCiBTaGVuZyBKaWFuZyAmbHQ7PGEgaHJlZj0ibWFpbHRvOmppYW5nc2hlbmdA
aHVhd2VpLmNvbSIgdGFyZ2V0PSJfYmxhbmsiPmppYW5nc2hlbmdAaHVhd2VpLmNvbTwvYT4mZ3Q7
PGJyPg0KPGI+U3ViamVjdDo8L2I+IFJlOiBbQW5pbWFdIERvZXMgQXV0b25vbWljIG5ldHdvcmsg
bmVlZCBzb21lIGludGVydmVudGlvbiBiZXlvbmQgaW50ZW50PC9zcGFuPjxzcGFuIGxhbmc9IkVO
LUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPHAgY2xhc3M9Ik1z
b05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9t
LWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1Rp
bWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFu
IGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNs
YXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2lu
LWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTom
cXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPkluIGdlbmVyYWws
IEkgYWdyZWUgd2l0aCBNaWNoYWVsLiBJJ2QgbGlrZSB0byBlbXBoYXNpemUgYW5kIGV4cGFuZCBv
biBhIGNvdXBsZSBvZiBrZXkgcG9pbnRzOjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48
L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5
bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48
c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21h
biZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdC
Ij48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9y
bWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0
OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMg
TmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij4mbmJzcDs8L3NwYW4+PHNwYW4gbGFu
Zz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNz
PSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJv
dHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0
O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztj
b2xvcjojMUY0OTdEIj5GaXJzdCwgQU5JTUEgaXNu4oCZdCB0aGUgb25seSB0aGluZyDigJxpbmZs
dWVuY2luZ+KAnSBvciDigJxzdGVlcmluZ+KAnSBhIG5ldHdvcmssIGF0IGxlYXN0IG5vdA0KIGlu
aXRpYWxseS4gVGhlcmUgaXMgc3RpbGwgdHJhZGl0aW9uYWwgY29uZmlnLCBwbHVzIGFsbCB0aGUg
ZXhpc3RpbmcgYW5kIGVtZXJnaW5nIFNETiBtb2RlbHMuIE15IHZpZXcgaGFzIGFsd2F5cyBiZWVu
IHRoYXQgQU5JTUEgc2V0cyBhIOKAnGRlZmF1bHTigJ0gcG9saWN5IGluIGEgbmV0d29yaywgYW5k
IGFsbCBvdGhlciBtZWFucyBhcmUgbW9yZSBzcGVjaWZpYyBhbmQgY2FuIG92ZXJyaWRlIGludGVu
dC4gVGhpcyBpcyBkZXNjcmliZWQgaW4gUkZDNzU3NS4NCiBNYWluIHBvaW50IGhlcmU6IEFOSU1B
IEludGVudCBpc27igJl0IHRoZSBvbmx5IHRvb2wgaW4gYSBuZXR3b3JrLiA8L3NwYW4+PHNwYW4g
bGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi
IHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0
byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcg
Um9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJF
Ti1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1z
b05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9t
LWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1Rp
bWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Jmx0O2pjcyZndDs8L3NwYW4+
PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4N
CjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28t
bWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZh
bWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPkFic29s
dXRlbHkgYWdyZWUuIE5vdGUgdGhhdCBldmVuIGluIHRoZSBDb25ncmVzcyBjYXNlIEkgbWVudGlv
bmVkPC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rp
dj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0
OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHls
ZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1
b3Q7Ij5lYXJsaWVyLCBpdCBpcyBOT1QgcHVzaGluZyBjb25maWcsIGFuZCBpdCBpcyBOT1Qgdmll
d2VkIGFzIHRoZTwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9w
Pg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4t
dG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1H
QiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtz
ZXJpZiZxdW90OyI+JnF1b3Q7b25seSZxdW90OyBwb2xpY3k8L3NwYW4+PHNwYW4gbGFuZz0iRU4t
R0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFz
cz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1i
b3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1
b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij4mbHQ7L2pjcyZndDs8
L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0K
PC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9w
LWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIg
c3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJp
ZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bh
bj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1h
cmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9
IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZx
dW90O3NlcmlmJnF1b3Q7Ij4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9v
OnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxl
PSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNw
YW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90
O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5JbnRl
bnQgY291bGQgcGVyZmVjdGx5IHdlbGwgY292ZXIgYSBoaWdoIGxldmVsIHBvbGljeSBzdWNoIGFz
IOKAnGFsbCBub2RlcyBvZiB0eXBlIHggZG8NCiB0aGlzOyB0eXBlIHkgZG9lcyB0aGF04oCdLiAo
V2hpY2ggSSB0aGluayBpcyB3aGF0IHlvdSBtZW50aW9uIGJlbG93LikgV2hlbiB3ZSBzdGFydGVk
IHdpdGggSW50ZW50LCB0aGUgdmlzaW9uIHdhcyB0byBrZWVwIGl0IHZlcnkgaGlnaCBsZXZlbCwg
bW9yZSBhbG9uZyB0aGUgbGluZSDigJxkbyB0aGUgcmlnaHQgdGhpbmfigJ0NCjwvc3Bhbj48c3Bh
biBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6V2luZ2Rp
bmdzO2NvbG9yOiMxRjQ5N0QiPko8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250
LXNpemU6MTEuMHB0O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1H
QiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1z
by1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBs
YW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90
OywmcXVvdDtzZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpw
PjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs
dDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVz
IE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Jmx0O2pjcyZndDs8L3NwYW4+PHNw
YW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxw
IGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFy
Z2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWls
eTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPkFncmVlZC4g
QWx0aG91Z2ggdG8gbWUsIHRoaXMgbWVhbnMgdGhhdCB3ZSBuZWVkIHRvIGNvbnNpZGVyIHRoZSB0
eXBlcyBvZiBhY3RvcnMgdGhhdCBhcmUgdXNpbmcgaW50ZW50LiBGb3IgZXhhbXBsZSwNCiBtb3N0
IG9mIG15IENDSUUgYW5kIEpDSUUgZnJpZW5kcyBoYXZlIG5vIGludGVyZXN0IGluIHVzaW5nIGFu
IGludGVudCBsYW5ndWFnZSBmb3IgcHVzaGluZyBjb25maWd1cmF0aW9uLiBPbiB0aGUgb3RoZXIg
aGFuZCwgd2hlbiBJIHRhbGsgdG8gb3BlcmF0b3JzIGFuZCBhcHBsaWNhdGlvbiBkZXZlbG9wZXJz
LCB0aGV5IHdvdWxkIGxvdmUgdG8gYmUgYWJsZSB0byB1c2UgYSBoaWdoLWxldmVsIGludGVudCBh
bmQgaGF2ZSB0aGUgaW50ZW50IGVuZ2luZQ0KIHRha2UgY2FyZSBvZiB0aGUgZGV0YWlscyBmb3Ig
dGhlbS48L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwv
ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1h
bHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0
eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYm
cXVvdDsiPiZsdDsvamNzJmd0Ozwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48
L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1z
by1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBs
YW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90
OywmcXVvdDtzZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpw
PjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFz
cz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1i
b3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1
b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0Qi
Pkl0IGhhcyBiZWNvbWUgY2xlYXIgdGhhdCBpbiB0b2RheeKAmXMgbmV0d29ya2luZywgcGVvcGxl
IHdpbGwgd2FudCB0byBwdXNoIG1vcmUg4oCccGFyYW1ldGVyc+KAnSB0aGFuIOKAnHBvbGljeeKA
nSwNCiBhbmQgSSB0aGluayB3ZeKAmWxsIGp1c3QgaGF2ZSB0byBhY2tub3dsZWRnZSB0aGF0Ljwv
c3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8
ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRv
O21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZv
bnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+
Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8
L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3At
YWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBz
dHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3Nlcmlm
JnF1b3Q7Ij4mbHQ7amNzJmd0Ozwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48
L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1z
by1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBs
YW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90
OywmcXVvdDtzZXJpZiZxdW90OyI+U29ycnkgTWljaGFlbCwgSSBkb24ndCBmb2xsb3cgaGVyZS4g
V2hhdCBkbyB5b3UgbWVhbjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3Nw
YW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1t
YXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5n
PSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90Oywm
cXVvdDtzZXJpZiZxdW90OyI+YnkgJnF1b3Q7cGFyYW1ldGVycyZxdW90Oz88L3NwYW4+PHNwYW4g
bGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNs
YXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2lu
LWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTom
cXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPiZsdDsvamNzJmd0
Ozwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+
DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDph
dXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9
ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90
OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+
DQo8L2Rpdj4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1t
YXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5n
PSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90Oywm
cXVvdDtzZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwv
bzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHls
ZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxz
cGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVv
dDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+UGVy
c29uYWxseSBJIHRoaW5rIGl04oCZbGwgcGFuIG91dCBsaWtlIHRoaXM6IEludGVudCB3aWxsIGhh
dmUgc2VjdGlvbnMgcGVyIGF1dG9ub21pYw0KIGZ1bmN0aW9uLiBBbiBJbnRlbnQgcGFyc2VyIHdp
bGwgdmFsaWRhdGUgb3ZlcmFsbCBjb3JyZWN0bmVzcyAoc2lnbmF0dXJlLCBldGMpLCB0aGVuIHNl
Z21lbnQgdGhlIEludGVudCBpbnRvIGl0cyBiaXRzOiBPbmUgcGFydCB3aWxsIGJlIHRoZSBhdXRv
bm9taWMgZmFicmljIGl0c2VsZiwgdGhlbiB0aGVyZSB3aWxsIGJlIGJpdHMgZm9yIGVhY2ggYXV0
b25vbWljIGZ1bmN0aW9uLg0KPC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwv
c3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0
OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHls
ZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1
b3Q7Ij4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwv
cD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2lu
LXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4t
R0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7
c2VyaWYmcXVvdDsiPiZsdDtqY3MmZ3Q7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwv
bzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHls
ZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxz
cGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFu
JnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij5EbyB5b3UgbWVhbiB0aGF0IHRoZSBpbnRlbnQgd3Jp
dGVyIHdpbGwgKiprbm93IHRoZSBkaWZmZXJlbmNlKio8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0Ii
PjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3Jt
YWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6
YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBO
ZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPmJldHdlZW4gdGhlIGZhYnJpYyBhbmQg
ZWFjaCBhdXRvbm9taWMgZnVuY3Rpb24/IEkgaG9wZSBub3QsPC9zcGFuPjxzcGFuIGxhbmc9IkVO
LUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNv
Tm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20t
YWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGlt
ZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij5idXQgcGVyaGFwcyBJJ20gbWlz
dW5kZXJzdGFuZGluZyB5b3UuLi48L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+
PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJt
c28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4g
bGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVv
dDssJnF1b3Q7c2VyaWYmcXVvdDsiPiZsdDsvamNzJmd0Ozwvc3Bhbj48c3BhbiBsYW5nPSJFTi1H
QiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs
dDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVz
IE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxh
bmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFz
cz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1i
b3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1
b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij4mbmJzcDs8L3NwYW4+
PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4N
CjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28t
bWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNp
emU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJp
ZiZxdW90Oztjb2xvcjojMUY0OTdEIj5TbyB0aGUgSW50ZW50IHBhcnNlciBvbiBhIG5vZGUgd2ls
bCBncmFiIGludGVudCwgc2VlIHRoZXJlIGlzIHNvbWUgSW50ZW50IGZvciBhdXRvbm9taWMNCiBm
dW5jdGlvbiBBLCB3aWxsIHNlZSB3aGV0aGVyIHRoaXMgaXMgcnVubmluZyBsb2NhbGx5LCBhbmQg
aWYgc28sIGl04oCZbGwganVzdCBwYXNzIHRoYXQgcGllY2Ugb2YgSW50ZW50IHRvIHRoYXQgZnVu
Y3Rpb24uDQo8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4N
CjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1h
cmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9
IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZx
dW90O3NlcmlmJnF1b3Q7Ij4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9v
OnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxl
PSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNw
YW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4m
cXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPiZsdDtqY3MmZ3Q7PC9zcGFuPjxzcGFuIGxhbmc9IkVO
LUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNv
Tm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20t
YWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGlt
ZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij5BZ3JlZWQuPC9zcGFuPjxzcGFu
IGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBj
bGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdp
bi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6
JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij4mbHQ7L2pjcyZn
dDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2
Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6
YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxl
PSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVv
dDsiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9w
Pg0KPC9kaXY+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28t
bWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFu
Zz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDss
JnF1b3Q7c2VyaWYmcXVvdDsiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48
L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5
bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48
c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1
b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPklu
IHRoaXMgd2F5LCBhIGZ1bmN0aW9uIG93bmVyIGRlZmluZXMgYm90aCB0aGUgaW50ZW50IGl0c2Vs
ZiwgYW5kIGhvdyBpdCBpcyBjb25zdW1lZA0KIGJ5IGhpcy9oZXIgZnVuY3Rpb24uIDwvc3Bhbj48
c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs
dDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVz
IE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxh
bmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFz
cz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1i
b3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1
b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij4mbHQ7amNzJmd0Ozwv
c3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8
ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRv
O21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZv
bnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+
SSdtIG5vdCBzdXJlIHdoYXQgeW91IG1lYW4gYnkgJnF1b3Q7ZnVuY3Rpb24gb3duZXImcXVvdDsu
IChTb3JyeSB0byBiZTwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+
PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJn
aW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJF
Ti1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVv
dDtzZXJpZiZxdW90OyI+cGlja3ksIEknbSBsaWtlbHkgdW5kZXItY2FmZmVpbmF0ZWQuKTwvc3Bh
bj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2
Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21z
by1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQt
ZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Jm5i
c3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rp
dj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0
OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHls
ZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1
b3Q7Ij5UaGUgYWJvdmUgc2VlbWVkIGxpa2UgYW4gYXV0b25vbW91cyBkZWNpc2lvbiwgbWFkZSBi
eSB0aGUgcGFyc2VyLjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+
PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJn
aW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJF
Ti1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVv
dDtzZXJpZiZxdW90OyI+SSBkb24ndCB0aGluayB0aGF0IHRoZSBpbnRlbnQgd3JpdGVyIHNob3Vs
ZCBiZSBjb2duaXphbnQgb2YgaG93IG1hbnkgb3I8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxv
OnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi
IHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0
byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcg
Um9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPndoYXQgdHlwZXMgb2YgYXV0b25vbWljIGZ1
bmN0aW9ucyBhcmUgcHJlc2VudC48L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+
PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJt
c28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4g
bGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVv
dDssJnF1b3Q7c2VyaWYmcXVvdDsiPiZsdDsvamNzJmd0Ozwvc3Bhbj48c3BhbiBsYW5nPSJFTi1H
QiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs
dDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVz
IE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxh
bmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFz
cz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1i
b3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1
b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij4mbmJzcDs8L3NwYW4+
PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4N
CjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1
dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0i
Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2Nv
bG9yOiMxRjQ5N0QiPlRoZSBkb3duc2lkZSBpcyB0aGF0IHRoZXJlIGlzIG5vIOKAnGNvbnRyb2zi
gJ0gYW5kIHBlb3BsZSBjYW4gZG8gd2hhdCB0aGV5IHdhbnQsIGluIHRoZSB3b3JzdCBjYXNlIGV2
ZW4gcHVzaA0KIGNvbmZpZywgd2hpY2ggaXMgY2xlYXJseSBOT1Qgd2hhdCBJbnRlbnQgaXMgbWVh
bnQgdG8gZG8uIDwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9w
Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21z
by1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQt
ZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Jm5i
c3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rp
dj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0
OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHls
ZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1
b3Q7Ij4mbHQ7amNzJmd0Ozwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3Nw
YW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1t
YXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5n
PSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90Oywm
cXVvdDtzZXJpZiZxdW90OyI+QW5kIGhlbmNlLCB0aGUgJnF1b3Q7SEFMIDIwMDAmcXVvdDsgc2Nl
bmFyaW8uIDotKTwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9w
Pg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4t
dG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1H
QiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtz
ZXJpZiZxdW90OyI+Jmx0Oy9qY3MmZ3Q7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwv
bzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHls
ZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxz
cGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFu
JnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0Ii
PjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0i
TXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0
b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7
VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij4mbmJzcDs8L3NwYW4+PHNw
YW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxw
IGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFy
Z2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWls
eTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3
RCI+SeKAmWQgaG9wZSB0aGF0IHBlb3BsZSBkb27igJl0IHN0YXJ0IHdyaXRpbmcgSW50ZW50IGxp
a2Ug4oCcbm9kZSAxNzogaGVyZSBpcyB5b3VyIENMSTsgbm9kZSAyMzogaGVyZSBpcyB5b3VyDQog
Q0xJ4oCdOyBub2RlIDM3OiBoZXJlIGlzIHlvdXIgQ0xJ4oCdLiBCdXQgd2hhdCBJIGRlc2NyaWJl
IGFib3ZlIENPVUxEIGJlIGFidXNlZCBpbiB0aGlzIHdheS4gUkZDNzU3NSBzYXlzIHRoaXMgaXMg
Tk9UIEludGVudC4NCjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+
PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJn
aW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJF
Ti1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVv
dDtzZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpw
Pjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0i
bXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFu
IGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1
b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij4mbHQ7amNzJmd0Ozwvc3Bhbj48c3BhbiBsYW5nPSJFTi1H
QiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs
dDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVz
IE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+QWJzb2x1dGVseSBhZ3JlZSE8L3Nw
YW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRp
dj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bztt
c28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250
LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPiZs
dDsvamNzJmd0Ozwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9w
Pg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4t
dG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1H
QiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtz
ZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwv
c3Bhbj48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+
DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10
b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdC
IiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7
LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+Jm5ic3A7cmVnYXJkcyw8L3Nw
YW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJN
c29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRv
bS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2Zv
bnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xv
cjojMUY0OTdEIj5Kb2huPC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bh
bj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1
dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0i
Zm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1b3Q7LCZxdW90O3NlcmlmJnF1b3Q7
O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286
cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9w
LWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIg
c3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5ldyBSb21hbiZxdW90OywmcXVvdDtzZXJp
ZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bh
bj48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxl
PSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNw
YW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4m
cXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+
PG86cD48L286cD48L3NwYW4+PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxl
PSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNw
YW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtUaW1lcyBOZXcgUm9tYW4m
cXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPk9uIFdlZCwgTWFyIDE2LCAyMDE2IGF0IDE6MTYgQU0s
IE1pY2hhZWwgQmVocmluZ2VyIChtYmVocmluZykgJmx0OzxhIGhyZWY9Im1haWx0bzptYmVocmlu
Z0BjaXNjby5jb20iIHRhcmdldD0iX2JsYW5rIj5tYmVocmluZ0BjaXNjby5jb208L2E+Jmd0Ow0K
IHdyb3RlOjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0K
PGJsb2NrcXVvdGUgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkIHdpbmRvd3Rl
eHQgMS4wcHQ7cGFkZGluZzowY20gMGNtIDBjbSA2LjBwdDttYXJnaW4tbGVmdDo0LjhwdDttYXJn
aW4tdG9wOjUuMHB0O21hcmdpbi1yaWdodDowY207bWFyZ2luLWJvdHRvbTo1LjBwdDtib3JkZXIt
Y29sb3I6Y3VycmVudENvbG9yIGN1cnJlbnRDb2xvciBjdXJyZW50Q29sb3IgcmdiKDIwNCwyMDQs
MjA0KSI+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFy
Z2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0i
RU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkm
cXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5IaSBab25ncGVuZywN
Cjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xh
c3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4t
Ym90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4w
cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7
O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286
cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9w
LWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIg
c3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90Oywm
cXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPlllcywgdGhpcyBpcyBhIHJlY3Vy
cmluZyBxdWVzdGlvbiBhbmQgdG9waWMuIExldCBtZSBnaXZlIG15IHZpZXcgb24gdHdvIGxldmVs
czoNCjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAg
Y2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJn
aW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZTox
MS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1
b3Q7O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48
L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4t
dG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1H
QiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90
OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPkZpcnN0LCBBTklNQSBpc27i
gJl0IHRoZSBvbmx5IHRoaW5nIOKAnGluZmx1ZW5jaW5n4oCdIG9yIOKAnHN0ZWVyaW5n4oCdIGEg
bmV0d29yaywgYXQgbGVhc3Qgbm90DQogaW5pdGlhbGx5LiBUaGVyZSBpcyBzdGlsbCB0cmFkaXRp
b25hbCBjb25maWcsIHBsdXMgYWxsIHRoZSBleGlzdGluZyBhbmQgZW1lcmdpbmcgU0ROIG1vZGVs
cy4gTXkgdmlldyBoYXMgYWx3YXlzIGJlZW4gdGhhdCBBTklNQSBzZXRzIGEg4oCcZGVmYXVsdOKA
nSBwb2xpY3kgaW4gYSBuZXR3b3JrLCBhbmQgYWxsIG90aGVyIG1lYW5zIGFyZSBtb3JlIHNwZWNp
ZmljIGFuZCBjYW4gb3ZlcnJpZGUgaW50ZW50LiBUaGlzIGlzIGRlc2NyaWJlZCBpbiBSRkM3NTc1
Lg0KIE1haW4gcG9pbnQgaGVyZTogQU5JTUEgSW50ZW50IGlzbuKAmXQgdGhlIG9ubHkgdG9vbCBp
biBhIG5ldHdvcmsuIDwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+
PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRv
O21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZv
bnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5z
LXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1H
QiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1z
by1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBs
YW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2Fs
aWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPkludGVudCBj
b3VsZCBwZXJmZWN0bHkgd2VsbCBjb3ZlciBhIGhpZ2ggbGV2ZWwgcG9saWN5IHN1Y2ggYXMg4oCc
YWxsIG5vZGVzIG9mIHR5cGUgeCBkbw0KIHRoaXM7IHR5cGUgeSBkb2VzIHRoYXTigJ0uIChXaGlj
aCBJIHRoaW5rIGlzIHdoYXQgeW91IG1lbnRpb24gYmVsb3cuKSBXaGVuIHdlIHN0YXJ0ZWQgd2l0
aCBJbnRlbnQsIHRoZSB2aXNpb24gd2FzIHRvIGtlZXAgaXQgdmVyeSBoaWdoIGxldmVsLCBtb3Jl
IGFsb25nIHRoZSBsaW5lIOKAnGRvIHRoZSByaWdodCB0aGluZ+KAnQ0KPC9zcGFuPjxzcGFuIGxh
bmc9IkVOLUdCIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTpXaW5nZGluZ3M7
Y29sb3I6IzFGNDk3RCI+Sjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6
ZToxMS4wcHQ7Y29sb3I6IzFGNDk3RCI+Jm5ic3A7IEl0IGhhcyBiZWNvbWUgY2xlYXIgdGhhdCBp
biB0b2RheeKAmXMgbmV0d29ya2luZywgcGVvcGxlIHdpbGwgd2FudCB0byBwdXNoIG1vcmUg4oCc
cGFyYW1ldGVyc+KAnSB0aGFuIOKAnHBvbGljeeKAnSwNCiBhbmQgSSB0aGluayB3ZeKAmWxsIGp1
c3QgaGF2ZSB0byBhY2tub3dsZWRnZSB0aGF0LiA8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxv
OnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFy
Z2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0i
RU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkm
cXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj4mbmJzcDs8L3NwYW4+
PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29O
b3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1h
bHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQt
ZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjoj
MUY0OTdEIj5QZXJzb25hbGx5IEkgdGhpbmsgaXTigJlsbCBwYW4gb3V0IGxpa2UgdGhpczogSW50
ZW50IHdpbGwgaGF2ZSBzZWN0aW9ucyBwZXIgYXV0b25vbWljDQogZnVuY3Rpb24uIEFuIEludGVu
dCBwYXJzZXIgd2lsbCB2YWxpZGF0ZSBvdmVyYWxsIGNvcnJlY3RuZXNzIChzaWduYXR1cmUsIGV0
YyksIHRoZW4gc2VnbWVudCB0aGUgSW50ZW50IGludG8gaXRzIGJpdHM6IE9uZSBwYXJ0IHdpbGwg
YmUgdGhlIGF1dG9ub21pYyBmYWJyaWMgaXRzZWxmLCB0aGVuIHRoZXJlIHdpbGwgYmUgYml0cyBm
b3IgZWFjaCBhdXRvbm9taWMgZnVuY3Rpb24uDQo8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxv
OnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFy
Z2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0i
RU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkm
cXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj4mbmJzcDs8L3NwYW4+
PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29O
b3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1h
bHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQt
ZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjoj
MUY0OTdEIj5TbyB0aGUgSW50ZW50IHBhcnNlciBvbiBhIG5vZGUgd2lsbCBncmFiIGludGVudCwg
c2VlIHRoZXJlIGlzIHNvbWUgSW50ZW50IGZvciBhdXRvbm9taWMNCiBmdW5jdGlvbiBBLCB3aWxs
IHNlZSB3aGV0aGVyIHRoaXMgaXMgcnVubmluZyBsb2NhbGx5LCBhbmQgaWYgc28sIGl04oCZbGwg
anVzdCBwYXNzIHRoYXQgcGllY2Ugb2YgSW50ZW50IHRvIHRoYXQgZnVuY3Rpb24uDQo8L3NwYW4+
PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29O
b3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1h
bHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQt
ZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjoj
MUY0OTdEIj4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFu
PjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0
bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJm
b250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fu
cy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5JbiB0aGlzIHdheSwgYSBmdW5jdGlvbiBvd25l
ciBkZWZpbmVzIGJvdGggdGhlIGludGVudCBpdHNlbGYsIGFuZCBob3cgaXQgaXMgY29uc3VtZWQN
CiBieSBoaXMvaGVyIGZ1bmN0aW9uLiA8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9v
OnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRv
cC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0Ii
IHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDss
JnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj4mbmJzcDs8L3NwYW4+PHNwYW4g
bGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi
IHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0
byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5
OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdE
Ij5UaGUgZG93bnNpZGUgaXMgdGhhdCB0aGVyZSBpcyBubyDigJxjb250cm9s4oCdIGFuZCBwZW9w
bGUgY2FuIGRvIHdoYXQgdGhleSB3YW50LCBpbiB0aGUNCiB3b3JzdCBjYXNlIGV2ZW4gcHVzaCBj
b25maWcsIHdoaWNoIGlzIGNsZWFybHkgTk9UIHdoYXQgSW50ZW50IGlzIG1lYW50IHRvIGRvLiBJ
4oCZZCBob3BlIHRoYXQgcGVvcGxlIGRvbuKAmXQgc3RhcnQgd3JpdGluZyBJbnRlbnQgbGlrZSDi
gJxub2RlIDE3OiBoZXJlIGlzIHlvdXIgQ0xJOyBub2RlIDIzOiBoZXJlIGlzIHlvdXIgQ0xJ4oCd
OyBub2RlIDM3OiBoZXJlIGlzIHlvdXIgQ0xJ4oCdLiBCdXQgd2hhdCBJIGRlc2NyaWJlIGFib3Zl
IENPVUxEIGJlIGFidXNlZA0KIGluIHRoaXMgd2F5LiBSRkM3NTc1IHNheXMgdGhpcyBpcyBOT1Qg
SW50ZW50LiA8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4N
CjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28t
bWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNp
emU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJp
ZiZxdW90Oztjb2xvcjojMUY0OTdEIj4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxv
OnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFy
Z2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0i
RU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkm
cXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5JbiBvdGhlciB3b3Jk
czogTGV04oCZcyBub3QgYmUgdG9vIHByZXNjcmlwdGl2ZSwgYW5kIGhvcGUgdGhhdCBmb2xrcyB3
aWxsIGRvIG1vcmUgb3IgbGVzcw0KIHJlYXNvbmFibGUgdGhpbmdzLiA8L3NwYW4+PHNwYW4gbGFu
Zz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0
eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+
PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZx
dW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj4m
bmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxw
IGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFy
Z2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6
MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZx
dW90Oztjb2xvcjojMUY0OTdEIj5J4oCZdmUgYmVlbiBtZWFuaW5nIHRvIGNvbW1lbnQgb24gdGhl
IEludGVudCBkcmFmdHMgYW5kIHBvc3NpYmx5IGNvbnRyaWJ1dGUsIEnigJltIGp1c3QNCiBzdHJ1
Z2dsaW5nIHdpdGggcHJpb3JpdGllcyBhdCB0aGUgbW9tZW50IChhcyB5b3UgY2FuIHNlZSBieSBt
eSByZWxhdGl2ZSBzaWxlbmNlIG92ZXIgdGhlIHBhc3Qgd2Vla3MpLg0KPC9zcGFuPjxzcGFuIGxh
bmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBz
dHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8i
PjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTom
cXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+
Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8
cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1h
cmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1zaXpl
OjExLjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYm
cXVvdDs7Y29sb3I6IzFGNDk3RCI+TWljaGFlbDwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86
cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJn
aW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJF
Ti1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZx
dW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48
c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05v
cm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFs
dDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1m
YW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMx
RjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+
PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRv
O21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZv
bnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5z
LXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1H
QiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPGRpdiBzdHlsZT0iYm9yZGVyOm5vbmU7Ym9yZGVy
LWxlZnQ6c29saWQgd2luZG93dGV4dCAxLjVwdDtwYWRkaW5nOjBjbSAwY20gMGNtIDQuMHB0O2Jv
cmRlci1jb2xvcjpjdXJyZW50Q29sb3IgY3VycmVudENvbG9yIGN1cnJlbnRDb2xvciBibHVlIj4N
CjxkaXY+DQo8ZGl2IHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItdG9wOnNvbGlkIHdpbmRvd3Rl
eHQgMS4wcHQ7cGFkZGluZzozLjBwdCAwY20gMGNtIDBjbTtib3JkZXItY29sb3I6Y3VycmVudENv
bG9yIj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0
bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PGI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxl
PSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7
c2Fucy1zZXJpZiZxdW90OyI+RnJvbTo8L3NwYW4+PC9iPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHls
ZT0iZm9udC1zaXplOjExLjBwdCI+IEFuaW1hIFttYWlsdG86PC9zcGFuPjxzcGFuIGxhbmc9IkVO
LUdCIj48YSBocmVmPSJtYWlsdG86YW5pbWEtYm91bmNlc0BpZXRmLm9yZyIgdGFyZ2V0PSJfYmxh
bmsiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdCI+YW5pbWEtYm91
bmNlc0BpZXRmLm9yZzwvc3Bhbj48L2E+PC9zcGFuPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0i
Zm9udC1zaXplOjExLjBwdCI+XQ0KPGI+T24gQmVoYWxmIE9mIDwvYj5EdXpvbmdwZW5nPGJyPg0K
PGI+U2VudDo8L2I+IDE2IE1hcmNoIDIwMTYgMDI6Mzg8YnI+DQo8Yj5Ubzo8L2I+IDwvc3Bhbj48
c3BhbiBsYW5nPSJFTi1HQiI+PGEgaHJlZj0ibWFpbHRvOmFuaW1hQGlldGYub3JnIiB0YXJnZXQ9
Il9ibGFuayI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0Ij5hbmlt
YUBpZXRmLm9yZzwvc3Bhbj48L2E+PC9zcGFuPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9u
dC1zaXplOjExLjBwdCI+PGJyPg0KPGI+Q2M6PC9iPiBMYXVyZW50IENpYXZhZ2xpYSAmbHQ7PC9z
cGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48YSBocmVmPSJtYWlsdG86bGF1cmVudC5jaWF2YWdsaWFA
bm9raWEuY29tIiB0YXJnZXQ9Il9ibGFuayI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250
LXNpemU6MTEuMHB0Ij5sYXVyZW50LmNpYXZhZ2xpYUBub2tpYS5jb208L3NwYW4+PC9hPjwvc3Bh
bj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQiPiZndDs7IErDqWZl
cnNvbg0KIENhbXBvcyBOb2JyZSAmbHQ7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48YSBocmVm
PSJtYWlsdG86amNub2JyZUBpbmYudWZyZ3MuYnIiIHRhcmdldD0iX2JsYW5rIj48c3BhbiBsYW5n
PSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQiPmpjbm9icmVAaW5mLnVmcmdzLmJyPC9z
cGFuPjwvYT48L3NwYW4+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0
Ij4mZ3Q7OyBTaGVuZyBKaWFuZyAmbHQ7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48YSBocmVm
PSJtYWlsdG86amlhbmdzaGVuZ0BodWF3ZWkuY29tIiB0YXJnZXQ9Il9ibGFuayI+PHNwYW4gbGFu
Zz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0Ij5qaWFuZ3NoZW5nQGh1YXdlaS5jb208
L3NwYW4+PC9hPjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4w
cHQiPiZndDs8YnI+DQo8Yj5TdWJqZWN0OjwvYj4gW0FuaW1hXSBEb2VzIEF1dG9ub21pYyBuZXR3
b3JrIG5lZWQgc29tZSBpbnRlcnZlbnRpb24gYmV5b25kIGludGVudDwvc3Bhbj48c3BhbiBsYW5n
PSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8L2Rpdj4NCjxwIGNsYXNz
PSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJv
dHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LXNpemU6MTAuNXB0
O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+
Jm5ic3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8
cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1h
cmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXpl
OjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYm
cXVvdDsiPkhlbGxvLCBldmVyeW9uZTwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286
cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9w
LWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1VUyIg
c3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90Oywm
cXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxv
OnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFy
Z2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0i
RU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkm
cXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i
c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7IEkgdW5kZXJzdGFuZCB0aGF0IEF1dG9ub21pYyBuZXR3b3Jr
IHNob3VsZCBiZSBhYmxlIHRvIHdvcmsgd2l0aCBhcyBsZXNzIGNvbmZpZ3VyYXRpb24gYXMNCiBw
b3NzaWJsZS4gSW4gdGhlIGlkZWFsIGNhc2UsIGFuIEF1dG9ub21pYyBuZXR3b3JrIHNob3VsZCBi
ZSBhYmxlIHRvIHdvcmsgd2VsbCB3aGlsZSBvbmx5IGludGVudCBpcyBuZWVkZWQgZnJvbSBodW1h
biBvcGVyYXRvcnMuPC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48
L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87
bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9u
dC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMt
c2VyaWYmcXVvdDsiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48
L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFs
dDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5
bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVv
dDtzYW5zLXNlcmlmJnF1b3Q7Ij4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsm
bmJzcDsmbmJzcDsgSG93ZXZlciwgSSB3b25kZXIgd2hldGhlciBBdXRvbm9taWMgbmV0d29yayBt
YXkgbmVlZCBzb21lIGludGVydmVudGlvbiBmcm9tIGh1bWFuIG9wZXJhdG9ycw0KIHRocm91Z2gg
c29tZSBsb3ctbGV2ZWwgcG9saWNpZXMgYmV5b25kIGludGVudHMuPC9zcGFuPjxzcGFuIGxhbmc9
IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHls
ZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxz
cGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVv
dDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPiZuYnNwOzwvc3Bhbj48c3Bh
biBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1h
bCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDph
dXRvIj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1p
bHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij4mbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsgRm9yIGV4YW1wbGUsIGluIHRo
ZQ0KPC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48YSBocmVmPSJodHRwczovL3Rvb2xzLmlldGYu
b3JnL2h0bWwvZHJhZnQtaWV0Zi1hbmltYS1wcmVmaXgtbWFuYWdlbWVudC0wMCIgdGFyZ2V0PSJf
YmxhbmsiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZh
bWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPmh0dHBzOi8v
dG9vbHMuaWV0Zi5vcmcvaHRtbC9kcmFmdC1pZXRmLWFuaW1hLXByZWZpeC1tYW5hZ2VtZW50LTAw
PC9zcGFuPjwvYT48L3NwYW4+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAu
NXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90
OyI+LA0KIGl0IGlzIHN1Z2dlc3RlZCB0aGF0IHRoZSBwcmVmaXggbGVuZ3RocyBmb3IgdGhlIENT
RywgQVNHLCBSU0cgKGRpZmZlcmVudCByb2xlcyBpbiBJUCBSQU4pIGNhbiBiZSBhc3NpZ25lZCBh
cyBhbiAmcXVvdDtpbnRlbnQmcXVvdDsuIFRoZXNlIGNvbmZpZ3VyYXRpb24gcGFyYW1ldGVycyBu
ZWVkIHRvIGJlIGRpc3RyaWJ1dGVkIGluIHRoZSBhdXRvbm9taWMgZG9tYWluIHRvIGluZmx1ZW5j
ZSB0aGUgZGV0YWlsIGNvbmZpZ3VyYXRpb25zIG9uIGVhY2ggYXV0b25vbWljDQogbm9kZS48L3Nw
YW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJN
c29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRv
bS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAuNXB0O2Zv
bnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+Jm5i
c3A7PC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBj
bGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdp
bi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjEw
LjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVv
dDsiPiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyBBcyBp
bnRlbnQgaXMgZGVzY3JpYmVkIGFzIGFuIGFic3RyYWN0LCBkZWNsYXJhdGl2ZSwgaGlnaC1sZXZl
bCBwb2xpY3kgdXNlZCB0byBvcGVyYXRlIGFuDQogYXV0b25vbWljIGRvbWFpbiwgc3VjaCBhcyBh
biBlbnRlcnByaXNlIG5ldHdvcmsuIFBlcmhhcHMgd2UgY2FuIHJlbmFtZSB0aGUg4oCcaW50ZW50
4oCdIGluIHRoZSDigJxwcmVmaXgtbWFuYWdlbWVudCBJROKAnSB0byBzb21lIG90aGVyIHdvcmRz
Ljwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xh
c3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4t
Ym90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMC41
cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7
Ij4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4N
CjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28t
bWFyZ2luLWJvdHRvbS1hbHQ6YXV0bzt0ZXh0LWluZGVudDoyMS4wcHQiPg0KPHNwYW4gbGFuZz0i
RU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkm
cXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+SXMg4oCcQVNBIHBhcmFtZXRlcnMmcXVvdDsg
b2sgaGVyZT8NCjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9w
Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21z
by1tYXJnaW4tYm90dG9tLWFsdDphdXRvO3RleHQtaW5kZW50OjIxLjBwdCI+DQo8c3BhbiBsYW5n
PSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJy
aSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij4mbmJzcDs8L3NwYW4+PHNwYW4gbGFuZz0i
RU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxl
PSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0bzt0ZXh0
LWluZGVudDoyMS4wcHQiPg0KPHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAu
NXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90
OyI+SSB0aGluayBzb21lIGNoYXJhY3RlcmlzdGljcyBmb3IgdGhpcyBraW5kIG9mIGxvdy1sZXZl
bCBwb2xpY2llcyBhcmU8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFu
PjwvcD4NCjxwIHN0eWxlPSJtYXJnaW4tbGVmdDozOS4wcHQiPjxzcGFuIGxhbmc9IkVOLVVTIiBz
dHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZx
dW90O3NhbnMtc2VyaWYmcXVvdDsiPjEuPC9zcGFuPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0i
Zm9udC1zaXplOjcuMHB0Ij4mbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsNCjwv
c3Bhbj48c3BhbiBsYW5nPSJFTi1VUyI+TmV0d29yayBsZXZlbCBwYXJhbWV0ZXJzLCBuZWVkIHRv
IGJlIGRpc3RyaWJ1dGVkIGJ5IEdSQVNQIGFuZCBBQ1A8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0Ii
PjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIHN0eWxlPSJtYXJnaW4tbGVmdDozOS4wcHQiPjxz
cGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVv
dDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPjIuPC9zcGFuPjxzcGFuIGxh
bmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjcuMHB0Ij4mbmJzcDsmbmJzcDsmbmJzcDsmbmJz
cDsmbmJzcDsmbmJzcDsNCjwvc3Bhbj48c3BhbiBsYW5nPSJFTi1VUyI+UmVsYXRpdmVseSBzdGF0
aWMgY29tcGFyZWQgdG8gaW50ZW50LCBwZXJoYXBzIG9ubHkgbmVlZGVkIHRvIGJlIGNvbmZpZ3Vy
ZWQgb25jZTwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0K
PHAgc3R5bGU9Im1hcmdpbi1sZWZ0OjM5LjBwdCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJm
b250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fu
cy1zZXJpZiZxdW90OyI+My48L3NwYW4+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNp
emU6Ny4wcHQiPiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOw0KPC9zcGFuPjxz
cGFuIGxhbmc9IkVOLVVTIj5VbmF2b2lkYWJsZSBpbiBjdXJyZW50IG5ldHdvcmsgZW52aXJvbm1l
bnQsIG1vc3RseSBmb3IgZXN0YWJsaXNoaW5nIHRoZSBuZXR3b3JrIGluZnJhc3RydWN0dXJlPC9z
cGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBzdHlsZT0i
bWFyZ2luLWxlZnQ6MzkuMHB0Ij48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZTox
MC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1
b3Q7Ij40Ljwvc3Bhbj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZTo3LjBwdCI+
Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7DQo8L3NwYW4+PHNwYW4gbGFuZz0i
RU4tVVMiPlJhcmVseSBuZWVkIGNvb3JkaW5hdGlvbnMgd2l0aCBvdGhlcnMgcGFyYW1ldGVycywg
dXN1YWxseSBpcyBjbG9zZWQgcmVsYXRlZCB0byBvbmUgc3BlY2lmaWMgYXV0b25vbWljIGZ1bmN0
aW9uPC9zcGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBj
bGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdp
bi1ib3R0b20tYWx0OmF1dG8iPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjEw
LjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVv
dDsiPiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOw0KPC9z
cGFuPjxzcGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0i
TXNvTm9ybWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0
b20tYWx0OmF1dG87dGV4dC1pbmRlbnQ6MjEuMHB0Ij4NCjxzcGFuIGxhbmc9IkVOLVVTIiBzdHls
ZT0iZm9udC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90
O3NhbnMtc2VyaWYmcXVvdDsiPldlbGNvbWUgZm9yIGNvbW1lbnQuIFlvdXIgc3VnZ2VzdGlvbnMg
d2lsbCBiZSBhcHByZWNpYXRlZC48L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+
PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1h
bHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0bzt0ZXh0LWluZGVudDoyMS4wcHQiPg0K
PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZx
dW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+Jm5ic3A7PC9zcGFuPjxz
cGFuIGxhbmc9IkVOLUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9y
bWFsIiBzdHlsZT0ibXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0
OmF1dG87dGV4dC1pbmRlbnQ6MjEuMHB0Ij4NCjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9u
dC1zaXplOjEwLjVwdDtmb250LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMt
c2VyaWYmcXVvdDsiPiZuYnNwOzwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286cD48
L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFs
dDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDphdXRvIj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5
bGU9ImZvbnQtc2l6ZToxMC41cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVv
dDtzYW5zLXNlcmlmJnF1b3Q7Ij5CZXN0IFJlZ2FyZHM8L3NwYW4+PHNwYW4gbGFuZz0iRU4tR0Ii
PjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28t
bWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2luLWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFu
Zz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTAuNXB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGli
cmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+Wm9uZ3BlbmcgRHU8L3NwYW4+PHNwYW4g
bGFuZz0iRU4tR0IiPjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rp
dj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bztt
YXJnaW4tYm90dG9tOjEyLjBwdCI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWls
eTomcXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPjxicj4NCl9f
X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fPGJyPg0KQW5pbWEg
bWFpbGluZyBsaXN0PGJyPg0KPGEgaHJlZj0ibWFpbHRvOkFuaW1hQGlldGYub3JnIiB0YXJnZXQ9
Il9ibGFuayI+QW5pbWFAaWV0Zi5vcmc8L2E+PGJyPg0KPGEgaHJlZj0iaHR0cHM6Ly93d3cuaWV0
Zi5vcmcvbWFpbG1hbi9saXN0aW5mby9hbmltYSIgdGFyZ2V0PSJfYmxhbmsiPmh0dHBzOi8vd3d3
LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vYW5pbWE8L2E+PC9zcGFuPjxzcGFuIGxhbmc9IkVO
LUdCIj48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8L2Jsb2NrcXVvdGU+DQo8L2Rpdj4NCjxwIGNs
YXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJtc28tbWFyZ2luLXRvcC1hbHQ6YXV0bzttc28tbWFyZ2lu
LWJvdHRvbS1hbHQ6YXV0byI+PHNwYW4gbGFuZz0iRU4tR0IiIHN0eWxlPSJmb250LWZhbWlseTom
cXVvdDtUaW1lcyBOZXcgUm9tYW4mcXVvdDssJnF1b3Q7c2VyaWYmcXVvdDsiPjxicj4NCjxiciBj
bGVhcj0iYWxsIj4NCjxicj4NCi0tIDwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+PG86cD48L286
cD48L3NwYW4+PC9wPg0KPGRpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0i
bXNvLW1hcmdpbi10b3AtYWx0OmF1dG87bXNvLW1hcmdpbi1ib3R0b20tYWx0OmF1dG8iPjxzcGFu
IGxhbmc9IkVOLUdCIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7VGltZXMgTmV3IFJvbWFuJnF1
b3Q7LCZxdW90O3NlcmlmJnF1b3Q7Ij5yZWdhcmRzLDwvc3Bhbj48c3BhbiBsYW5nPSJFTi1HQiI+
PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1h
bCIgc3R5bGU9Im1zby1tYXJnaW4tdG9wLWFsdDphdXRvO21zby1tYXJnaW4tYm90dG9tLWFsdDph
dXRvIj48c3BhbiBsYW5nPSJFTi1HQiIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O1RpbWVzIE5l
dyBSb21hbiZxdW90OywmcXVvdDtzZXJpZiZxdW90OyI+Sm9objwvc3Bhbj48c3BhbiBsYW5nPSJF
Ti1HQiI+PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9k
aXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIiBzdHlsZT0i
dGV4dC1hbGlnbjpqdXN0aWZ5O3RleHQtanVzdGlmeTppbnRlci1pZGVvZ3JhcGgiPjxzcGFuIGxh
bmc9IkVOLVVTIj48YnI+DQo8YnIgY2xlYXI9ImFsbCI+DQo8YnI+DQotLSA8bzpwPjwvbzpwPjwv
c3Bhbj48L3A+DQo8ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiIHN0eWxlPSJ0ZXh0
LWFsaWduOmp1c3RpZnk7dGV4dC1qdXN0aWZ5OmludGVyLWlkZW9ncmFwaCI+PHNwYW4gbGFuZz0i
RU4tVVMiPnJlZ2FyZHMsPG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAg
Y2xhc3M9Ik1zb05vcm1hbCIgc3R5bGU9InRleHQtYWxpZ246anVzdGlmeTt0ZXh0LWp1c3RpZnk6
aW50ZXItaWRlb2dyYXBoIj48c3BhbiBsYW5nPSJFTi1VUyI+Sm9objxvOnA+PC9vOnA+PC9zcGFu
PjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9ib2R5Pg0KPC9odG1sPg0K

--_000_BAFEC9523F57BC48A51C20226A5589575FDFCA51nkgeml514mbxchi_--


From nobody Fri Mar 25 03:24:45 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8C4B212D11B for ; Fri, 25 Mar 2016 03:24:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.621
X-Spam-Level: 
X-Spam-Status: No, score=-2.621 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1OAmlYGKXg5T for ; Fri, 25 Mar 2016 03:24:41 -0700 (PDT)
Received: from lb2-smtp-cloud3.xs4all.net (lb2-smtp-cloud3.xs4all.net [194.109.24.26]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D47F212D534 for ; Fri, 25 Mar 2016 03:24:40 -0700 (PDT)
Received: from webmail.xs4all.nl ([194.109.20.203]) by smtp-cloud3.xs4all.net with ESMTP id aAQe1s00x4NtgTm01AQeho; Fri, 25 Mar 2016 11:24:38 +0100
Received: from 2001:983:a264:1:41f7:64f7:e281:ad9c by webmail.xs4all.nl with HTTP (HTTP/1.1 POST); Fri, 25 Mar 2016 11:24:38 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 7bit
Date: Fri, 25 Mar 2016 11:24:38 +0100
From: peter van der Stok 
To: Laurent Ciavaglia 
Organization: vanderstok consultancy
Mail-Reply-To: consultancy@vanderstok.org
In-Reply-To: <56F45A42.4090207@alcatel-lucent.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com> <56F409D0.9030900@joelhalpern.com> <56F41351.8080309@alcatel-lucent.com> <56F4437C.7000803@gmail.com> <56F45A42.4090207@alcatel-lucent.com>
Message-ID: <10a6361c10281b6314bdd0d7bc710ab1@xs4all.nl>
X-Sender: stokcons@xs4all.nl (wtyqunkCsBTtWOf75YkVvKtC5qeHsuBm)
User-Agent: XS4ALL Webmail
Archived-At: 
Cc: Duzongpeng , "Michael Behringer \(mbehring\)" , "EXT Joel M. Halpern" , anima@ietf.org, Sheng Jiang 
Subject: Re: [Anima] Re-flooding [was: Does Autonomic network need some intervention beyond intent]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
Reply-To: consultancy@vanderstok.org
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 10:24:43 -0000

I would not worry too much about for example sleepy nodes.
Flooding is OK as long it is possible to to limit the distribution 
(based on MC address)
For example, Sleepy nodes will have private (probably standardized) 
mechanisms to obtain the info when needed from its parent.

Peter


>> Actually the same applies to sleeping nodes: at some frequency
>> they should wake up and synchronize intent, or any other flooded
>> objective that they care about. This is already specified in GRASP:
>> "Sleeping nodes that require an objective subject to flooding
>> SHOULD
>> periodically request unicast synchronization for that objective."
>> 
>> That doesn't mean than periodic re-flooding is forbidden, but the
>> frequency
>> can be kept quite low.
> 
> Agreed.


From nobody Fri Mar 25 04:43:39 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1FA4D12D65B for ; Fri, 25 Mar 2016 04:43:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.23
X-Spam-Level: 
X-Spam-Status: No, score=-4.23 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HWjKAno8GdgM for ; Fri, 25 Mar 2016 04:43:32 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D27E412D54F for ; Fri, 25 Mar 2016 04:43:30 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml703-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGO49490; Fri, 25 Mar 2016 11:43:27 +0000 (GMT)
Received: from NKGEML413-HUB.china.huawei.com (10.98.56.74) by lhreml703-cah.china.huawei.com (10.201.5.104) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 25 Mar 2016 11:43:22 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by NKGEML413-HUB.china.huawei.com ([10.98.56.74]) with mapi id 14.03.0235.001; Fri, 25 Mar 2016 19:43:11 +0800
From: Duzongpeng 
To: John Strassner , "Alexander Clemm (alex)" 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwATdGeIAAHjH6AAAOqg2AABRir4AAT8o7UA==
Date: Fri, 25 Mar 2016 11:43:10 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> 
In-Reply-To: 
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.149.226]
Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFCAAAnkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A020201.56F52460.0263, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 75564e734a57cfeb0e99f48b48eaa534
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , "Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 11:43:37 -0000

--_000_BAFEC9523F57BC48A51C20226A5589575FDFCAAAnkgeml514mbxchi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFCAAAnkgeml514mbxchi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFCAAAnkgeml514mbxchi_--


From nobody Fri Mar 25 06:40:05 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BF52512D69E for ; Fri, 25 Mar 2016 06:40:04 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ov5dOHvY6Sm2 for ; Fri, 25 Mar 2016 06:40:03 -0700 (PDT)
Received: from rcdn-iport-5.cisco.com (rcdn-iport-5.cisco.com [173.37.86.76]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2EC8012D1EC for ; Fri, 25 Mar 2016 06:40:03 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1708; q=dns/txt; s=iport; t=1458913203; x=1460122803; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=bBLpoYQQC/5sLdhwmUecZyj7niUsmcM7iWvSN7lgA8Q=; b=OENXaTNFMMd4DEmed71bY1hhWSc8DZ38P12CFGaL3rzpX9g+757GqiuC YEpIgFQ+S3jBUCxHMnNywICTEHtEjlvpTiPo3DoTwCgF87YABi1GWUkm2 T5j4yRlQVnkLQpbkpBVUPTRXCSXazrigi9G/+cv9t5UE10Y6kfj3uhtM4 g=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0BGBQAPP/VW/5FdJa1dgy6BUAavCItcg?= =?us-ascii?q?XCGDQIcgRQ5EwEBAQEBAQFkJ4RBAQEBBCMRRQwEAgEIEQQBAQECAiMDAgICHxE?= =?us-ascii?q?UAQgIAgQBDQUIiAoDErEmi1INhHcBAQEBAQEBAQEBAQEBAQEBAQEBAQEVfIUih?= =?us-ascii?q?ESCPoFfAQEzgmqCVgWSYYROMQGMDoFugW2NJYYPgSaHVAEiAT+DZWwBh2g2fgE?= =?us-ascii?q?BAQ?=
X-IronPort-AV: E=Sophos;i="5.24,391,1454976000"; d="scan'208";a="86740646"
Received: from rcdn-core-9.cisco.com ([173.37.93.145]) by rcdn-iport-5.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 25 Mar 2016 13:40:02 +0000
Received: from XCH-RCD-010.cisco.com (xch-rcd-010.cisco.com [173.37.102.20]) by rcdn-core-9.cisco.com (8.14.5/8.14.5) with ESMTP id u2PDe2YT010478 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 25 Mar 2016 13:40:02 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-010.cisco.com (173.37.102.20) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 25 Mar 2016 08:40:01 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 25 Mar 2016 08:40:01 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , John Strassner , Duzongpeng 
Thread-Topic: Another use case for Intent: "Freeze network enrolment"
Thread-Index: AQHRhgPpzIwvl+WeaEyrb+KXVUZG1J9qK3OA
Date: Fri, 25 Mar 2016 13:40:01 +0000
Message-ID: 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>
In-Reply-To: <56F440D1.3060702@gmail.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 13:40:04 -0000
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From nobody Fri Mar 25 07:17:15 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BFD6712D667 for ; Fri, 25 Mar 2016 07:17:13 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.4
X-Spam-Level: 
X-Spam-Status: No, score=-2.4 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FREEMAIL_FORGED_FROMDOMAIN=0.199, FREEMAIL_FROM=0.001, HEADER_FROM_DIFFERENT_DOMAINS=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H2=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id X4i5DDDbbD7y for ; Fri, 25 Mar 2016 07:17:11 -0700 (PDT)
Received: from mail-io0-f181.google.com (mail-io0-f181.google.com [209.85.223.181]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 74F6512D1D3 for ; Fri, 25 Mar 2016 07:17:05 -0700 (PDT)
Received: by mail-io0-f181.google.com with SMTP id c63so118001074iof.0 for ; Fri, 25 Mar 2016 07:17:05 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to; bh=tIXhVlwBuUqOVzdOIp2GMGy6suU+BRBvMRnOekv5ZXo=; b=JFsSFyN3LWJsamnIjlYc/dIjZKO6TSet3nRPXaWaFUqCD+2a75nJE2vFLbocrVPzmm dgHH6HAwK2dBpc13FdPpJb0wrm8imb2+9gqdnQgdYoBn5UpBwowAm92LqoO8U4LqVk2E eMNshcEj5VpnVVEtDIGDcDkK6X+dZpnfBoMuIHyzuNvwHOnq6HmHyCEQ6vSaXw36gtDP 7wERxOdfNfuhNGXZ/eCwH980y1oUHNbMK9pNCjOFtNMMVonl79shZnyIyelDWtA2TVIg BJNsYxQPSZr3vSYgBXjBmCz3+gq1S/szB9espiHeqejWRzEkpMPzNToUS5SAT8eUvXhb Rbuw==
X-Gm-Message-State: AD7BkJI164pogCRLSOzRExoFShGoS8ZNhp9yXELeMRO1p4wL2i5rEDJbXrsKoG8Fhy2A0WmzGbLyRQa/YOY6YA==
X-Received: by 10.107.26.201 with SMTP id a192mr15037336ioa.13.1458915424196;  Fri, 25 Mar 2016 07:17:04 -0700 (PDT)
MIME-Version: 1.0
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com> <56F4718D.8010104@alcatel-lucent.com>
In-Reply-To: <56F4718D.8010104@alcatel-lucent.com>
From: =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= 
Date: Fri, 25 Mar 2016 14:16:54 +0000
Message-ID: 
To: Laurent Ciavaglia ,  "EXT Michael Behringer (mbehring)" , "anima@ietf.org" 
Content-Type: multipart/alternative; boundary=001a113ff0bc9e0844052ee037c3
Archived-At: 
Subject: Re: [Anima] Reference Model Issues
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 14:17:14 -0000

--001a113ff0bc9e0844052ee037c3
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi.
+ 1 for the grouping.
Michael, are you thinking in maintaining subsections in the future work
section?
Best.
J=C3=A9ferson

Em qui, 24 de mar de 2016 =C3=A0s 20:00, Laurent Ciavaglia <
laurent.ciavaglia@nokia.com> escreveu:

> Dear Michael, all,
>
> Thanks for your continuous care of the document and taking some hindsight=
!
>
> I am also more in favor of grouping non-chartered aspects into a future
> work section, and allow the text there to be meaningful for 1) the
> relationship with (chartered) ANIMA components, 2) the overall picture, a=
nd
> 3) describing the implications of these future items with enough depth.
>
> I think the reference document is a pivot one in ANIMA to prepare what
> will come next (in ANIMA or elsewhere) and improve our collective
> understanding on future areas to investigate.
> I think it would be a mistake to neglect this opportunity by restricting
> the document on chartered items.
>
Thanks, Laurent.
> ps: of course my above support has nothing to do with the fact that I'm
> too lazy busy to condense the section I am editing ;)
>
> On 24/03/2016 18:52, EXT Michael Behringer (mbehring) wrote:
>
> I just read through the entire document again, noting down things that ne=
ed fixing, or discussion. The below is my personal view, for discussion. I =
cleaned up the github repository, and created a new file, corresponding to =
the latest version plus some small edits I made today while reading through=
 again.
>
> One important point: Our chairs want us to take out or reduce drastically=
 all the stuff that doesn't belong into the current charter. I agree that i=
t is not always 100% clear if something is out of charter. But I'm having a=
 hard time reducing some sections. What if we make a new section "Future wo=
rk" (which clearly indicates out of scope for this charter), where we colle=
ct those points, but then are a bit more liberal about the volume of each s=
ection? I think that would be a good compromise. Sheng?
>
> I'll summarise the below points in the WG meeting in BA. Maybe we can alr=
eady clarify some stuff before.
>
> Michael
>
>
>
>
> Reference model issues
> 24 Mar 2016
>
> 4.1 Naming:
> - "It is recommended that the names are generated by the autonomic nodes
>    themselves." - how? Should names / addresses not come from the registr=
ar?
>    Suggestion: Name *and* ACP address should be assigned by the registrar
>    at enrolment time. zone-id is dynamic, rest is fixed.
>
> - "a specific naming convention is out of scope":
>    Suggestion: We define a default naming and addressing scheme.
>    1st we settle on an addressing scheme from the ACP draft.
>    That defines a "device-ID" (last n bits of ACP address).
>    convert that to a string, and use it as a name. This would result in
>    something like: 0123-4567-89ab-0001.example.com.
>    The name would go into the domain certificate --> need to put this int=
o
>    the bootstrap draft as well.
>
> - to write in the doc: "registrar picks a naming scheme; all registrars
>   in a domain must use the same scheme. Example is: ...
>
> 4.4 ACP
> - should move to the end of section 4, it breaks the flow where it is rig=
ht now.
>
> 4.7 Intent Distribution
> - This section should only explain the basics, real work is out of scope =
for this phase.
> - should point to the intent distribution draft
> - should explain that intent is flooded to all nodes in a domain (if we a=
gree),
>   or other methods (if not).
> - should explain that we expect Intent to have a long life time (months),=
 thus
>   Intent distribution is expected to be very infrequent.
> - should explain that the entire Intent file is flooded in one go (if we =
agree).
>
> 5. Functional Overview
> - title is not good. This section really describes how an autonomic node =
behaves. maybe call it "Behaviour of an autonomic node"?
> - we need to describe the bring-up better, specifically insecure discover=
y,
>   ACP negotiation, ACP bringup, and subsequent operations, and which prot=
ocols to use
>   where. Or should this go into the ACP draft??
>
> 6. Security and Trust Infrastructure
> - needs a review from someone in the security space.
> - we need to describe the certificate format; this should likely go into =
bootstrap draft. This draft should explain where the certificate format is =
described.
> - the domain certificate should also contain the ACP IP address (where th=
e zone bits are set to zero).
>
> 7.1
> - broken paragraph; needs fixing.
> - there is a mention of constrained devices - take out.
>
> 7.2 Specific ASAs
> - should explain MUST implement
> - generally, need to work through text in this section; relate back to se=
ction 5.
>
> 8.1
> - remove last reference to RFC7575.
>
> 8.2 Intent
> - explain not part of current charter first.
> - not sure whether we should keep this level of detail, or shorten furthe=
r?
>
> 8.3 Aggregated Reporting
> - explain not part of current charter first.
> - my feeling: Should be reduced to 1-2 paragraphs, only high level idea.
>
> General:
> - should we combine all items that are not part of the current charter in=
to a separate section "Future Topics" or something like that? Then all (*) =
topic would be nicely contained. Yes, I think I'd prefer that to taking the=
m out completely, or shortening them to next-to-nothing.
>
> 10. Security Considerations
> - needs to be worked over.
>
> General:
> - uppercase Intent everywhere in the doc.
>
> _______________________________________________
> Anima mailing listAnima@ietf.orghttps://www.ietf.org/mailman/listinfo/ani=
ma
>
>
> --
>
> Laurent Ciavaglia
>
> Senior Research Manager
>
> Bell Labs, Nokia
>
>
>
> +33 160 402 636
>
> Route de Villejust | 91620 Nozay | France
>
> LinkedIn: laurentciavaglia 
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>

--001a113ff0bc9e0844052ee037c3
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi.
+ 1 for the =
grouping.
Michael, are you thinking in maintainin=
g subsections in the future work section?
Best.
J=
=C3=A9ferson

Em qui, 24 de=
 mar de 2016 =C3=A0s 20:00, Laurent Ciavaglia <laurent.ciavaglia@nokia.com> escreveu:

 =20
   =20
 =20
  

    Dear Michael, all,

      

      Thanks for your continuous care of the document and taking some
      hindsight!

      

      I am also more in favor of grouping non-chartered aspects into a
      future work section, and allow the text there to be meaningful for
      1) the relationship with (chartered) ANIMA components, 2) the
      overall picture, and 3) describing the implications of these
      future items with enough depth.

      

      I think the reference document is a pivot one in ANIMA to prepare
      what will come next (in ANIMA or elsewhere) and improve our
      collective understanding on future areas to investigate.

      I think it would be a mistake to neglect this opportunity by
      restricting the document on chartered items.

      Thanks, Laurent.

      ps: of course my above support has nothing to do with the fact
      that I'm too lazy busy to condense the section I
      am editing ;)

    


    
On 24/03/2016 18:52, EXT Michael
      Behringer (mbehring) wrote:

    

    

      
I just read through the entire document again, noting down thing=
s that need fixing, or discussion. The below is my personal view, for discu=
ssion. I cleaned up the github repository, and created a new file, correspo=
nding to the latest version plus some small edits I made today while readin=
g through again.=20

One important point: Our chairs want us to take out or reduce drastically a=
ll the stuff that doesn't belong into the current charter. I agree that=
 it is not always 100% clear if something is out of charter. But I'm ha=
ving a hard time reducing some sections. What if we make a new section &quo=
t;Future work" (which clearly indicates out of scope for this charter)=
, where we collect those points, but then are a bit more liberal about the =
volume of each section? I think that would be a good compromise. Sheng?=20

I'll summarise the below points in the WG meeting in BA. Maybe we can a=
lready clarify some stuff before.=20

Michael




Reference model issues
24 Mar 2016

4.1 Naming:=20
- "It is recommended that the names are generated by the autonomic nod=
es
   themselves." - how? Should names / addresses not come from the regi=
strar?=20
   Suggestion: Name *and* ACP address should be assigned by the registrar
   at enrolment time. zone-id is dynamic, rest is fixed.=20

- "a specific naming convention is out of scope":=20
   Suggestion: We define a default naming and addressing scheme.=20
   1st we settle on an addressing scheme from the ACP draft.=20
   That defines a "device-ID" (last n bits of ACP address).=20
   convert that to a string, and use it as a name. This would result in
   something like: 0123-4567-89ab-0001.example.com.=20
   The name would go into the domain certificate --> need to put this in=
to=20
   the bootstrap draft as well.=20
  =20
- to write in the doc: "registrar picks a naming scheme; all registrar=
s=20
  in a domain must use the same scheme. Example is: ...=20
  =20
4.4 ACP
- should move to the end of section 4, it breaks the flow where it is right=
 now.=20

4.7 Intent Distribution
- This section should only explain the basics, real work is out of scope fo=
r this phase.
- should point to the intent distribution draft
- should explain that intent is flooded to all nodes in a domain (if we agr=
ee),=20
  or other methods (if not).=20
- should explain that we expect Intent to have a long life time (months), t=
hus
  Intent distribution is expected to be very infrequent.=20
- should explain that the entire Intent file is flooded in one go (if we ag=
ree).=20

5. Functional Overview
- title is not good. This section really describes how an autonomic node be=
haves. maybe call it "Behaviour of an autonomic node"?=20
- we need to describe the bring-up better, specifically insecure discovery,=
=20
  ACP negotiation, ACP bringup, and subsequent operations, and which protoc=
ols to use
  where. Or should this go into the ACP draft??=20

6. Security and Trust Infrastructure
- needs a review from someone in the security space.
- we need to describe the certificate format; this should likely go into bo=
otstrap draft. This draft should explain where the certificate format is de=
scribed.=20
- the domain certificate should also contain the ACP IP address (where the =
zone bits are set to zero).=20

7.1=20
- broken paragraph; needs fixing.=20
- there is a mention of constrained devices - take out.=20

7.2 Specific ASAs
- should explain MUST implement
- generally, need to work through text in this section; relate back to sect=
ion 5.=20

8.1=20
- remove last reference to RFC7575.=20

8.2 Intent
- explain not part of current charter first.=20
- not sure whether we should keep this level of detail, or shorten further?=
=20

8.3 Aggregated Reporting
- explain not part of current charter first.=20
- my feeling: Should be reduced to 1-2 paragraphs, only high level idea.

General:=20
- should we combine all items that are not part of the current charter into=
 a separate section "Future Topics" or something like that? Then =
all (*) topic would be nicely contained. Yes, I think I'd prefer that t=
o taking them out completely, or shortening them to next-to-nothing.

10. Security Considerations
- needs to be worked over.=20

General:=20
- uppercase Intent everywhere in the doc.=20

_______________________________________________
Anima mailing list
Anima@ietf.org
h=
ttps://www.ietf.org/mailman/listinfo/anima



    

    

    
-- 

     =20
     =20
     =20
     =20
     =20
     =20
     =20
     =20
     =20
     =20
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
=C2=A0
        
+33=C2=A0160 402=C2=A0636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: <=
span lang=3D"EN-GB">laurentciavaglia

        
=C2=A0

      

    

  


_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima




--001a113ff0bc9e0844052ee037c3--


From nobody Fri Mar 25 07:23:26 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3DD4112DA88 for ; Fri, 25 Mar 2016 07:23:25 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.53
X-Spam-Level: 
X-Spam-Status: No, score=-14.53 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oeOuqn9DASRS for ; Fri, 25 Mar 2016 07:23:23 -0700 (PDT)
Received: from rcdn-iport-6.cisco.com (rcdn-iport-6.cisco.com [173.37.86.77]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 868DC12D673 for ; Fri, 25 Mar 2016 07:23:22 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=53716; q=dns/txt; s=iport; t=1458915802; x=1460125402; h=from:to:subject:date:message-id:references:in-reply-to: mime-version; bh=dOLskKphpYiEZ/bppx0w+cPyMH15WoQiIxdqLH8a690=; b=LdG4z+0exEFi8SZx2Qi3EvbJyGA00pK8bd6hafWx/grdlVrTZr6LVJ4f /tsVIG8SLqRH2m3bVDQTw6rveULGOrAKhte7BRZwkTBE4PvSg182AqXOY j8cqcICO0zLccVH9bSHp/E/CGy7HzMrUW9hR+o9Cj4leuG4JYnck9VWsQ Q=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0DnAgCBSfVW/4sNJK1dgmJMU30GqjWQI?= =?us-ascii?q?QENgW0DFwEJhWwCHIEUOBQBAQEBAQEBZCeEQQEBAQQBAQEgBAZBBBcCAQgRBAE?= =?us-ascii?q?BIQEGAwICAiULFAkIAQEEARIIEYgODrEgkFIBAQEBAQEBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEVhh6DRX+EUhcJgkqCVgWNPIVKhFoBhXCIDIFth3SFMY8JAQ8PAQFCg2VsiB9?= =?us-ascii?q?+AQEB?=
X-IronPort-AV: E=Sophos; i="5.24,391,1454976000"; d="scan'208,217"; a="86598667"
Received: from alln-core-6.cisco.com ([173.36.13.139]) by rcdn-iport-6.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 25 Mar 2016 14:23:21 +0000
Received: from XCH-ALN-007.cisco.com (xch-aln-007.cisco.com [173.36.7.17]) by alln-core-6.cisco.com (8.14.5/8.14.5) with ESMTP id u2PENLvM025387 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 25 Mar 2016 14:23:21 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-007.cisco.com (173.36.7.17) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 25 Mar 2016 09:23:20 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Fri, 25 Mar 2016 09:23:20 -0500
From: "Michael Behringer (mbehring)" 
To: =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , Laurent Ciavaglia , "anima@ietf.org" 
Thread-Topic: [Anima] Reference Model Issues
Thread-Index: AdGF9G4U4OT93HH3SYixSiiMjjRm3AAVnB6AACABZwAACmaHwA==
Date: Fri, 25 Mar 2016 14:23:20 +0000
Message-ID: 
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com> <56F4718D.8010104@alcatel-lucent.com> 
In-Reply-To: 
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: multipart/alternative; boundary="_000_ac2fbeb4120e4b5ab31579c5ff56b7b5XCHRCD006ciscocom_"
MIME-Version: 1.0
Archived-At: 
Subject: Re: [Anima] Reference Model Issues
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 14:23:26 -0000

--_000_ac2fbeb4120e4b5ab31579c5ff56b7b5XCHRCD006ciscocom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_ac2fbeb4120e4b5ab31579c5ff56b7b5XCHRCD006ciscocom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_ac2fbeb4120e4b5ab31579c5ff56b7b5XCHRCD006ciscocom_--


From nobody Fri Mar 25 11:18:34 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 65D9712D0CF for ; Fri, 25 Mar 2016 11:18:33 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.53
X-Spam-Level: 
X-Spam-Status: No, score=-14.53 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OL6faw9g8UR9 for ; Fri, 25 Mar 2016 11:18:29 -0700 (PDT)
Received: from rcdn-iport-2.cisco.com (rcdn-iport-2.cisco.com [173.37.86.73]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2834A12D0DC for ; Fri, 25 Mar 2016 11:18:29 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=128272; q=dns/txt; s=iport; t=1458929908; x=1460139508; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=UlHvaDBoQpuJ35JFz4iss17oPquEpPIk+HBBpXUOPE4=; b=jwZM7E8BLaFkI85RALQSxUihEWNin11eMDPBObwNuD/vumdY0DpaLPMK IvNxMvodKAXCxEHFLoy11ttnGlr3B/XFTtLs52rnKG0pFFaYmgfXYDs8X 37oa5DbdpURJPPw6PhEERTW5mv0gm5XN+XgKlDyeNB2JUVVmpZYbV4NEP I=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AfAgC2f/VW/5FdJa1egmJMU30GrwiLT?= =?us-ascii?q?gENgW0DFwEJhWwCHIEVOBQBAQEBAQEBZCeEQQEBAQQBAQEXAQgKQQsQAgEGAhE?= =?us-ascii?q?EAQEhAQYDAgICHwYLFAkIAgQBDQUIE4d3AxIOk2qdF4tBDYRuAQEBAQEBAQEBA?= =?us-ascii?q?QEBAQEBAQEBAQEBEQSKYoI+ghQXCYJKglYFjTyFSoQpMQGFcIYegW6BbYRMiFm?= =?us-ascii?q?HNYdUAR4BAUKDZWyHYiUYfgEBAQ?=
X-IronPort-AV: E=Sophos; i="5.24,392,1454976000"; d="scan'208,217"; a="89539282"
Received: from rcdn-core-9.cisco.com ([173.37.93.145]) by rcdn-iport-2.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 25 Mar 2016 18:18:27 +0000
Received: from XCH-RTP-008.cisco.com (xch-rtp-008.cisco.com [64.101.220.148]) by rcdn-core-9.cisco.com (8.14.5/8.14.5) with ESMTP id u2PIIQFR027551 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Fri, 25 Mar 2016 18:18:27 GMT
Received: from xch-rtp-001.cisco.com (64.101.220.141) by XCH-RTP-008.cisco.com (64.101.220.148) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Fri, 25 Mar 2016 14:18:26 -0400
Received: from xch-rtp-001.cisco.com ([64.101.220.141]) by XCH-RTP-001.cisco.com ([64.101.220.141]) with mapi id 15.00.1104.009; Fri, 25 Mar 2016 14:18:26 -0400
From: "Alexander Clemm (alex)" 
To: Duzongpeng , John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwAVBrxYAAHjH7AAAF5yGgAB0lm4AAQyTvAAAFQ5LQ
Date: Fri, 25 Mar 2016 18:18:25 +0000
Message-ID: <1dd14a9abe20419592e877235979585e@XCH-RTP-001.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>  
In-Reply-To: 
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.154.160.185]
Content-Type: multipart/alternative; boundary="_000_1dd14a9abe20419592e877235979585eXCHRTP001ciscocom_"
MIME-Version: 1.0
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , "Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 18:18:33 -0000

--_000_1dd14a9abe20419592e877235979585eXCHRTP001ciscocom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64

SGkgWm9uZ3BlbmcsDQoNCmxvb2tpbmcgYXQgeW91ciBjb21tZW50cywgSSBjb21wbGV0ZWx5IGFn
cmVlLiAgSSB0aGluayB3ZeKAmXJlIG9uIHRoZSBzYW1lIHBhZ2UuDQoNClJlZ2FyZHMNCi0tLSBB
bGV4DQoNCkZyb206IER1em9uZ3BlbmcgW21haWx0bzpkdXpvbmdwZW5nQGh1YXdlaS5jb21dDQpT
ZW50OiBGcmlkYXksIE1hcmNoIDI1LCAyMDE2IDQ6NDMgQU0NClRvOiBKb2huIFN0cmFzc25lciA8
c3RyYXpwZGpAZ21haWwuY29tPjsgQWxleGFuZGVyIENsZW1tIChhbGV4KSA8YWxleEBjaXNjby5j
b20+DQpDYzogTWljaGFlbCBCZWhyaW5nZXIgKG1iZWhyaW5nKSA8bWJlaHJpbmdAY2lzY28uY29t
PjsgTGF1cmVudCBDaWF2YWdsaWEgPGxhdXJlbnQuY2lhdmFnbGlhQG5va2lhLmNvbT47IFNoZW5n
IEppYW5nIDxqaWFuZ3NoZW5nQGh1YXdlaS5jb20+OyBhbmltYUBpZXRmLm9yZzsgSsOpZmVyc29u
IENhbXBvcyBOb2JyZSA8amNub2JyZUBpbmYudWZyZ3MuYnI+DQpTdWJqZWN0OiBSRTogW0FuaW1h
XSBEb2VzIEF1dG9ub21pYyBuZXR3b3JrIG5lZWQgc29tZSBpbnRlcnZlbnRpb24gYmV5b25kIGlu
dGVudA0KDQpIaSwgSm9obiwgQWxleDoNCg0KICAgICAgICAgU29tZSBwZXJzb25hbCB1bmRlcnN0
YW5kaW5nIGJlbG93LiBQbGVhc2Ugc2VlIGlubGluZS4NCg0KQmVzdCByZWdhcmRzDQpab25ncGVu
ZyBEdQ0KDQpGcm9tOiBKb2huIFN0cmFzc25lciBbbWFpbHRvOnN0cmF6cGRqQGdtYWlsLmNvbV0N
ClNlbnQ6IFRodXJzZGF5LCBNYXJjaCAyNCwgMjAxNiAxMTo0MSBBTQ0KVG86IEFsZXhhbmRlciBD
bGVtbSAoYWxleCk7IEpvaG4gU3RyYXNzbmVyDQpDYzogTWljaGFlbCBCZWhyaW5nZXIgKG1iZWhy
aW5nKTsgRHV6b25ncGVuZzsgTGF1cmVudCBDaWF2YWdsaWE7IFNoZW5nIEppYW5nOyBhbmltYUBp
ZXRmLm9yZzxtYWlsdG86YW5pbWFAaWV0Zi5vcmc+OyBKw6lmZXJzb24gQ2FtcG9zIE5vYnJlDQpT
dWJqZWN0OiBSZTogW0FuaW1hXSBEb2VzIEF1dG9ub21pYyBuZXR3b3JrIG5lZWQgc29tZSBpbnRl
cnZlbnRpb24gYmV5b25kIGludGVudA0KDQpIaSBBbGV4LA0KDQpJJ20gbm90IHN1cmUgSSBmb2xs
b3cgYSBjb3VwbGUgb2YgeW91ciBwb2ludHMsIHBsZWFzZSBzZWUgaW5saW5lLi4udGhvdWdoIGlu
IHJldHJvc3BlY3QsIHBlcmhhcHMgdGhpcyBpcyBtZXJlbHkgYSBjYXNlIG9mIHdobyB0aGUgYWN0
b3IgaXMgKGluIG15IGNhc2UsIEkgYW0gd29ycmllZCBhYm91dCB0aGUgd3JpdGVyIGFuZCB1c2Vy
IG9mIGludGVudCwgbm90IGhvdyB3ZSBidWlsZCBpdCkuDQoNCg0KSSByZWFsbHkgbGlrZSB5b3Vy
IHN1Z2dlc3Rpb24gb2YgZmxvb2RpbmcgaW50ZW50IHRvIGFsbCBub2RlcyAoZXZlbiBpZiB0aGVy
ZSBtYXkgYmUgb3B0aW1pemF0aW9ucyBsYXRlcikuICBJdCBwcm92aWRlcyBhIGxldmVsIG9mIGNv
bmNlcHR1YWwgY2xhcml0eS4gIEV2ZXJ5IG5vZGUgc2hvdWxkIGJlIGF3YXJlIG9mIG92ZXJhbGwg
aW50ZW50IChhbmQgbmVlZCB0byBkZWNpZGUgZm9yIGl0c2VsZiBpZiB0aGVyZSBhcmUgYWN0aW9u
cyBmb3IgaXQgdG8gdGFrZSkuDQo8amNzPg0KV2h5IHNob3VsZCAqKmV2ZXJ5Kiogbm9kZSBiZSBh
d2FyZSBvZiBhbGwgaW50ZW50PyBNYW55IHRpbWVzLCB0aGlzIHdvbid0IGJlIGFwcGxpY2FibGUu
DQoNClRoZSBvbmx5IHJlYXNvbiBJIGNhbiBzZWUgZm9yIGRvaW5nIHRoaXMgaXMgdG8gZW5hYmxl
IGNvbGxhYm9yYXRpb24gYW5kL29yIGVtZXJnZW50IGJlaGF2aW9yIChpbiB0aGUgQUkgc2Vuc2Up
LiBIb3dldmVyLCBpZiB5b3Ugd2FudCBlbWVyZ2VudCBiZWhhdmlvciBmcm9tIGF1dG9ub21pYyBu
b2RlcywgdGhlbiBlYWNoIG5vZGUgdGhhdCBpcyBwYXJ0IG9mIHRoYXQgY29sbGFib3JhdGlvbiBt
dXN0IGludGVyYWN0LiBJIGRvbid0IHRoaW5rIHdlIGtub3cgZW5vdWdoIGFib3V0IHRoaXMgdG8g
c3VwcG9ydCBvciBwcm9oaWJpdCBzdWNoIGJlaGF2aW9yLCBvdGhlciB0aGFuIHRoaXMgcGxhY2Vz
IGFuIGV4dHJhIHJlcXVpcmVtZW50IG9uIHRoZSBjb21wdXRhdGlvbmFsIGFuZCByZWFzb25pbmcg
YWJpbGl0aWVzIG9mIHRoZSBub2RlLg0KDQpJZiB5b3UgaGF2ZSBhIGRpZmZlcmVudCBwdXJwb3Nl
LCBwbGVhc2UgZG8gZXhwbGFpbi4NCjwvamNzPg0KW2R1em9uZ3BlbmddDQpJIGFncmVlIHRoYXQg
d2UgY2FuIGhhdmUgZXZlcnkgbm9kZSBpbiB0aGUgYXV0b25vbWljIGRvbWFpbiBoYXZlIHRoZSBz
YW1lIGludGVudC4gV2l0aGluIG15IHVuZGVyc3RhbmRpbmcsIGl0IGNhbiBiZSBhIGNhbmRpZGF0
ZSBzb2x1dGlvbi4NCg0KVGhlIGFkdmFudGFnZSBvZiB0aGUgc29sdXRpb24gaXMgc2ltcGxlLiBC
eSB1c2luZyBzb21lIHN5bmNocm9uaXphdGlvbiBtZWNoYW5pc20sIGl0IG1ha2VzIGFsbCB0aGUg
bm9kZSBpbiB0aGUgYXV0b25vbWljIGRvbWFpbiBjYW4gd29yayBjb29wZXJhdGl2ZWx5Lg0KDQpB
bm90aGVyIGNvbnNpZGVyYXRpb24gaXMgdGhhdCB3aGVuIGEgbmV3IG5vZGUgam9pbiBpbiB0aGUg
YXV0b25vbWljIGRvbWFpbiwgcGVyaGFwcyBpdCBjYW4gcmVjZWl2ZSB0aGUgaW50ZW50IGZyb20g
aXRzIG5laWdoYm9yLg0KDQpJIG1lYW5zIEFkbWluaXN0cmF0aXZlIEludGVudCBoZXJlLCBub3Qg
YSBzZXJ2aWNlIGludGVudC4gKFNlY3Rpb24gOCBvZiBodHRwczovL3Rvb2xzLmlldGYub3JnL2h0
bWwvZHJhZnQtZHUtYW5pbWEtYW4taW50ZW50LTAzKQ0KDQpQZXJoYXBzIHRoZSBsYXRlciBvbmx5
IG5lZWRzIHRvIGJlIGRpc3RyaWJ1dGVkIHRvIHRoZSByZWxhdGVkIG5vZGUsIG5vdCBhbGwgdGhl
IG5vZGUuDQoNClRoZSBzZXJ2aWNlIGludGVudCBtYXkgZXZlbiBiZSByZWxhdGVkIHRvIG11bHRp
cGxlIGF1dG9ub21pYyBkb21haW5zLg0KWy9kdXpvbmdwZW5nXQ0KDQpJZiB0aGVyZSBpcyBzb21l
IGludGVudCB0aGF0IGFwcGxpZXMgb25seSB0byBzb21lIHBhcnRzIG9mIHRoZSBuZXR3b3JrIGFu
ZCBub3Qgb3RoZXJzLCB0aGF0IGNhbiBiZSBhZGRyZXNzZWQgYnkgdXNpbmcgZGlmZmVyZW50IEF1
dG9ub21pYyBEb21haW5zLiAgQW5kLCBvZiBjb3Vyc2UsIEF1dG9ub21pYyBEb21haW5zIG1heSBi
ZSBhc3NvY2lhdGVkIHdpdGggY2VydGFpbiBBdXRvbm9taWMgRnVuY3Rpb25zLiAgSW50ZXJ2ZW50
aW9uIGJleW9uZCBpbnRlbnQgc2hvdWxkIE5PVCBiZSBhbGxvd2VkIGZvciB0aGUgQXV0b25vbWlj
IEZ1bmN0aW9uKHMpIGJlaW5nIHNlcnZlZCBhbiBBdXRvbm9taWMgRG9tYWluLCBidXQgb2YgY291
cnNlIG90aGVyIGludGVydmVudGlvbnMgKGZvciBkaWZmZXJlbnQgZnVuY3Rpb25zLCBoZW5jZSBv
dXRzaWRlIHRoZSBzY29wZSkgYXJlIHN0aWxsIHBvc3NpYmxlLg0KPGpjcz4NCkkgdGhpbmsgdGhh
dCBBdXRvbm9taWMgRG9tYWlucyBzaG91bGQgKG11c3Q/KSBiZSBoaWRkZW4gZnJvbSB0aGUgdXNl
ci9pbnRlbnQgd3JpdGVyLiBUaGV5IGFyZSBjbGVhcmx5IG9mIGh1Z2UgaW50ZXJuYWwgYmVuZWZp
dCwgYnV0IHdoeSBzaG91bGQgdGhlIHVzZXIgb2YgaW50ZW50IGtub3cgYWJvdXQgdGhlbT8gSW4g
YWRkaXRpb24sIEkgdGhpbmsgdGhhdCB3ZSBzaG91bGQgdXNlIHJvbGVzIGFzIGFuIGluZGlyZWN0
aW9uIG1lY2hhbmlzbSBmb3Igc2NhbGFiaWxpdHkgd2hlbiB3b3JraW5nIHdpdGggQXV0b25vbWlj
IERvbWFpbnMuDQo8L2pjcz4NCltkdXpvbmdwZW5nXQ0KV2l0aGluIG15IHVuZGVyc3RhbmRpbmcs
IGFuIGF1dG9ub21pYyBkb21haW4gd2lsbCBoYXZlIGFuIGludGVudCwgYW5kIGl0IGNhbiBvbmx5
IHRha2UgZWZmZWN0IHdpdGhpbiB0aGlzIGRvbWFpbi4NCg0KV2hlbiBhIHVzZXIgaXNzdWVzIGFu
IGludGVudCwgc2hlL2hlIHNob3VsZCBrbm93IHRoZSBkb21haW4gbmFtZS4gSSBtZWFuIGFuIGlu
dGVudCBjYW4gY29udGFpbiB0aGUgZG9tYWluIG5hbWUgYXMgb25lIG9mIGl0cyBlbGVtZW50Lg0K
DQpJIGFncmVlIHRoYXQgdXNpbmcgcm9sZXMgaXMgZ29vZCBmb3Igc2NhbGFiaWxpdHkuDQpbL2R1
em9uZ3BlbmddDQoNCg0KcmVnYXJkcywNCkpvaG4NCg0KT24gV2VkLCBNYXIgMjMsIDIwMTYgYXQg
MTA6NTYgQU0sIEFsZXhhbmRlciBDbGVtbSAoYWxleCkgPGFsZXhAY2lzY28uY29tPG1haWx0bzph
bGV4QGNpc2NvLmNvbT4+IHdyb3RlOg0KSSByZWFsbHkgbGlrZSB5b3VyIHN1Z2dlc3Rpb24gb2Yg
Zmxvb2RpbmcgaW50ZW50IHRvIGFsbCBub2RlcyAoZXZlbiBpZiB0aGVyZSBtYXkgYmUgb3B0aW1p
emF0aW9ucyBsYXRlcikuICBJdCBwcm92aWRlcyBhIGxldmVsIG9mIGNvbmNlcHR1YWwgY2xhcml0
eS4gIEV2ZXJ5IG5vZGUgc2hvdWxkIGJlIGF3YXJlIG9mIG92ZXJhbGwgaW50ZW50IChhbmQgbmVl
ZCB0byBkZWNpZGUgZm9yIGl0c2VsZiBpZiB0aGVyZSBhcmUgYWN0aW9ucyBmb3IgaXQgdG8gdGFr
ZSkuDQoNCklmIHRoZXJlIGlzIHNvbWUgaW50ZW50IHRoYXQgYXBwbGllcyBvbmx5IHRvIHNvbWUg
cGFydHMgb2YgdGhlIG5ldHdvcmsgYW5kIG5vdCBvdGhlcnMsIHRoYXQgY2FuIGJlIGFkZHJlc3Nl
ZCBieSB1c2luZyBkaWZmZXJlbnQgQXV0b25vbWljIERvbWFpbnMuICBBbmQsIG9mIGNvdXJzZSwg
QXV0b25vbWljIERvbWFpbnMgbWF5IGJlIGFzc29jaWF0ZWQgd2l0aCBjZXJ0YWluIEF1dG9ub21p
YyBGdW5jdGlvbnMuICBJbnRlcnZlbnRpb24gYmV5b25kIGludGVudCBzaG91bGQgTk9UIGJlIGFs
bG93ZWQgZm9yIHRoZSBBdXRvbm9taWMgRnVuY3Rpb24ocykgYmVpbmcgc2VydmVkIGFuIEF1dG9u
b21pYyBEb21haW4sIGJ1dCBvZiBjb3Vyc2Ugb3RoZXIgaW50ZXJ2ZW50aW9ucyAoZm9yIGRpZmZl
cmVudCBmdW5jdGlvbnMsIGhlbmNlIG91dHNpZGUgdGhlIHNjb3BlKSBhcmUgc3RpbGwgcG9zc2li
bGUuDQoNCi0tLSBBbGV4DQoNCkZyb206IEFuaW1hIFttYWlsdG86YW5pbWEtYm91bmNlc0BpZXRm
Lm9yZzxtYWlsdG86YW5pbWEtYm91bmNlc0BpZXRmLm9yZz5dIE9uIEJlaGFsZiBPZiBNaWNoYWVs
IEJlaHJpbmdlciAobWJlaHJpbmcpDQpTZW50OiBXZWRuZXNkYXksIE1hcmNoIDIzLCAyMDE2IDM6
NTcgQU0NClRvOiBKb2huIFN0cmFzc25lciA8c3RyYXpwZGpAZ21haWwuY29tPG1haWx0bzpzdHJh
enBkakBnbWFpbC5jb20+Pg0KQ2M6IER1em9uZ3BlbmcgPGR1em9uZ3BlbmdAaHVhd2VpLmNvbTxt
YWlsdG86ZHV6b25ncGVuZ0BodWF3ZWkuY29tPj47IExhdXJlbnQgQ2lhdmFnbGlhIDxsYXVyZW50
LmNpYXZhZ2xpYUBub2tpYS5jb208bWFpbHRvOmxhdXJlbnQuY2lhdmFnbGlhQG5va2lhLmNvbT4+
OyBTaGVuZyBKaWFuZyA8amlhbmdzaGVuZ0BodWF3ZWkuY29tPG1haWx0bzpqaWFuZ3NoZW5nQGh1
YXdlaS5jb20+PjsgYW5pbWFAaWV0Zi5vcmc8bWFpbHRvOmFuaW1hQGlldGYub3JnPjsgSsOpZmVy
c29uIENhbXBvcyBOb2JyZSA8amNub2JyZUBpbmYudWZyZ3MuYnI8bWFpbHRvOmpjbm9icmVAaW5m
LnVmcmdzLmJyPj4NClN1YmplY3Q6IFJlOiBbQW5pbWFdIERvZXMgQXV0b25vbWljIG5ldHdvcmsg
bmVlZCBzb21lIGludGVydmVudGlvbiBiZXlvbmQgaW50ZW50DQoNCkpvaG4sDQoNCkkgdGhpbmsg
d2XigJlyZSBvbiB0aGUgc2FtZSBwYWdlLiBDb3VwbGUgb2YgcG9pbnRzLCB0b3AgcG9zdGluZyBo
ZXJlLCBvdGhlcndpc2UgaXQgZ2V0cyB0b28gbWVzc3kuDQoNCldoaWxlICpDSUVzIGRvbuKAmXQg
bGlrZSBtYWdpYywgdGhleSBkbyB1bmRlcnN0YW5kIHRoZSBjb25jZXB0IG9mIGEgZGVmYXVsdCBi
ZWhhdmlvdXIgd2hpY2ggdGhleSBjYW4gaW5mbHVlbmNlIGlmIG5lZWRlZC4gQW55IHJlYXNvbmFi
bGUgbmV0d29yayBtYW5hZ2VtZW50IGFwcHJvYWNoIHRyaWVzIHRvIGRlZmluZSDigJxkZWZhdWx0
4oCdIGJlaGF2aW91ciBhbmQg4oCcZXhjZXB0aW9uc+KAnS4gVGhlIEFOIHN0b3J5IGZpbGxzIHRo
ZSDigJxkZWZhdWx04oCdIHBhcnQgdmVyeSBuaWNlbHksIGFuZCAqQ0lFcyBnZXQgdGhhdC4NCg0K
VGhlIOKAnHBhcmFtZXRlcuKAnSB2ZXJzdXMg4oCccG9saWN54oCdIGRpc2N1c3Npb246IE15IHZp
ZXcgaXMgdGhhdCB3ZSBzaG91bGRu4oCZdCB0cnkgdG8gZGVmaW5lIGV4YWN0bHkgKndoYXQqIGlu
dGVudCBhbGxvd3MgYW5kIGRvZXNu4oCZdCBhbGxvdywgc2luY2Ugd2XigJl2ZSBoZWFyZCBmcm9t
IG1hbnkgc2lkZXMgdGhhdCB0aGUgYm91bmRhcnkgaXMgbm90IGNsZWFyLCBhbmQgaXQgbWF5IGJl
IGltcG9zc2libGUgdG8gaGF2ZSBhIHJlYWxseSBjbGVhciwgdXNhYmxlIGRlZmluaXRpb24uIFNv
IG15IHBvaW50IHdhczogTGV04oCZcyBmb2N1cyBvbiB0aGUgZGlzdHJpYnV0aW9uIG1vZGVsOiBJ
bnRlbnQgaXMgZmxvb2RlZCB0byBhbGwgbm9kZXMuIExldOKAmXMgTk9UIGRvIHNlbGVjdGl2ZSBm
bG9vZGluZywgdW5pY2FzdCBJbnRlbnQgdG8gc3BlY2lmaWMgbm9kZXMsIGV0YywgYmVjYXVzZSB0
aGF0IHdvdWxkIGJsdXIgdGhlIGJvdW5kYXJ5IGV2ZW4gd29yc2UuDQoNCk15IHN1Z2dlc3Rpb246
IEludGVudCBpcyBmbG9vZGVkIHRvIGFsbCBub2RlcyBpbiBhIGRvbWFpbi4gVGhlIEFTQSBvd25l
cnMgZGVmaW5lIHRoZSBjb250ZW50IG9mIHdoYXQgdGhleSB3YW50IHRvIHB1c2ggaW4gdGhlaXIg
Yml0IG9mIHRoZSBJbnRlbnQuIElmIHdlIGluc2lzdCBvZiBmdWxsIGZsb29kaW5nLCB3ZSBlbmNv
dXJhZ2UgdGhlIHJpZ2h0IGJlaGF2aW91ci4gQW5kIGlmIHNvbWVvbmUgcmVhbGx5IHB1dHMgaW50
byBoaXMgSW50ZW50IOKAnG5vZGUgQTogZG8gdGhpczsgbm9kZSBCOiBkbyB0aGUgb3RoZXI7IG5v
ZGUgQzogZG8gdGhlIHRoaXJk4oCdIGl0IGlzIGF3a3dhcmQgZW5vdWdoIGZvciBwZW9wbGUgdG8g
dGFrZSBhIHN0ZXAgYmFjayBhbmQgdGhpbmsgd2hldGhlciB0aGV54oCZcmUgZG9pbmcgdGhlIHJp
Z2h0IHRoaW5nIOKYug0KDQo8amNzPg0KRG8geW91IG1lYW4gdGhhdCB0aGUgaW50ZW50IHdyaXRl
ciB3aWxsICoqa25vdyB0aGUgZGlmZmVyZW5jZSoqDQpiZXR3ZWVuIHRoZSBmYWJyaWMgYW5kIGVh
Y2ggYXV0b25vbWljIGZ1bmN0aW9uPyBJIGhvcGUgbm90LA0KYnV0IHBlcmhhcHMgSSdtIG1pc3Vu
ZGVyc3RhbmRpbmcgeW91Li4uDQo8L2pjcz4NCg0KTm90IHN1cmUgSSB1bmRlcnN0YW5kIHlvdXIg
Y29uY2Vybi4gWWVzLCBpbiBteSBtb2RlbCB0aGVyZSBpcyBhIHNlY3Rpb24gZm9yIHRoZSBmYWJy
aWMsIGFuZCBzZWN0aW9ucyBmb3IgdGhlIHZhcmlvdXMgZnVuY3Rpb25zLiBTZWVtcyBhIG5hdHVy
YWwgc3RydWN0dXJlPyE/DQoNCuKAnEZ1bmN0aW9uIG93bmVy4oCdOiBZZXMsIEnigJltIHN1Z2dl
c3RpbmcgdG8gc3RydWN0dXJlIEludGVudCBhcm91bmQgQXV0b25vbWljIEZ1bmN0aW9ucy4gVGhp
cyBpcyBub3Qgb3B0aW1hbCBmcm9tIGEgc2NpZW50aWZpYyBwb2ludCBvZiB2aWV3LCBidXQgYWxp
Z25zIHZlcnkgd2VsbCB3aXRoIHRoZSB3YXkgaG93IHRoaW5ncyBhcmUgb3JnYW5pc2VkLiBQcm9i
YWJseSB3ZSBuZWVkIHRvIGRpc2N1c3MgdGhhdCBpbiBCQS4gQnV0IHJlZmVyIHRvIG15IFlBTkcg
bW9kZWwgc2VudCBhIGZldyBkYXlzIGFnby4NCg0KV2Ugc2hvdWxkIHNjaGVkdWxlIHNvbWUgZGlz
Y3Vzc2lvbiB0aW1lIG91dHNpZGUgdGhlIEFOSU1BIG1lZXRpbmcgd2luZG93Li4uDQoNCk1pY2hh
ZWwNCg0KDQpGcm9tOiBKb2huIFN0cmFzc25lciBbbWFpbHRvOnN0cmF6cGRqQGdtYWlsLmNvbV0N
ClNlbnQ6IDIyIE1hcmNoIDIwMTYgMjE6MzINClRvOiBNaWNoYWVsIEJlaHJpbmdlciAobWJlaHJp
bmcpIDxtYmVocmluZ0BjaXNjby5jb208bWFpbHRvOm1iZWhyaW5nQGNpc2NvLmNvbT4+OyBKb2hu
IFN0cmFzc25lciA8c3RyYXpwZGpAZ21haWwuY29tPG1haWx0bzpzdHJhenBkakBnbWFpbC5jb20+
Pg0KQ2M6IER1em9uZ3BlbmcgPGR1em9uZ3BlbmdAaHVhd2VpLmNvbTxtYWlsdG86ZHV6b25ncGVu
Z0BodWF3ZWkuY29tPj47IGFuaW1hQGlldGYub3JnPG1haWx0bzphbmltYUBpZXRmLm9yZz47IExh
dXJlbnQgQ2lhdmFnbGlhIDxsYXVyZW50LmNpYXZhZ2xpYUBub2tpYS5jb208bWFpbHRvOmxhdXJl
bnQuY2lhdmFnbGlhQG5va2lhLmNvbT4+OyBKw6lmZXJzb24gQ2FtcG9zIE5vYnJlIDxqY25vYnJl
QGluZi51ZnJncy5icjxtYWlsdG86amNub2JyZUBpbmYudWZyZ3MuYnI+PjsgU2hlbmcgSmlhbmcg
PGppYW5nc2hlbmdAaHVhd2VpLmNvbTxtYWlsdG86amlhbmdzaGVuZ0BodWF3ZWkuY29tPj4NClN1
YmplY3Q6IFJlOiBbQW5pbWFdIERvZXMgQXV0b25vbWljIG5ldHdvcmsgbmVlZCBzb21lIGludGVy
dmVudGlvbiBiZXlvbmQgaW50ZW50DQoNCkluIGdlbmVyYWwsIEkgYWdyZWUgd2l0aCBNaWNoYWVs
LiBJJ2QgbGlrZSB0byBlbXBoYXNpemUgYW5kIGV4cGFuZCBvbiBhIGNvdXBsZSBvZiBrZXkgcG9p
bnRzOg0KDQoNCkZpcnN0LCBBTklNQSBpc27igJl0IHRoZSBvbmx5IHRoaW5nIOKAnGluZmx1ZW5j
aW5n4oCdIG9yIOKAnHN0ZWVyaW5n4oCdIGEgbmV0d29yaywgYXQgbGVhc3Qgbm90IGluaXRpYWxs
eS4gVGhlcmUgaXMgc3RpbGwgdHJhZGl0aW9uYWwgY29uZmlnLCBwbHVzIGFsbCB0aGUgZXhpc3Rp
bmcgYW5kIGVtZXJnaW5nIFNETiBtb2RlbHMuIE15IHZpZXcgaGFzIGFsd2F5cyBiZWVuIHRoYXQg
QU5JTUEgc2V0cyBhIOKAnGRlZmF1bHTigJ0gcG9saWN5IGluIGEgbmV0d29yaywgYW5kIGFsbCBv
dGhlciBtZWFucyBhcmUgbW9yZSBzcGVjaWZpYyBhbmQgY2FuIG92ZXJyaWRlIGludGVudC4gVGhp
cyBpcyBkZXNjcmliZWQgaW4gUkZDNzU3NS4gTWFpbiBwb2ludCBoZXJlOiBBTklNQSBJbnRlbnQg
aXNu4oCZdCB0aGUgb25seSB0b29sIGluIGEgbmV0d29yay4NCg0KPGpjcz4NCkFic29sdXRlbHkg
YWdyZWUuIE5vdGUgdGhhdCBldmVuIGluIHRoZSBDb25ncmVzcyBjYXNlIEkgbWVudGlvbmVkDQpl
YXJsaWVyLCBpdCBpcyBOT1QgcHVzaGluZyBjb25maWcsIGFuZCBpdCBpcyBOT1Qgdmlld2VkIGFz
IHRoZQ0KIm9ubHkiIHBvbGljeQ0KPC9qY3M+DQoNCg0KSW50ZW50IGNvdWxkIHBlcmZlY3RseSB3
ZWxsIGNvdmVyIGEgaGlnaCBsZXZlbCBwb2xpY3kgc3VjaCBhcyDigJxhbGwgbm9kZXMgb2YgdHlw
ZSB4IGRvIHRoaXM7IHR5cGUgeSBkb2VzIHRoYXTigJ0uIChXaGljaCBJIHRoaW5rIGlzIHdoYXQg
eW91IG1lbnRpb24gYmVsb3cuKSBXaGVuIHdlIHN0YXJ0ZWQgd2l0aCBJbnRlbnQsIHRoZSB2aXNp
b24gd2FzIHRvIGtlZXAgaXQgdmVyeSBoaWdoIGxldmVsLCBtb3JlIGFsb25nIHRoZSBsaW5lIOKA
nGRvIHRoZSByaWdodCB0aGluZ+KAnSDimLoNCg0KPGpjcz4NCkFncmVlZC4gQWx0aG91Z2ggdG8g
bWUsIHRoaXMgbWVhbnMgdGhhdCB3ZSBuZWVkIHRvIGNvbnNpZGVyIHRoZSB0eXBlcyBvZiBhY3Rv
cnMgdGhhdCBhcmUgdXNpbmcgaW50ZW50LiBGb3IgZXhhbXBsZSwgbW9zdCBvZiBteSBDQ0lFIGFu
ZCBKQ0lFIGZyaWVuZHMgaGF2ZSBubyBpbnRlcmVzdCBpbiB1c2luZyBhbiBpbnRlbnQgbGFuZ3Vh
Z2UgZm9yIHB1c2hpbmcgY29uZmlndXJhdGlvbi4gT24gdGhlIG90aGVyIGhhbmQsIHdoZW4gSSB0
YWxrIHRvIG9wZXJhdG9ycyBhbmQgYXBwbGljYXRpb24gZGV2ZWxvcGVycywgdGhleSB3b3VsZCBs
b3ZlIHRvIGJlIGFibGUgdG8gdXNlIGEgaGlnaC1sZXZlbCBpbnRlbnQgYW5kIGhhdmUgdGhlIGlu
dGVudCBlbmdpbmUgdGFrZSBjYXJlIG9mIHRoZSBkZXRhaWxzIGZvciB0aGVtLg0KPC9qY3M+DQoN
Ckl0IGhhcyBiZWNvbWUgY2xlYXIgdGhhdCBpbiB0b2RheeKAmXMgbmV0d29ya2luZywgcGVvcGxl
IHdpbGwgd2FudCB0byBwdXNoIG1vcmUg4oCccGFyYW1ldGVyc+KAnSB0aGFuIOKAnHBvbGljeeKA
nSwgYW5kIEkgdGhpbmsgd2XigJlsbCBqdXN0IGhhdmUgdG8gYWNrbm93bGVkZ2UgdGhhdC4NCg0K
PGpjcz4NClNvcnJ5IE1pY2hhZWwsIEkgZG9uJ3QgZm9sbG93IGhlcmUuIFdoYXQgZG8geW91IG1l
YW4NCmJ5ICJwYXJhbWV0ZXJzIj8NCjwvamNzPg0KDQoNClBlcnNvbmFsbHkgSSB0aGluayBpdOKA
mWxsIHBhbiBvdXQgbGlrZSB0aGlzOiBJbnRlbnQgd2lsbCBoYXZlIHNlY3Rpb25zIHBlciBhdXRv
bm9taWMgZnVuY3Rpb24uIEFuIEludGVudCBwYXJzZXIgd2lsbCB2YWxpZGF0ZSBvdmVyYWxsIGNv
cnJlY3RuZXNzIChzaWduYXR1cmUsIGV0YyksIHRoZW4gc2VnbWVudCB0aGUgSW50ZW50IGludG8g
aXRzIGJpdHM6IE9uZSBwYXJ0IHdpbGwgYmUgdGhlIGF1dG9ub21pYyBmYWJyaWMgaXRzZWxmLCB0
aGVuIHRoZXJlIHdpbGwgYmUgYml0cyBmb3IgZWFjaCBhdXRvbm9taWMgZnVuY3Rpb24uDQoNCjxq
Y3M+DQpEbyB5b3UgbWVhbiB0aGF0IHRoZSBpbnRlbnQgd3JpdGVyIHdpbGwgKiprbm93IHRoZSBk
aWZmZXJlbmNlKioNCmJldHdlZW4gdGhlIGZhYnJpYyBhbmQgZWFjaCBhdXRvbm9taWMgZnVuY3Rp
b24/IEkgaG9wZSBub3QsDQpidXQgcGVyaGFwcyBJJ20gbWlzdW5kZXJzdGFuZGluZyB5b3UuLi4N
CjwvamNzPg0KDQoNClNvIHRoZSBJbnRlbnQgcGFyc2VyIG9uIGEgbm9kZSB3aWxsIGdyYWIgaW50
ZW50LCBzZWUgdGhlcmUgaXMgc29tZSBJbnRlbnQgZm9yIGF1dG9ub21pYyBmdW5jdGlvbiBBLCB3
aWxsIHNlZSB3aGV0aGVyIHRoaXMgaXMgcnVubmluZyBsb2NhbGx5LCBhbmQgaWYgc28sIGl04oCZ
bGwganVzdCBwYXNzIHRoYXQgcGllY2Ugb2YgSW50ZW50IHRvIHRoYXQgZnVuY3Rpb24uDQoNCjxq
Y3M+DQpBZ3JlZWQuDQo8L2pjcz4NCg0KDQpJbiB0aGlzIHdheSwgYSBmdW5jdGlvbiBvd25lciBk
ZWZpbmVzIGJvdGggdGhlIGludGVudCBpdHNlbGYsIGFuZCBob3cgaXQgaXMgY29uc3VtZWQgYnkg
aGlzL2hlciBmdW5jdGlvbi4NCg0KPGpjcz4NCkknbSBub3Qgc3VyZSB3aGF0IHlvdSBtZWFuIGJ5
ICJmdW5jdGlvbiBvd25lciIuIChTb3JyeSB0byBiZQ0KcGlja3ksIEknbSBsaWtlbHkgdW5kZXIt
Y2FmZmVpbmF0ZWQuKQ0KDQpUaGUgYWJvdmUgc2VlbWVkIGxpa2UgYW4gYXV0b25vbW91cyBkZWNp
c2lvbiwgbWFkZSBieSB0aGUgcGFyc2VyLg0KSSBkb24ndCB0aGluayB0aGF0IHRoZSBpbnRlbnQg
d3JpdGVyIHNob3VsZCBiZSBjb2duaXphbnQgb2YgaG93IG1hbnkgb3INCndoYXQgdHlwZXMgb2Yg
YXV0b25vbWljIGZ1bmN0aW9ucyBhcmUgcHJlc2VudC4NCjwvamNzPg0KDQoNClRoZSBkb3duc2lk
ZSBpcyB0aGF0IHRoZXJlIGlzIG5vIOKAnGNvbnRyb2zigJ0gYW5kIHBlb3BsZSBjYW4gZG8gd2hh
dCB0aGV5IHdhbnQsIGluIHRoZSB3b3JzdCBjYXNlIGV2ZW4gcHVzaCBjb25maWcsIHdoaWNoIGlz
IGNsZWFybHkgTk9UIHdoYXQgSW50ZW50IGlzIG1lYW50IHRvIGRvLg0KDQo8amNzPg0KQW5kIGhl
bmNlLCB0aGUgIkhBTCAyMDAwIiBzY2VuYXJpby4gOi0pDQo8L2pjcz4NCg0KDQpJ4oCZZCBob3Bl
IHRoYXQgcGVvcGxlIGRvbuKAmXQgc3RhcnQgd3JpdGluZyBJbnRlbnQgbGlrZSDigJxub2RlIDE3
OiBoZXJlIGlzIHlvdXIgQ0xJOyBub2RlIDIzOiBoZXJlIGlzIHlvdXIgQ0xJ4oCdOyBub2RlIDM3
OiBoZXJlIGlzIHlvdXIgQ0xJ4oCdLiBCdXQgd2hhdCBJIGRlc2NyaWJlIGFib3ZlIENPVUxEIGJl
IGFidXNlZCBpbiB0aGlzIHdheS4gUkZDNzU3NSBzYXlzIHRoaXMgaXMgTk9UIEludGVudC4NCg0K
PGpjcz4NCkFic29sdXRlbHkgYWdyZWUhDQo8L2pjcz4NCg0KIHJlZ2FyZHMsDQpKb2huDQoNCg0K
DQpPbiBXZWQsIE1hciAxNiwgMjAxNiBhdCAxOjE2IEFNLCBNaWNoYWVsIEJlaHJpbmdlciAobWJl
aHJpbmcpIDxtYmVocmluZ0BjaXNjby5jb208bWFpbHRvOm1iZWhyaW5nQGNpc2NvLmNvbT4+IHdy
b3RlOg0KSGkgWm9uZ3BlbmcsDQoNClllcywgdGhpcyBpcyBhIHJlY3VycmluZyBxdWVzdGlvbiBh
bmQgdG9waWMuIExldCBtZSBnaXZlIG15IHZpZXcgb24gdHdvIGxldmVsczoNCg0KRmlyc3QsIEFO
SU1BIGlzbuKAmXQgdGhlIG9ubHkgdGhpbmcg4oCcaW5mbHVlbmNpbmfigJ0gb3Ig4oCcc3RlZXJp
bmfigJ0gYSBuZXR3b3JrLCBhdCBsZWFzdCBub3QgaW5pdGlhbGx5LiBUaGVyZSBpcyBzdGlsbCB0
cmFkaXRpb25hbCBjb25maWcsIHBsdXMgYWxsIHRoZSBleGlzdGluZyBhbmQgZW1lcmdpbmcgU0RO
IG1vZGVscy4gTXkgdmlldyBoYXMgYWx3YXlzIGJlZW4gdGhhdCBBTklNQSBzZXRzIGEg4oCcZGVm
YXVsdOKAnSBwb2xpY3kgaW4gYSBuZXR3b3JrLCBhbmQgYWxsIG90aGVyIG1lYW5zIGFyZSBtb3Jl
IHNwZWNpZmljIGFuZCBjYW4gb3ZlcnJpZGUgaW50ZW50LiBUaGlzIGlzIGRlc2NyaWJlZCBpbiBS
RkM3NTc1LiBNYWluIHBvaW50IGhlcmU6IEFOSU1BIEludGVudCBpc27igJl0IHRoZSBvbmx5IHRv
b2wgaW4gYSBuZXR3b3JrLg0KDQpJbnRlbnQgY291bGQgcGVyZmVjdGx5IHdlbGwgY292ZXIgYSBo
aWdoIGxldmVsIHBvbGljeSBzdWNoIGFzIOKAnGFsbCBub2RlcyBvZiB0eXBlIHggZG8gdGhpczsg
dHlwZSB5IGRvZXMgdGhhdOKAnS4gKFdoaWNoIEkgdGhpbmsgaXMgd2hhdCB5b3UgbWVudGlvbiBi
ZWxvdy4pIFdoZW4gd2Ugc3RhcnRlZCB3aXRoIEludGVudCwgdGhlIHZpc2lvbiB3YXMgdG8ga2Vl
cCBpdCB2ZXJ5IGhpZ2ggbGV2ZWwsIG1vcmUgYWxvbmcgdGhlIGxpbmUg4oCcZG8gdGhlIHJpZ2h0
IHRoaW5n4oCdIOKYuiAgSXQgaGFzIGJlY29tZSBjbGVhciB0aGF0IGluIHRvZGF54oCZcyBuZXR3
b3JraW5nLCBwZW9wbGUgd2lsbCB3YW50IHRvIHB1c2ggbW9yZSDigJxwYXJhbWV0ZXJz4oCdIHRo
YW4g4oCccG9saWN54oCdLCBhbmQgSSB0aGluayB3ZeKAmWxsIGp1c3QgaGF2ZSB0byBhY2tub3ds
ZWRnZSB0aGF0Lg0KDQpQZXJzb25hbGx5IEkgdGhpbmsgaXTigJlsbCBwYW4gb3V0IGxpa2UgdGhp
czogSW50ZW50IHdpbGwgaGF2ZSBzZWN0aW9ucyBwZXIgYXV0b25vbWljIGZ1bmN0aW9uLiBBbiBJ
bnRlbnQgcGFyc2VyIHdpbGwgdmFsaWRhdGUgb3ZlcmFsbCBjb3JyZWN0bmVzcyAoc2lnbmF0dXJl
LCBldGMpLCB0aGVuIHNlZ21lbnQgdGhlIEludGVudCBpbnRvIGl0cyBiaXRzOiBPbmUgcGFydCB3
aWxsIGJlIHRoZSBhdXRvbm9taWMgZmFicmljIGl0c2VsZiwgdGhlbiB0aGVyZSB3aWxsIGJlIGJp
dHMgZm9yIGVhY2ggYXV0b25vbWljIGZ1bmN0aW9uLg0KDQpTbyB0aGUgSW50ZW50IHBhcnNlciBv
biBhIG5vZGUgd2lsbCBncmFiIGludGVudCwgc2VlIHRoZXJlIGlzIHNvbWUgSW50ZW50IGZvciBh
dXRvbm9taWMgZnVuY3Rpb24gQSwgd2lsbCBzZWUgd2hldGhlciB0aGlzIGlzIHJ1bm5pbmcgbG9j
YWxseSwgYW5kIGlmIHNvLCBpdOKAmWxsIGp1c3QgcGFzcyB0aGF0IHBpZWNlIG9mIEludGVudCB0
byB0aGF0IGZ1bmN0aW9uLg0KDQpJbiB0aGlzIHdheSwgYSBmdW5jdGlvbiBvd25lciBkZWZpbmVz
IGJvdGggdGhlIGludGVudCBpdHNlbGYsIGFuZCBob3cgaXQgaXMgY29uc3VtZWQgYnkgaGlzL2hl
ciBmdW5jdGlvbi4NCg0KVGhlIGRvd25zaWRlIGlzIHRoYXQgdGhlcmUgaXMgbm8g4oCcY29udHJv
bOKAnSBhbmQgcGVvcGxlIGNhbiBkbyB3aGF0IHRoZXkgd2FudCwgaW4gdGhlIHdvcnN0IGNhc2Ug
ZXZlbiBwdXNoIGNvbmZpZywgd2hpY2ggaXMgY2xlYXJseSBOT1Qgd2hhdCBJbnRlbnQgaXMgbWVh
bnQgdG8gZG8uIEnigJlkIGhvcGUgdGhhdCBwZW9wbGUgZG9u4oCZdCBzdGFydCB3cml0aW5nIElu
dGVudCBsaWtlIOKAnG5vZGUgMTc6IGhlcmUgaXMgeW91ciBDTEk7IG5vZGUgMjM6IGhlcmUgaXMg
eW91ciBDTEnigJ07IG5vZGUgMzc6IGhlcmUgaXMgeW91ciBDTEnigJ0uIEJ1dCB3aGF0IEkgZGVz
Y3JpYmUgYWJvdmUgQ09VTEQgYmUgYWJ1c2VkIGluIHRoaXMgd2F5LiBSRkM3NTc1IHNheXMgdGhp
cyBpcyBOT1QgSW50ZW50Lg0KDQpJbiBvdGhlciB3b3JkczogTGV04oCZcyBub3QgYmUgdG9vIHBy
ZXNjcmlwdGl2ZSwgYW5kIGhvcGUgdGhhdCBmb2xrcyB3aWxsIGRvIG1vcmUgb3IgbGVzcyByZWFz
b25hYmxlIHRoaW5ncy4NCg0KSeKAmXZlIGJlZW4gbWVhbmluZyB0byBjb21tZW50IG9uIHRoZSBJ
bnRlbnQgZHJhZnRzIGFuZCBwb3NzaWJseSBjb250cmlidXRlLCBJ4oCZbSBqdXN0IHN0cnVnZ2xp
bmcgd2l0aCBwcmlvcml0aWVzIGF0IHRoZSBtb21lbnQgKGFzIHlvdSBjYW4gc2VlIGJ5IG15IHJl
bGF0aXZlIHNpbGVuY2Ugb3ZlciB0aGUgcGFzdCB3ZWVrcykuDQoNCk1pY2hhZWwNCg0KDQoNCkZy
b206IEFuaW1hIFttYWlsdG86YW5pbWEtYm91bmNlc0BpZXRmLm9yZzxtYWlsdG86YW5pbWEtYm91
bmNlc0BpZXRmLm9yZz5dIE9uIEJlaGFsZiBPZiBEdXpvbmdwZW5nDQpTZW50OiAxNiBNYXJjaCAy
MDE2IDAyOjM4DQpUbzogYW5pbWFAaWV0Zi5vcmc8bWFpbHRvOmFuaW1hQGlldGYub3JnPg0KQ2M6
IExhdXJlbnQgQ2lhdmFnbGlhIDxsYXVyZW50LmNpYXZhZ2xpYUBub2tpYS5jb208bWFpbHRvOmxh
dXJlbnQuY2lhdmFnbGlhQG5va2lhLmNvbT4+OyBKw6lmZXJzb24gQ2FtcG9zIE5vYnJlIDxqY25v
YnJlQGluZi51ZnJncy5icjxtYWlsdG86amNub2JyZUBpbmYudWZyZ3MuYnI+PjsgU2hlbmcgSmlh
bmcgPGppYW5nc2hlbmdAaHVhd2VpLmNvbTxtYWlsdG86amlhbmdzaGVuZ0BodWF3ZWkuY29tPj4N
ClN1YmplY3Q6IFtBbmltYV0gRG9lcyBBdXRvbm9taWMgbmV0d29yayBuZWVkIHNvbWUgaW50ZXJ2
ZW50aW9uIGJleW9uZCBpbnRlbnQNCg0KSGVsbG8sIGV2ZXJ5b25lDQoNCiAgICAgICAgIEkgdW5k
ZXJzdGFuZCB0aGF0IEF1dG9ub21pYyBuZXR3b3JrIHNob3VsZCBiZSBhYmxlIHRvIHdvcmsgd2l0
aCBhcyBsZXNzIGNvbmZpZ3VyYXRpb24gYXMgcG9zc2libGUuIEluIHRoZSBpZGVhbCBjYXNlLCBh
biBBdXRvbm9taWMgbmV0d29yayBzaG91bGQgYmUgYWJsZSB0byB3b3JrIHdlbGwgd2hpbGUgb25s
eSBpbnRlbnQgaXMgbmVlZGVkIGZyb20gaHVtYW4gb3BlcmF0b3JzLg0KDQogICAgICAgICBIb3dl
dmVyLCBJIHdvbmRlciB3aGV0aGVyIEF1dG9ub21pYyBuZXR3b3JrIG1heSBuZWVkIHNvbWUgaW50
ZXJ2ZW50aW9uIGZyb20gaHVtYW4gb3BlcmF0b3JzIHRocm91Z2ggc29tZSBsb3ctbGV2ZWwgcG9s
aWNpZXMgYmV5b25kIGludGVudHMuDQoNCiAgICAgICAgIEZvciBleGFtcGxlLCBpbiB0aGUgaHR0
cHM6Ly90b29scy5pZXRmLm9yZy9odG1sL2RyYWZ0LWlldGYtYW5pbWEtcHJlZml4LW1hbmFnZW1l
bnQtMDAsIGl0IGlzIHN1Z2dlc3RlZCB0aGF0IHRoZSBwcmVmaXggbGVuZ3RocyBmb3IgdGhlIENT
RywgQVNHLCBSU0cgKGRpZmZlcmVudCByb2xlcyBpbiBJUCBSQU4pIGNhbiBiZSBhc3NpZ25lZCBh
cyBhbiAiaW50ZW50Ii4gVGhlc2UgY29uZmlndXJhdGlvbiBwYXJhbWV0ZXJzIG5lZWQgdG8gYmUg
ZGlzdHJpYnV0ZWQgaW4gdGhlIGF1dG9ub21pYyBkb21haW4gdG8gaW5mbHVlbmNlIHRoZSBkZXRh
aWwgY29uZmlndXJhdGlvbnMgb24gZWFjaCBhdXRvbm9taWMgbm9kZS4NCg0KICAgICAgICAgQXMg
aW50ZW50IGlzIGRlc2NyaWJlZCBhcyBhbiBhYnN0cmFjdCwgZGVjbGFyYXRpdmUsIGhpZ2gtbGV2
ZWwgcG9saWN5IHVzZWQgdG8gb3BlcmF0ZSBhbiBhdXRvbm9taWMgZG9tYWluLCBzdWNoIGFzIGFu
IGVudGVycHJpc2UgbmV0d29yay4gUGVyaGFwcyB3ZSBjYW4gcmVuYW1lIHRoZSDigJxpbnRlbnTi
gJ0gaW4gdGhlIOKAnHByZWZpeC1tYW5hZ2VtZW50IElE4oCdIHRvIHNvbWUgb3RoZXIgd29yZHMu
DQoNCklzIOKAnEFTQSBwYXJhbWV0ZXJzIiBvayBoZXJlPw0KDQpJIHRoaW5rIHNvbWUgY2hhcmFj
dGVyaXN0aWNzIGZvciB0aGlzIGtpbmQgb2YgbG93LWxldmVsIHBvbGljaWVzIGFyZQ0KDQoxLiAg
ICAgICBOZXR3b3JrIGxldmVsIHBhcmFtZXRlcnMsIG5lZWQgdG8gYmUgZGlzdHJpYnV0ZWQgYnkg
R1JBU1AgYW5kIEFDUA0KDQoyLiAgICAgICBSZWxhdGl2ZWx5IHN0YXRpYyBjb21wYXJlZCB0byBp
bnRlbnQsIHBlcmhhcHMgb25seSBuZWVkZWQgdG8gYmUgY29uZmlndXJlZCBvbmNlDQoNCjMuICAg
ICAgIFVuYXZvaWRhYmxlIGluIGN1cnJlbnQgbmV0d29yayBlbnZpcm9ubWVudCwgbW9zdGx5IGZv
ciBlc3RhYmxpc2hpbmcgdGhlIG5ldHdvcmsgaW5mcmFzdHJ1Y3R1cmUNCg0KNC4gICAgICAgUmFy
ZWx5IG5lZWQgY29vcmRpbmF0aW9ucyB3aXRoIG90aGVycyBwYXJhbWV0ZXJzLCB1c3VhbGx5IGlz
IGNsb3NlZCByZWxhdGVkIHRvIG9uZSBzcGVjaWZpYyBhdXRvbm9taWMgZnVuY3Rpb24NCg0KV2Vs
Y29tZSBmb3IgY29tbWVudC4gWW91ciBzdWdnZXN0aW9ucyB3aWxsIGJlIGFwcHJlY2lhdGVkLg0K
DQoNCkJlc3QgUmVnYXJkcw0KWm9uZ3BlbmcgRHUNCg0KX19fX19fX19fX19fX19fX19fX19fX19f
X19fX19fX19fX19fX19fX19fX19fX18NCkFuaW1hIG1haWxpbmcgbGlzdA0KQW5pbWFAaWV0Zi5v
cmc8bWFpbHRvOkFuaW1hQGlldGYub3JnPg0KaHR0cHM6Ly93d3cuaWV0Zi5vcmcvbWFpbG1hbi9s
aXN0aW5mby9hbmltYQ0KDQoNCg0KLS0NCnJlZ2FyZHMsDQpKb2huDQoNCg0KDQotLQ0KcmVnYXJk
cywNCkpvaG4NCg==

--_000_1dd14a9abe20419592e877235979585eXCHRTP001ciscocom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_1dd14a9abe20419592e877235979585eXCHRTP001ciscocom_--


From nobody Fri Mar 25 12:55:51 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B4D6A12D1A6 for ; Fri, 25 Mar 2016 12:55:49 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.911
X-Spam-Level: 
X-Spam-Status: No, score=-1.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id P0LhnGqf8iYL for ; Fri, 25 Mar 2016 12:55:47 -0700 (PDT)
Received: from tuna.sandelman.ca (tuna.sandelman.ca [209.87.249.19]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BA18812D1C0 for ; Fri, 25 Mar 2016 12:55:47 -0700 (PDT)
Received: from sandelman.ca (obiwan.sandelman.ca [IPv6:2607:f0b0:f:2::247]) by tuna.sandelman.ca (Postfix) with ESMTP id 95BEA203C8; Fri, 25 Mar 2016 15:58:37 -0400 (EDT)
Received: from obiwan.sandelman.ca (localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id 6921563755; Fri, 25 Mar 2016 15:55:46 -0400 (EDT)
From: Michael Richardson 
To: "anima\@ietf.org" 
In-Reply-To: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <56F28403.6060401@alcatel-lucent.com> 
X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2
X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m
Archived-At: 
Cc: John Strassner 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 19:55:50 -0000

--=-=-=
Content-Type: text/plain


John Strassner  wrote:
    >  Hopefully, intent will not change frequently (or else it probably
    > is too low level in nature). I personally like using pub-sub, but that
    > is because it enables me to do additional pre- and post-processing of
    > intent. However, I think that this should be an option.  < /jcs>

Why not flood the list of intents (by suitable-hash of intent file), and
along with an origin for the intent.  (this winds up looking a lot like
OSPF, and is what DNCP is good at...)

Then, any node which does not have a copy of that intent can ask for it using
any of our well known protocols (i.e. HTTP).

--
Michael Richardson , Sandelman Software Works
 -= IPv6 IoT consulting =-




--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBVvWXv4CLcPvd0N1lAQJLnAf8CNqPsnG1cPiWhSBf68Y9EmulvezICIMy
TMKJDl+XmD5zGcT/YoJMfs6084UCMgbSWRsQdhfjHsZ7imh/C+LxjCSDORaJCCJD
xOf1jA7//4ZI7hF8O9IjJDI2nXl3XMr8bRurmV4jsXbWiE7FFEYycsOkaCvQrjjp
tmatws1NQK9dSsCUo5Yx4Spyp5CwN/jX4pyj0JeSCNFFQA+JF04jRGlZbLrdQepj
3FsT6JsiZ56z/cWdx7Ah010LjsrqNSkTjAppe9xAk/KFEofPsy4m30t9in1pkU62
CKVC822qGkk0nFi/zvrQ/3NCXyR2XAWj+WeoLoEK+SOwFK6dPOUWpA==
=KT9u
-----END PGP SIGNATURE-----
--=-=-=--


From nobody Fri Mar 25 13:06:16 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7397E12D1C0 for ; Fri, 25 Mar 2016 13:06:14 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gJX0-1-tL-Fk for ; Fri, 25 Mar 2016 13:06:12 -0700 (PDT)
Received: from mail-pa0-x231.google.com (mail-pa0-x231.google.com [IPv6:2607:f8b0:400e:c03::231]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 781EF12D13C for ; Fri, 25 Mar 2016 13:06:12 -0700 (PDT)
Received: by mail-pa0-x231.google.com with SMTP id td3so52173974pab.2 for ; Fri, 25 Mar 2016 13:06:12 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=gqkzJT6XdomJi/4RRM3WjSvF7IB8VzeBty8C5StmKTE=; b=Jl0LEd+0YR7YZzUzKscp90i0lGLAQ46HP/NTWKXUdofuawHdSlVUEhaWaHnZhi3/GD cLCTP93o4fi1stNphjvW8nqxnO1u1gBozB+6/ZtsMveReis8SSO0/3Qloo9jZyT9JOMF WWS+Xu3TlpmrWp2RrZ5nKRNlIYrfLz/DgySOftTRJQaTei4XSeCl+G6LKQm3nVW3xcpq IyfaQBBSM1C2bgzjahyr8poc/O0fuAAxYaMU61Tah0F3uQfa76KPvelr8uAz1X+41mwM /ewxcSRJHqIagsYd3maUBgR4C4m4SrG70LzJL1eRErENYOhYEWKjJvAcVVSCQWN+zzcX 83Xw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=gqkzJT6XdomJi/4RRM3WjSvF7IB8VzeBty8C5StmKTE=; b=Y+n5DxyGQxC67Dsotlsg27LC1HuEUCFu4lCV/EcsBe0NJjUeuAFXqN5hjhgySGEnIG eVBiXD5kr2pawPXuD2I6ZUY/5M/XprBRad5m7G0Moj/7DyJUrSHPY2kZjY3PWIdyKM5A LTGeXgOayDk6ltUGt0KKv/MR5Q+EPMOoujG3KWXIb6Pa0sriemARi7aNm4fpqYXiEDKr zZa42YbIuc71b9tpuErYmUSdOlo9bDwkcNcEocEj6wKWvW0Kywmyc827IG04NAj9I/Z5 pcwgdXiEOctXkhY02o8C/nsPsP+hrNxqXvJNWOc1NEcThtEpITq7mNztdTxTByN+ngvd 4rIw==
X-Gm-Message-State: AD7BkJLeNFXoG+ohB9+3Jqk07OndAOz4fmnvYD3YfDrwmSFEIuXTJeHUDauWUnTCbbxqsg==
X-Received: by 10.66.147.74 with SMTP id ti10mr23981466pab.128.1458936372055;  Fri, 25 Mar 2016 13:06:12 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id fn3sm18241015pab.20.2016.03.25.13.06.07 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 13:06:10 -0700 (PDT)
To: "Michael Behringer (mbehring)" , Laurent Ciavaglia , "EXT Joel M. Halpern" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com> <56F409D0.9030900@joelhalpern.com> <56F41351.8080309@alcatel-lucent.com> <56F4437C.7000803@gmail.com> 
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F59A3A.3050807@gmail.com>
Date: Sat, 26 Mar 2016 09:06:18 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Duzongpeng , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Re-flooding [was: Does Autonomic network need some intervention beyond intent]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 20:06:14 -0000

Hi Michael,

On 25/03/2016 20:54, Michael Behringer (mbehring) wrote:
>> -----Original Message-----
>> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
>> Sent: 24 March 2016 20:44
>> To: Laurent Ciavaglia ; EXT Joel M. Halpern
>> ; Michael Behringer (mbehring)
>> 
>> Cc: Duzongpeng ; anima@ietf.org; Sheng Jiang
>> 
>> Subject: Re-flooding [was: Does Autonomic network need some intervention
>> beyond intent]
>>
>> On 25/03/2016 05:18, Laurent Ciavaglia wrote:
>> ...
>>> It is not because an intent policy is not changed, that recipients (e.g. new
>> ASA instances) should not be informed about the intent policies, and thus
>> generates a new broadcast.
>>
>> A newly started ASA can (and probably should) perform the following logic:
>>
>>  - discover the Intent server (just another ASA, of course)
>>  - request unicast synchronization of the latest intent
>>
>> Actually the same applies to sleeping nodes: at some frequency they should
>> wake up and synchronize intent, or any other flooded objective that they
>> care about. This is already specified in GRASP:
>>   "Sleeping nodes that require an objective subject to flooding SHOULD
>>    periodically request unicast synchronization for that objective."
> 
> Hmmm... To me, keeping Intent up to date is not a function of the ASAs at all. 

Well, that requirement is about a sleeping *node* which could contain any
number of ASAs.

> 
> In my mind, Intent management, synchronisation etc are functions of the AN Infrastructure. In the AN Infrastructure there is a functional block "Intent management". That keeps Intent in sync with neighbouring nodes. Up to this point the ASAs aren't even involved.

Yes, if we model Intent as a single data structure distributed to all nodes.
(I would actually model it as a generic ASA called "Intent" that handles a generic
objective called "Intent", because that way we can use bog-standard GRASP
to handle Intent.)

> 
> An ASA has an interface into the ANI, which allows it to request "its" Intent. Or, the ANI pushes the respective Intent to the ASA. 

Sure. But again we can model it in the same way, "Intent.PrefixManager" being the
Intent relevant to the "PrefixManager" objective.

> But my main point: ASAs shouldn't know about synchronisation or distribution. They just consume what's on the node (in the ANI). 

All ASAs know about synchronization because it's a generic function.
However, again we can use the same model. To be very concrete:

 ok, intent = grasp.synchronize(asa_nonce, grasp.objective("Intent.PrefixManager"), None, None)

That will fetch the requested intent from the local flood cache, if it's there,
or figure out where to fetch it if it's not there.

I could code an Intent system in Python in a couple of days, I think.

>> That doesn't mean than periodic re-flooding is forbidden, but the frequency
>> can be kept quite low.
> 
> And regarding periodic flooding: Depends on the overall model. If we keep reliable connections between nodes (a la BGP) then re-flooding is not required as long as the session is up. 

I have a tendency to prefer stateless solutions where possible, because under
adverse conditions such as intermittent connectivity the tend to survive better.

> If we use unreliable transport, then I personally would prefer a two phase process: 
> 1) a "ping" just with the version number of Intent

That works if there is a single source of Intent. It gets more tricky
if there are multiple sources.

> 2) If one side has a lower than the other, it pulls it reliably.

Sure. That's actually why I prefer TCP for GRASP unicast operations, which
will always be the backup if flooding fails.

Rgds
   Brian

> 
> There are many other ways of doing it, I guess that's a discussion for BA. 
> 
> Michael
> 
>>     Brian
>>
>>
> 


From nobody Fri Mar 25 13:17:53 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8313F12D17E for ; Fri, 25 Mar 2016 13:17:52 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.911
X-Spam-Level: 
X-Spam-Status: No, score=-1.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2hwQOlANSZbs for ; Fri, 25 Mar 2016 13:17:50 -0700 (PDT)
Received: from tuna.sandelman.ca (tuna.sandelman.ca [IPv6:2607:f0b0:f:3:216:3eff:fe7c:d1f3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9694012D104 for ; Fri, 25 Mar 2016 13:17:50 -0700 (PDT)
Received: from sandelman.ca (obiwan.sandelman.ca [209.87.249.21]) by tuna.sandelman.ca (Postfix) with ESMTP id 1244B203C8; Fri, 25 Mar 2016 16:20:41 -0400 (EDT)
Received: from obiwan.sandelman.ca (localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id CB93663755; Fri, 25 Mar 2016 16:17:49 -0400 (EDT)
From: Michael Richardson 
To: "anima\@ietf.org" 
In-Reply-To: <56F35918.4080701@gmail.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com> <485.1458564774@obiwan.sandelman.ca> <56F0BDB4.1040904@gmail.com> <56F35918.4080701@gmail.com>
X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2
X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m
Archived-At: 
Subject: Re: [Anima] Who's a proxy? [was: need for some ASA definitions definitions to help bootstrap]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 20:17:52 -0000

--=-=-=
Content-Type: text/plain


Brian E Carpenter  wrote:
    >> That doesn't really seem right, because it assumes that the NOC has
    >> perfect knowledge of the topology. Potentially, any node with multiple
    >> link interfaces could be a proxy, and if the topology is
    >> non-redundant, they would all need to be proxies.

    > This has been niggling at me.

    > 1. Does it matter if there is more than one proxy on a given LAN?

No, it's fine.
What matters is that there is at least one *legitimate* proxy on a given LAN.

That means that a potential proxy can not just look for the *insecure* proxy
announcement (it could be an attacker), it has to communicate securely to
indicate to other proxies to indicate it is legit.  Or, just always become a
proxy.

    >    1. Do I have more than one LAN interface?  If no, I cannot be a
    > proxy; exit 2.

All machines that participate in the ACP have more than one interface!
Okay, change that to more than one physical interface...

I can see your point that if a device has only one physical interface
(accounting for multiple wifi channels and multiple lambdas...) that being a
proxy is pointless... the other device at the other end of the near ACP hop
ought to be legitimate and can operate a proxy.

--
Michael Richardson , Sandelman Software Works
 -= IPv6 IoT consulting =-




--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBVvWc6oCLcPvd0N1lAQLLuQgAp61epmJxGW9ZVTDJRQV/v8qEwxjxNK2n
ZQFrXAy9KUCQTdI55TAs2KhGMa0uIabmJ99SqGOzMBBxCI1YagArF/046Bm4HPxC
Ww7SFWFY35vitci0Q5RhGedhMFMGQZ8aWUFiZ4KNnX5bmC0Y2zDI14nvSI7IX9TW
sxjXrbEuIwSQxl2J5YzNjbXDnkhJbdhwrq0xgFsIiS+IC1SzNmAGE1IZwUwiqbgL
btN3pmRlq8UcFaPycW/zt1989NBEfi2HN8e2qq/5AtDZWmsagRvRVFmJBdrCgqXg
9j5x7hnBvLm/fSrAltVfbN6i5j8OF7uC1+nUQvrIA87QGiHch2ld3g==
=gvJj
-----END PGP SIGNATURE-----
--=-=-=--


From nobody Fri Mar 25 13:20:20 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 379A112D69D for ; Fri, 25 Mar 2016 13:20:18 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.911
X-Spam-Level: 
X-Spam-Status: No, score=-1.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Tmqcww_hZgDg for ; Fri, 25 Mar 2016 13:20:17 -0700 (PDT)
Received: from tuna.sandelman.ca (tuna.sandelman.ca [209.87.249.19]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D855E12D104 for ; Fri, 25 Mar 2016 13:20:16 -0700 (PDT)
Received: from sandelman.ca (obiwan.sandelman.ca [209.87.249.21]) by tuna.sandelman.ca (Postfix) with ESMTP id 4EF9C203C8 for ; Fri, 25 Mar 2016 16:23:07 -0400 (EDT)
Received: from obiwan.sandelman.ca (localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id 0D39E63755 for ; Fri, 25 Mar 2016 16:20:16 -0400 (EDT)
From: Michael Richardson 
To: "anima\@ietf.org" 
In-Reply-To: <259458c3b8804541aabb0ddf0b151054@XCH-RCD-006.cisco.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com> <485.1458564774@obiwan.sandelman.ca> <56F0BDB4.1040904@gmail.com> <56F35918.4080701@gmail.com> <259458c3b8804541aabb0ddf0b151054@XCH-RCD-006.cisco.com>
X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2
X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m
Archived-At: 
Subject: Re: [Anima] Who's a proxy? [was: need for some ASA definitions definitions to help bootstrap]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 20:20:18 -0000

--=-=-=
Content-Type: text/plain

Michael Behringer (mbehring)  wrote:
    >> 1. Does it matter if there is more than one proxy on a given LAN?

    > I don't see why it would matter. In fact, if you have many proxies, and
    > if new devices pick at random, you'd even get some load balancing (if
    > that matters).

The other reason is that if you have more legitimate proxies than rogue
proxies the odds of the new node picking a legitimate proxy goes up.

    > Simplicity rules. I suggest:
    > - every device is always a proxy, by default
    > - admins have two options to stop the network expanding:
    > 1) Intent "this network shall not expand any further" - will turn all proxies off everywhere
    > (here is where the power of AN comes out!)

maybe the intent would say, "maximum layer-3 hops is X"

    > 2) Local override through traditional means (e.g., config)
    > on an interface: "no autonomic"
    > on device level: "no autonomic proxy"


--
Michael Richardson , Sandelman Software Works
 -= IPv6 IoT consulting =-




--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBVvWdfICLcPvd0N1lAQJK7wgAkxHVqgCe1Qam/QQxz5hBz7WVqoIgMJUG
Uaej8JJLOVJu7HZGB8e19TSlWrvesYJMr5EIK4Ws+ZoB4P+RG2A4bRCCRK6QdgY4
ZvAQ+TzN0TN6iDA3FE08Ylq4lBNUniJ+S7+gjU21/yxFAK3F7EBtWsUuuZyFtF9i
AOd0o6OznA7mDk8JATL0dnyAS6RL3pRu68UsTmvOA4eqzyhdPizVBfzKao0qs6wQ
i+RhVS9qAltRdDlJxifopm3VLdmu2KH31on6XF2WLfwSN9yo8BtjmSPoWvUQVmmA
RRfe/kOc2orSfgtexvB5+f5WxXBnSxdQJ6NkNi8tekWwLeD33DdGlQ==
=fIu3
-----END PGP SIGNATURE-----
--=-=-=--


From nobody Fri Mar 25 15:43:59 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 55B9512D861 for ; Fri, 25 Mar 2016 15:43:57 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id MsT5HwXclxSm for ; Fri, 25 Mar 2016 15:43:54 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 65CDB12D860 for ; Fri, 25 Mar 2016 15:43:54 -0700 (PDT)
Received: from fr712umx3.dmz.alcatel-lucent.com (unknown [135.245.210.42]) by Websense Email Security Gateway with ESMTPS id AA1F64E5B4B0F; Fri, 25 Mar 2016 22:43:47 +0000 (GMT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (fr711usmtp1.zeu.alcatel-lucent.com [135.239.2.122]) by fr712umx3.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2PMhpKu012682 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Fri, 25 Mar 2016 22:43:52 GMT
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id u2PMhpXR000344 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Fri, 25 Mar 2016 23:43:51 +0100
Received: from [135.224.206.200] (135.239.27.41) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Fri, 25 Mar 2016 23:43:50 +0100
To: "EXT Michael Behringer (mbehring)" , "anima@ietf.org" 
References: <90c950d5911040b28da16e33e2ca4d05@XCH-RCD-006.cisco.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56F5BF25.4020100@alcatel-lucent.com>
Date: Fri, 25 Mar 2016 23:43:49 +0100
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <90c950d5911040b28da16e33e2ca4d05@XCH-RCD-006.cisco.com>
Content-Type: multipart/alternative; boundary="------------010902090208070201090300"
X-Originating-IP: [135.239.27.41]
Archived-At: 
Subject: Re: [Anima] Content Distribution
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 25 Mar 2016 22:43:57 -0000

--------------010902090208070201090300
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit

Dear Michael, all,

Thanks for the nice example.
I now understand much better your intent ;) with "flooding".
I am fine with "unicast from/to that neighbor", and this solves my 
concerns/misunderstanding on that aspect (intent file distribution).
Your proposal seems to work fine to have all nodes sharing a same content.

So let's say that all nodes have a copy of the intent file.
My follow-up question relates to your conception of what happens next...
If the intent file contains all the intent policies entered by the 
operator, then it will be up to the nodes and/or ASAs to process the 
policies and act accordingly.
This is a strong assumption which implies that nodes/ASAs know how to 
interpret the content of the intent policies, which part shall be 
processed, how to resolve conflicts, how to coordinate with other 
nodes/ASAs impacted by the intent policies, etc...

I am not saying that this is not feasible, neither to be investigated 
(and maybe finally the way to adopt), but I have the following issues in 
mind:
-this approach seems to imply a replication of the capability to 
process/interpret intent policy in all nodes / ASAs, while it might be 
capability that is factorized at some (TBD) level.
-beyond the simple distribution mechanism (which is fine), this approach 
seems to move the complexity into nodes/ASAs for the processing of 
intent without indicating/considering the policy hierarchy and how 
translation is achieved.
-this approach seems to make it difficult to "create" / manipulate the 
mapping between the different levels of the policy continuum, e.g. 
referring to John example " one could imagine that a single Service 
intent would need multiple Autonomic Functions to be implemented", how 
the nodes/ASAs will know / coordinate wrt. the services they contribute to.

I think John also pointed to some interrogations in replies to Alex 
which are relevant here.

In conclusion:
-Ok with the mechanism proposed for content distribution.
-We need to dig more into the whole picture to assess if we are on the 
right track. I am not suggesting that your example would solve the above 
issues, but that investigating these issues might change how a intent is 
conceived, process, where/by whom and the corresponding design for ANIMA 
components, interactions, protocols.

Thanks, best regards, Laurent.

On 25/03/2016 09:55, EXT Michael Behringer (mbehring) wrote:
> Just writing down how I see Intent distribution as a whole process, for discussion. And, I want to abstract from distributing "Intent" to distributing "content", where "Intent" is one type only.
>
> Starting point: a node is booted; say it has no Intent initially;
>     $content_type = "Intent"
>     $my_version = 0;
>     it consults the AN adjacency table (see section 5 of the reference model draft).
>     it has n neighbours in the same domain --> n ACP connections.
>
> Run every t sec, or when I just received new content:
> - For each neighbour in the adjacency table with an ACP connection:
>     - send $content_type; $my_version
>     - receive $content_type; $remote _version
>     - if  $my_version = $remote _version
>             exit;
>     - if  $my_version < $remote_ version
>             pull $content (unicast from that neighbour)
>     - if  $my _version > $remote_version
>             push $content (unicast to that neighbour)
>
> All of the above happens in the ANI; ASAs are not involved in this operation.
> Probably GRASP does the version comparison; not sure about the file transfer.
>
> The $content_type allows us to use the same mechanism to flood other things; this gives us a lot of flexibility to distribute other content types later on; and, if we find out that we *really* need to separate Intent for different autonomic functions, we could then say $content_type = "Intent for function a" for example (of course this is schematic only; we may want sub-fields).
>
> The Intent is stored as a file. it is that file that gets re-distributed. This file is locally validated (signature) and parsed, and then passed to the ASAs. And a portion of Intent will affect the ANI itself.
>
> How far off am I with this thinking? I think this is very simple and robust.
>
> draft-liu-anima-grasp-distribution-01 has I think the same fundamental thinking, but specified a lot of things that are not needed, IMO:
> - I do not think we need selective distribution (2.2)
> - I do not think we need incremental distribution (2.3)
> - 3.1 is a bit redundant, if we assume distribution in the ACP. I would rather state
>    as above "to all ACP neighbors", then we have the boundary detection implicit.
> - I think the above algorithm avoids loops automatically.
> - I don't think we need selective flooding (3.4)
>
> Michael
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------010902090208070201090300
Content-Type: text/html; charset="windows-1252"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Dear Michael, all,

      

      Thanks for the nice example.

      I now understand much better your intent ;) with "flooding".

      I am fine with "unicast from/to that neighbor", and this solves my
      concerns/misunderstanding on that aspect (intent file
      distribution).

      Your proposal seems to work fine to have all nodes sharing a same
      content.

      

      So let's say that all nodes have a copy of the intent file.

      My follow-up question relates to your conception of what happens
      next...

      If the intent file contains all the intent policies entered by the
      operator, then it will be up to the nodes and/or ASAs to process
      the policies and act accordingly.

      This is a strong assumption which implies that nodes/ASAs know how
      to interpret the content of the intent policies, which part shall
      be processed, how to resolve conflicts, how to coordinate with
      other nodes/ASAs impacted by the intent policies, etc...

      

      I am not saying that this is not feasible, neither to be
      investigated (and maybe finally the way to adopt), but I have the
      following issues in mind:

      -this approach seems to imply a replication of the capability to
      process/interpret intent policy in all nodes / ASAs, while it
      might be capability that is factorized at some (TBD) level.

      -beyond the simple distribution mechanism (which is fine), this
      approach seems to move the complexity into nodes/ASAs for the
      processing of intent without indicating/considering the policy
      hierarchy and how translation is achieved. 

      -this approach seems to make it difficult to "create" / manipulate
      the mapping between the different levels of the policy continuum,
      e.g. referring to John example " one could imagine that a single
      Service intent would need multiple Autonomic Functions to be
      implemented", how the nodes/ASAs will know / coordinate wrt. the
      services they contribute to.

      

      I think John also pointed to some interrogations in replies to
      Alex which are relevant here.

      

      In conclusion:

      -Ok with the mechanism proposed for content distribution.

      -We need to dig more into the whole picture to assess if we are on
      the right track. I am not suggesting that your example would solve
      the above issues, but that investigating these issues might change
      how a intent is conceived, process, where/by whom and the
      corresponding design for ANIMA components, interactions,
      protocols.

      

      Thanks, best regards, Laurent.

    

    
On 25/03/2016 09:55, EXT Michael
      Behringer (mbehring) wrote:

    

    

      
Just writing down how I see Intent distribution as a whole process, for discussion. And, I want to abstract from distributing "Intent" to distributing "content", where "Intent" is one type only. 

Starting point: a node is booted; say it has no Intent initially; 
   $content_type = "Intent"
   $my_version = 0;
   it consults the AN adjacency table (see section 5 of the reference model draft).
   it has n neighbours in the same domain --> n ACP connections. 

Run every t sec, or when I just received new content: 
- For each neighbour in the adjacency table with an ACP connection:  
   - send $content_type; $my_version
   - receive $content_type; $remote _version
   - if  $my_version = $remote _version
           exit;
   - if  $my_version < $remote_ version
           pull $content (unicast from that neighbour)
   - if  $my _version > $remote_version
           push $content (unicast to that neighbour)

All of the above happens in the ANI; ASAs are not involved in this operation. 
Probably GRASP does the version comparison; not sure about the file transfer.

The $content_type allows us to use the same mechanism to flood other things; this gives us a lot of flexibility to distribute other content types later on; and, if we find out that we *really* need to separate Intent for different autonomic functions, we could then say $content_type = "Intent for function a" for example (of course this is schematic only; we may want sub-fields). 

The Intent is stored as a file. it is that file that gets re-distributed. This file is locally validated (signature) and parsed, and then passed to the ASAs. And a portion of Intent will affect the ANI itself. 

How far off am I with this thinking? I think this is very simple and robust. 

draft-liu-anima-grasp-distribution-01 has I think the same fundamental thinking, but specified a lot of things that are not needed, IMO: 
- I do not think we need selective distribution (2.2)
- I do not think we need incremental distribution (2.3)
- 3.1 is a bit redundant, if we assume distribution in the ACP. I would rather state
  as above "to all ACP neighbors", then we have the boundary detection implicit. 
- I think the above algorithm avoids loops automatically. 
- I don't think we need selective flooding (3.4)

Michael

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima



    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------010902090208070201090300--


From nobody Fri Mar 25 17:30:00 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B992212D0C9 for ; Fri, 25 Mar 2016 17:29:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IqUpbW_z5Uxm for ; Fri, 25 Mar 2016 17:29:57 -0700 (PDT)
Received: from mail-pa0-x233.google.com (mail-pa0-x233.google.com [IPv6:2607:f8b0:400e:c03::233]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3867412D739 for ; Fri, 25 Mar 2016 17:29:57 -0700 (PDT)
Received: by mail-pa0-x233.google.com with SMTP id fl4so55685404pad.0 for ; Fri, 25 Mar 2016 17:29:57 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=QDNmt/ECyZTRSvPVo0Spf5Q0FKLG585wWWjzKWig6XQ=; b=hu8IyLIqN5XqWtaDsPvbDHbnONvA5OYMvfsMbgvGMhSV4flpweS0dNQfTE6skhSw0U fyyl+TgO2UExSiw5+2nXLAxL6wEA5Cn720mUV7ll+KVoYuEe863sGJbrfKygntyjPe5J 17PWHZsnm/q7eR77t6wbU+InjzSnXLnCFVT6df0RtUgHskNIMrnwcWga/richT74YbR1 7eHmW67NlgIakfm2ytWGIBeooAaxnBUCodi2ROv6EJARBy9c1Q4M81YUkCR6gC0lW+9/ 6FXGqdYfMe0SHOQDFcbHk+kBI29zR11reETklIdvVPelGEEQ6M9X7jT8+fWVBqhK91pT j5Uw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=QDNmt/ECyZTRSvPVo0Spf5Q0FKLG585wWWjzKWig6XQ=; b=Pyo+X3IwTIOaJONhqvlwkUXdrytH8zG1dJpEiyjV58+WBrmGG/XoKLx0V1J+DLxZ66 QiQrdeNZRmc60OekzLE21nYa619B3gGwWikMGpsfyt++wNqPsriNkyvqsxrM33NOuS0f tx+UZF87VMCxo3IUvA94KH0dBoYDbQv9PoGTrQBLIHqIPd1IkeMWx60VxNTBz63DdfS6 8RyvYJ/VOKLK6/d1TMhWl1L2D0EvZPCebEDWpe9m9nbKmUXB/gWo/CDQDzvxVYuI+eTH kbT9q+syVLtOCvGYesbuaOSLHrBaEbVXM9Li6yGSdA5ND6EKbdv3qijj9cKdw7jQR0jy E8/g==
X-Gm-Message-State: AD7BkJLlrhoR1DLgUijlZhZmcyBfCHdlf1CnAEB5JISTsjMJM3jMcFMTBOf9TVWZf2l1XQ==
X-Received: by 10.66.249.70 with SMTP id ys6mr24987639pac.5.1458952196832; Fri, 25 Mar 2016 17:29:56 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id d12sm18800463pfj.85.2016.03.25.17.29.52 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 17:29:55 -0700 (PDT)
To: "Michael Behringer (mbehring)" , Laurent Ciavaglia , "EXT Joel M. Halpern" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com> <56F2E3DA.9030204@joelhalpern.com> <96d097c6115d4aac84d523469d190176@XCH-RCD-006.cisco.com> <56F409D0.9030900@joelhalpern.com> <56F41351.8080309@alcatel-lucent.com> <56F4437C.7000803@gmail.com> <56F45A42.4090207@alcatel-lucent.com> 
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F5D80C.5040301@gmail.com>
Date: Sat, 26 Mar 2016 13:30:04 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Duzongpeng , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Re-flooding [was: Does Autonomic network need some intervention beyond intent]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 00:29:59 -0000

On 25/03/2016 21:00, Michael Behringer (mbehring) wrote:
> I think we have a misunderstanding in terminology.
>=20
> When I say =E2=80=9Cflooding=E2=80=9D I mean that the same Intent data =
structure is passed on hop by hop through the entire network and every no=
de sees exactly the same data structure.
>=20
> This may be implemented by passing the Intent file hop by hop in a unic=
ast transmission. The key being =E2=80=9Chop by hop=E2=80=9D.
>=20
> When I hear the word =E2=80=9Cunicast=E2=80=9D I suspect that some peop=
le mean =E2=80=9Cunicast to a remote node=E2=80=9D or to an =E2=80=9CInte=
nt server=E2=80=9D. That I don=E2=80=99t think is necessary, and will mak=
e things more complex.

But we already need that for specific ASA-to-ASA synchronization (or nego=
tiation,
but I assume Intent is non-negotiable). So it really isn't additional com=
plexity.
As the example in my previous mail suggests, an ASA in need of intent doe=
sn't
really care whether the GRASP engine got the Intent by flooding or by poi=
nt-to-point
synchronization.

I do agree that the ASA shouldn't need to know the details; it just needs=
 a way
to get the current Intent.

   Brian

>=20
> Michael
>=20
> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Laurent Ciavag=
lia
> Sent: 24 March 2016 22:21
> To: EXT Brian E Carpenter ; Laurent Ciavag=
lia ; EXT Joel M. Halpern ; Michael Behringer (mbehring) 
> Cc: Duzongpeng ; anima@ietf.org; Sheng Jiang 
> Subject: Re: [Anima] Re-flooding [was: Does Autonomic network need some=
 intervention beyond intent]
>=20
> Brian,
>=20
> On 24/03/2016 20:43, EXT Brian E Carpenter wrote:
>=20
> On 25/03/2016 05:18, Laurent Ciavaglia wrote:
>=20
> ...
>=20
> It is not because an intent policy is not changed, that recipients (e.g=
=2E new ASA instances) should not be informed about the intent policies, =
and thus generates a new broadcast.
>=20
>=20
>=20
> A newly started ASA can (and probably should) perform the following log=
ic:
>=20
>=20
>=20
>  - discover the Intent server (just another ASA, of course)
>=20
>  - request unicast synchronization of the latest intent
>=20
>=20
>=20
> Actually the same applies to sleeping nodes: at some frequency
>=20
> they should wake up and synchronize intent, or any other flooded
>=20
> objective that they care about. This is already specified in GRASP:
>=20
>   "Sleeping nodes that require an objective subject to flooding SHOULD
>=20
>    periodically request unicast synchronization for that objective."
>=20
>=20
>=20
> That doesn't mean than periodic re-flooding is forbidden, but the frequ=
ency
>=20
> can be kept quite low.
>=20
> Agreed.
> So, there are at least two mechanisms involved concerning intent "distr=
ibution":
>     -flooding (when intent is created, updated, etc.)
> and
>     -unicast or pub/sub or ... for the cases we described.
>=20
> Michael B./all: do you agree with the above? does it make a case for co=
nsidering more than only flooding as the mechanism for intent distributio=
n?
>=20
> Thanks, best regards,
> Laurent.
>=20


From nobody Fri Mar 25 17:41:38 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ADBE412D0B5 for ; Fri, 25 Mar 2016 17:41:36 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NXUxND70mvgb for ; Fri, 25 Mar 2016 17:41:35 -0700 (PDT)
Received: from mail-pa0-x233.google.com (mail-pa0-x233.google.com [IPv6:2607:f8b0:400e:c03::233]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 102B312D099 for ; Fri, 25 Mar 2016 17:41:35 -0700 (PDT)
Received: by mail-pa0-x233.google.com with SMTP id fe3so55903009pab.1 for ; Fri, 25 Mar 2016 17:41:35 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=BSAuwvvzspAdpQR1gqo6XATGx/lUq2IEx3E/PqCN3kE=; b=jVAC1ZtBQir43AHgPPrPfMKKsBZrsJ863gT+mEWj2fTgfpSbVDU5hO8dBAgxmHfI/Y 4Fl+JU/6xRX3V4+x2GhA0DbGxj+g0uIAxNIYvuG0tXk6MSUphwChCZbN+UUbC7CaWCTj KKd2x9G6tfl857pIYikTbqQIu1l0R/G2qDhNCa9HdjNyJiTwgjdEZPEihyyLsCAfUHw6 xMDPDCExAM/IBDIuIdnNtuy6ID6VssI7FsHIRZVPSHjKJSP1lKSVML/A/0NAnlfTRNSt 9p+1SXh2wURC3KkIACT/ouKhBxLpEGjNKUo7ZFErzDDjU9jCBXY2S7k0SQpPMYFBr2pm dKiA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=BSAuwvvzspAdpQR1gqo6XATGx/lUq2IEx3E/PqCN3kE=; b=L/cX2Qz1QrVGPkeMcqFiZrKLa0rXtdBGkTer/8jTYOU/zT0JtcxcVvnGO42SlnsLwi nOPdDyE3c6jvDHTwJiW2wpQ1g8sPHUR9+9cVrOp/fz0+ZfyyHsVYDNoX83KsmaO7whGk F51KdoCUBWlGhq+LY5xQtSA+2ceZ3TNjYRhj9+DZVUYn0vS9hQ9mjqGHhsV813XPOpZA OJL1HbWEj4AhknHN6S9Z1QrzOqyZZYkSg2BgER/9d2xDicjeihTgv9kAklOJQoTDS4d7 zAgblLAGtR0HvhY1iHiTh8buvS4c85LdDI35nnGPr0EvLVgN3dDJd4GSyWcAOuOjufCg n9IQ==
X-Gm-Message-State: AD7BkJK1T/N6G3C6AgQgd/LXtC2NApAqwAMrn1ZapcFQDswJAD3c3wO7eafAwssfGeLzMg==
X-Received: by 10.66.228.1 with SMTP id se1mr24835195pac.43.1458952894665; Fri, 25 Mar 2016 17:41:34 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id 83sm18855116pfn.46.2016.03.25.17.41.30 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 17:41:32 -0700 (PDT)
To: Michael Richardson , "anima@ietf.org" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <56F28403.6060401@alcatel-lucent.com>  <6723.1458935746@obiwan.sandelman.ca>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F5DAC6.4050708@gmail.com>
Date: Sat, 26 Mar 2016 13:41:42 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <6723.1458935746@obiwan.sandelman.ca>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: John Strassner 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 00:41:36 -0000

On 26/03/2016 08:55, Michael Richardson wrote:
> 
> John Strassner  wrote:
>     >  Hopefully, intent will not change frequently (or else it probably
>     > is too low level in nature). I personally like using pub-sub, but that
>     > is because it enables me to do additional pre- and post-processing of
>     > intent. However, I think that this should be an option.  < /jcs>
> 
> Why not flood the list of intents (by suitable-hash of intent file), and
> along with an origin for the intent.  (this winds up looking a lot like
> OSPF, and is what DNCP is good at...)

Sure, if it turns out that we need such a mechanism, there are plenty
of precedents.

> Then, any node which does not have a copy of that intent can ask for it using
> any of our well known protocols (i.e. HTTP).

Well, this would also fit immediately into GRASP too. Just add a "virtual"
objective called "Intent.Hash" which you could synchronize any time, and
if it doesn't match your local hash, you know you need to resynch "Intent".

    Brian


From nobody Fri Mar 25 17:47:09 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EEAF812D0B4 for ; Fri, 25 Mar 2016 17:47:07 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.419
X-Spam-Level: 
X-Spam-Status: No, score=-2.419 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FREEMAIL_FORGED_FROMDOMAIN=0.199, FREEMAIL_FROM=0.001, HEADER_FROM_DIFFERENT_DOMAINS=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YjTaF7ey6wOM for ; Fri, 25 Mar 2016 17:47:04 -0700 (PDT)
Received: from mail-ig0-f173.google.com (mail-ig0-f173.google.com [209.85.213.173]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6E1D312D099 for ; Fri, 25 Mar 2016 17:47:04 -0700 (PDT)
Received: by mail-ig0-f173.google.com with SMTP id cl4so21095277igb.0 for ; Fri, 25 Mar 2016 17:47:04 -0700 (PDT)
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to; bh=5rsU0w0mqqid+qK9BWQ78X8l6pKhuWbsshMHEG0759Q=; b=bCyLsQKvXHiMDRfWm0+OrRvCPnD13aI20oXD8PFOZpmdjZKvlA98r+5HRWssXeP9gt ZJHQv3IZBx93JMm17LLNJxxKNw5ekQIEi7CnAeg+BrmSwuDhHYNlQvWiXtQCYgxFWdBE HHddGF44mxWX4xA29vCWOsjKZo062yZ75EJ14lEYN0yPtwU5Ihsu/418RpJX9FvCKnAi q0ICW0CZqs54Mt6pRlGjFLbSRKDUWX8xmxWNOlvC6vTBaFJQPMFY32i/E9MtJYJJODfH vAqHQDRBOUv2Xo026RB79bS7cK8fHMeOxe2tJVnqDk5E5SpvH2IjNi7+mt0EZHG43S2J L1Kg==
X-Gm-Message-State: AD7BkJJCLJbj2+yIp0zZeJ62hyQri1vCZkE8Q6KZeA4S9peqJUQwm7KzoE4d+xcYcuJD5dbqg5rMwFGldJoFEw==
X-Received: by 10.50.65.74 with SMTP id v10mr1011805igs.33.1458953223694; Fri, 25 Mar 2016 17:47:03 -0700 (PDT)
MIME-Version: 1.0
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com> <56F4718D.8010104@alcatel-lucent.com>  
In-Reply-To: 
From: =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= 
Date: Sat, 26 Mar 2016 00:46:54 +0000
Message-ID: 
To: "Michael Behringer (mbehring)" , Laurent Ciavaglia ,  "anima@ietf.org" 
Content-Type: multipart/alternative; boundary=047d7b3a9b2ca47345052ee9041d
Archived-At: 
Subject: Re: [Anima] Reference Model Issues
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 00:47:08 -0000

--047d7b3a9b2ca47345052ee9041d
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Michael.
I agree with your proposal to move everything that is outside the charter
into a future work/steps section.
I can further short the section I wrote (intent and aggregated reporting).
If you need additional help in other sections, just let me know.
Best.
J=C3=A9ferson

Em sex, 25 de mar de 2016 =C3=A0s 11:23, Michael Behringer (mbehring) <
mbehring@cisco.com> escreveu:

> Yes, I=E2=80=99m thinking that we move EVERYTHING that isn=E2=80=99t in t=
he current
> charter into a =E2=80=9Cfutures=E2=80=9D section. So that would include:
>
> -          Intent
>
> -          feedback loops
>
> -          aggregated reporting
>
> -          APIs
>
> -          current section 9
>
> -          constrained devices
>
> -          cross-domain behaviour
>
> -          etc
>
>
>
> This way the document would be very clearly structured, and we could stil=
l
> keep the current content, which, frankly, I don=E2=80=99t want to lose.
>
>
>
> Michael
>
>
>
>
>
> *From:* J=C3=A9ferson Campos Nobre [mailto:jcnobre@inf.ufrgs.br]
> *Sent:* 25 March 2016 15:17
> *To:* Laurent Ciavaglia ; Michael Behringer
> (mbehring) ; anima@ietf.org
> *Subject:* Re: [Anima] Reference Model Issues
>
>
>
> Hi.
>
> + 1 for the grouping.
>
> Michael, are you thinking in maintaining subsections in the future work
> section?
>
> Best.
>
> J=C3=A9ferson
>
>
>
> Em qui, 24 de mar de 2016 =C3=A0s 20:00, Laurent Ciavaglia <
> laurent.ciavaglia@nokia.com> escreveu:
>
> Dear Michael, all,
>
> Thanks for your continuous care of the document and taking some hindsight=
!
>
> I am also more in favor of grouping non-chartered aspects into a future
> work section, and allow the text there to be meaningful for 1) the
> relationship with (chartered) ANIMA components, 2) the overall picture, a=
nd
> 3) describing the implications of these future items with enough depth.
>
> I think the reference document is a pivot one in ANIMA to prepare what
> will come next (in ANIMA or elsewhere) and improve our collective
> understanding on future areas to investigate.
> I think it would be a mistake to neglect this opportunity by restricting
> the document on chartered items.
>
> Thanks, Laurent.
> ps: of course my above support has nothing to do with the fact that I'm
> too lazy busy to condense the section I am editing ;)
>
>
>
> On 24/03/2016 18:52, EXT Michael Behringer (mbehring) wrote:
>
> I just read through the entire document again, noting down things that ne=
ed fixing, or discussion. The below is my personal view, for discussion. I =
cleaned up the github repository, and created a new file, corresponding to =
the latest version plus some small edits I made today while reading through=
 again.
>
>
>
> One important point: Our chairs want us to take out or reduce drastically=
 all the stuff that doesn't belong into the current charter. I agree that i=
t is not always 100% clear if something is out of charter. But I'm having a=
 hard time reducing some sections. What if we make a new section "Future wo=
rk" (which clearly indicates out of scope for this charter), where we colle=
ct those points, but then are a bit more liberal about the volume of each s=
ection? I think that would be a good compromise. Sheng?
>
>
>
> I'll summarise the below points in the WG meeting in BA. Maybe we can alr=
eady clarify some stuff before.
>
>
>
> Michael
>
>
>
>
>
>
>
>
>
> Reference model issues
>
> 24 Mar 2016
>
>
>
> 4.1 Naming:
>
> - "It is recommended that the names are generated by the autonomic nodes
>
>    themselves." - how? Should names / addresses not come from the registr=
ar?
>
>    Suggestion: Name *and* ACP address should be assigned by the registrar
>
>    at enrolment time. zone-id is dynamic, rest is fixed.
>
>
>
> - "a specific naming convention is out of scope":
>
>    Suggestion: We define a default naming and addressing scheme.
>
>    1st we settle on an addressing scheme from the ACP draft.
>
>    That defines a "device-ID" (last n bits of ACP address).
>
>    convert that to a string, and use it as a name. This would result in
>
>    something like: 0123-4567-89ab-0001.example.com.
>
>    The name would go into the domain certificate --> need to put this int=
o
>
>    the bootstrap draft as well.
>
>
>
> - to write in the doc: "registrar picks a naming scheme; all registrars
>
>   in a domain must use the same scheme. Example is: ...
>
>
>
> 4.4 ACP
>
> - should move to the end of section 4, it breaks the flow where it is rig=
ht now.
>
>
>
> 4.7 Intent Distribution
>
> - This section should only explain the basics, real work is out of scope =
for this phase.
>
> - should point to the intent distribution draft
>
> - should explain that intent is flooded to all nodes in a domain (if we a=
gree),
>
>   or other methods (if not).
>
> - should explain that we expect Intent to have a long life time (months),=
 thus
>
>   Intent distribution is expected to be very infrequent.
>
> - should explain that the entire Intent file is flooded in one go (if we =
agree).
>
>
>
> 5. Functional Overview
>
> - title is not good. This section really describes how an autonomic node =
behaves. maybe call it "Behaviour of an autonomic node"?
>
> - we need to describe the bring-up better, specifically insecure discover=
y,
>
>   ACP negotiation, ACP bringup, and subsequent operations, and which prot=
ocols to use
>
>   where. Or should this go into the ACP draft??
>
>
>
> 6. Security and Trust Infrastructure
>
> - needs a review from someone in the security space.
>
> - we need to describe the certificate format; this should likely go into =
bootstrap draft. This draft should explain where the certificate format is =
described.
>
> - the domain certificate should also contain the ACP IP address (where th=
e zone bits are set to zero).
>
>
>
> 7.1
>
> - broken paragraph; needs fixing.
>
> - there is a mention of constrained devices - take out.
>
>
>
> 7.2 Specific ASAs
>
> - should explain MUST implement
>
> - generally, need to work through text in this section; relate back to se=
ction 5.
>
>
>
> 8.1
>
> - remove last reference to RFC7575.
>
>
>
> 8.2 Intent
>
> - explain not part of current charter first.
>
> - not sure whether we should keep this level of detail, or shorten furthe=
r?
>
>
>
> 8.3 Aggregated Reporting
>
> - explain not part of current charter first.
>
> - my feeling: Should be reduced to 1-2 paragraphs, only high level idea.
>
>
>
> General:
>
> - should we combine all items that are not part of the current charter in=
to a separate section "Future Topics" or something like that? Then all (*) =
topic would be nicely contained. Yes, I think I'd prefer that to taking the=
m out completely, or shortening them to next-to-nothing.
>
>
>
> 10. Security Considerations
>
> - needs to be worked over.
>
>
>
> General:
>
> - uppercase Intent everywhere in the doc.
>
>
>
> _______________________________________________
>
> Anima mailing list
>
> Anima@ietf.org
>
> https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
>
> --
>
> Laurent Ciavaglia
>
> Senior Research Manager
>
> Bell Labs, Nokia
>
>
>
> +33 160 402 636
>
> Route de Villejust | 91620 Nozay | France
>
> LinkedIn: laurentciavaglia 
>
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>

--047d7b3a9b2ca47345052ee9041d
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi Michael.
I agree with your proposal to move everyth=
ing that is outside the charter into a future work/steps section.
I can further short the section I wrote (intent and aggregated reporting).=

If you need additional help in other sections, just let me know.=

Best.
J=C3=A9ferson

=

Em sex, 25 de mar de 2016 =C3=A0s 11:23, Michael Behringer=
 (mbehring) <mbehring@cisco.com> escreveu:











=
Yes, I=E2=80=99m thinking that we move EVERYTHING that isn=
=E2=80=99t in the current charter into a =E2=80=9Cfutures=E2=80=9D
 section. So that would include: 


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0
Intent


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0
feedback loops


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0
aggregated reporting=



-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0
APIs


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0
current section 9


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0
constrained devices<=
/font>


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0
cross-domain behaviour


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0
etc


=
=C2=A0


=
This way the document would be very clearly structured, and =
we could still keep the current
 content, which, frankly, I don=E2=80=99t want to lose. 


=
=C2=A0


=
Michael


=
=C2=A0


=
=C2=A0








From:=

 J=C3=A9ferson Campos Nobre [mailto:jcnobre@inf.ufrgs.br] 

Sent: 25 March 2016 15:17
To: Laurent Ciavaglia <laurent.ciav=
aglia@nokia.com>; Michael Behringer (mbehring) <mbehring@cisco.com>; anima@ietf.org

Subject: Re: [Anima] Referen=
ce Model Issues




=



=C2=A0








Hi.






+ 1 for the grouping.





Michael, are you thinking in maintaining subsections=
 in the future work section?






Best.






J=C3=A9ferson






=C2=A0






Em qui, 24 de mar de 2016 =C3=A0s 20:00, Laurent Cia=
vaglia <laurent.ciavaglia@nokia.com> escreveu:<=
/p>







Dear Michael, all,


Thanks for your continuous care of the document and taking some hindsight!<=
br>


I am also more in favor of grouping non-chartered aspects into a future wor=
k section, and allow the text there to be meaningful for 1) the relationshi=
p with (chartered) ANIMA components, 2) the overall picture, and 3) describ=
ing the implications of these future
 items with enough depth.



I think the reference document is a pivot one in ANIMA to prepare what will=
 come next (in ANIMA or elsewhere) and improve our collective understanding=
 on future areas to investigate.

I think it would be a mistake to neglect this opportunity by restricting th=
e document on chartered items.










Thanks, Laurent.

ps: of course my above support has nothing to do with the fact that I'm=
 too lazy busy to condense the section I am editing ;)=







=C2=A0




On 24/03/2016 18:52, EXT Michael Behringer (mbehring=
) wrote:






I just read through the entire document again, noting down things that nee=
d fixing, or discussion. The below is my personal view, for discussion. I c=
leaned up the github repository, and created a new file, corresponding to t=
he latest version plus some small edits I made today while reading through =
again. 


=C2=A0


One important point: Our chairs want us to take out or reduce drastically =
all the stuff that doesn't belong into the current charter. I agree tha=
t it is not always 100% clear if something is out of charter. But I'm h=
aving a hard time reducing some sections. What if we make a new section &qu=
ot;Future work" (which clearly indicates out of scope for this charter=
), where we collect those points, but then are a bit more liberal about the=
 volume of each section? I think that would be a good compromise. Sheng? 


=C2=A0


I'll summarise the below points in the WG meeting in BA. Maybe we can =
already clarify some stuff before. 


=C2=A0


Michael


=C2=A0


=C2=A0


=C2=A0


=C2=A0


Reference model issues


24 Mar 2016


=C2=A0


4.1 Naming: 


- "It is recommended that the names are generated by the autonomic no=
des


=C2=A0=C2=A0 themselves." - how? Should names / addresses not come fr=
om the registrar? 


=C2=A0=C2=A0=C2=A0Suggestion: Name *and* ACP address should be assigned by=
 the registrar


=C2=A0=C2=A0 at enrolment time. zone-id is dynamic, rest is fixed. =



=C2=A0


- "a specific naming convention is out of scope": =



=C2=A0=C2=A0=C2=A0Suggestion: We define a default naming and addressing sc=
heme. 


=C2=A0=C2=A0=C2=A01st we settle on an addressing scheme from the ACP draft=
. 


=C2=A0=C2=A0=C2=A0That defines a "device-ID" (last n bits of ACP=
 address). 


=C2=A0=C2=A0=C2=A0convert that to a string, and use it as a name. This wou=
ld result in


=C2=A0=C2=A0 something like: 0123-4567-89ab-0001.example.com. =



=C2=A0=C2=A0=C2=A0The name would go into the domain certificate --> nee=
d to put this into 


=C2=A0=C2=A0=C2=A0the bootstrap draft as well. 


=C2=A0=C2=A0=C2=A0


- to write in the doc: "registrar picks a naming scheme; all registra=
rs 


=C2=A0=C2=A0in a domain must use the same scheme. Example is: ... <=
u>


=C2=A0=C2=A0=C2=A0


4.4 ACP


- should move to the end of section 4, it breaks the flow where it is righ=
t now. 


=C2=A0


4.7 Intent Distribution


- This section should only explain the basics, real work is out of scope f=
or this phase.


- should point to the intent distribution draft


- should explain that intent is flooded to all nodes in a domain (if we ag=
ree), 


=C2=A0=C2=A0or other methods (if not). 


- should explain that we expect Intent to have a long life time (months), =
thus


=C2=A0 Intent distribution is expected to be very infrequent. 


- should explain that the entire Intent file is flooded in one go (if we a=
gree). 


=C2=A0


5. Functional Overview


- title is not good. This section really describes how an autonomic node b=
ehaves. maybe call it "Behaviour of an autonomic node"? 


- we need to describe the bring-up better, specifically insecure discovery=
, 


=C2=A0=C2=A0ACP negotiation, ACP bringup, and subsequent operations, and w=
hich protocols to use


=C2=A0 where. Or should this go into the ACP draft?? =



=C2=A0


6. Security and Trust Infrastructure


- needs a review from someone in the security space.<=
/font>


- we need to describe the certificate format; this should likely go into b=
ootstrap draft. This draft should explain where the certificate format is d=
escribed. 


- the domain certificate should also contain the ACP IP address (where the=
 zone bits are set to zero). 


=C2=A0


7.1 


- broken paragraph; needs fixing. 


- there is a mention of constrained devices - take out. 


=C2=A0


7.2 Specific ASAs


- should explain MUST implement


- generally, need to work through text in this section; relate back to sec=
tion 5. 


=C2=A0


8.1 


- remove last reference to RFC7575. 


=C2=A0


8.2 Intent


- explain not part of current charter first. <=
/pre>

- not sure whether we should keep this level of detail, or shorten further=
? 


=C2=A0


8.3 Aggregated Reporting


- explain not part of current charter first. <=
/pre>

- my feeling: Should be reduced to 1-2 paragraphs, only high level idea.


=C2=A0


General: 


- should we combine all items that are not part of the current charter int=
o a separate section "Future Topics" or something like that? Then=
 all (*) topic would be nicely contained. Yes, I think I'd prefer that =
to taking them out completely, or shortening them to next-to-nothing.


=C2=A0


10. Security Considerations


- needs to be worked over. 


=C2=A0


General: 


- uppercase Intent everywhere in the doc. 

=C2=A0


_______________________________________________


Anima mailing list


Anima@ietf.org<=
/u>


=
https://www.ietf.org/mailman/listinfo/anima=



=C2=A0




=C2=A0








--





Laurent Ciavaglia


Senior Research Manager<=
/p>

Bell Labs, Nokia


=C2=A0

+33=C2=A0160 402=C2=A0636



Route de Villejust | 91620 Nozay | France<=
/u>


LinkedIn:
<=
span lang=3D"FR">laurentciavaglia

=C2=A0<=
/span>








_______________________________________________

Anima mailing list

Anima@ietf.org

h=
ttps://www.ietf.org/mailman/listinfo/anima<=
/p>









_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima




--047d7b3a9b2ca47345052ee9041d--


From nobody Fri Mar 25 17:48:36 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8CB0812D105 for ; Fri, 25 Mar 2016 17:48:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0ZVJEOWwxVG3 for ; Fri, 25 Mar 2016 17:48:33 -0700 (PDT)
Received: from mail-pf0-x229.google.com (mail-pf0-x229.google.com [IPv6:2607:f8b0:400e:c00::229]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B721A12D0F9 for ; Fri, 25 Mar 2016 17:48:32 -0700 (PDT)
Received: by mail-pf0-x229.google.com with SMTP id n5so92490858pfn.2 for ; Fri, 25 Mar 2016 17:48:32 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=rFHMENwyF5iGvVBjDCi4LAKyyGKiSVcVvxnzgj15THM=; b=UEultbHi0SjxCUd0Zl4WVE7V4NDPaZaKMYjn1SU8FgnrdCI0xR59C6frxcWULyZiuF CiEqMj1ycQj+9l+E1kF3WjG4iCXTYTazF4rYbaEh7CfFaPW0gpezq4BapaDcHr4aj8ks kj5NtcGY7XP9s04wdu2yX0USFQf8eQtpZoGSdc2t1uf7xNB4u5m059iNK7Yi6EdEcbIn gtHMmBOwNZY2sgyiTeOGr2jsfBzas3t6BVxtamXYDc/oU//JSfcOvZwoXjNPEYYzq+fd dvB6PRYOsHZgqxYIXvBKszG9HXUEVCrwLGJveVAxKUbb5Qo8gjAlHKIq5hTd6hQDfk9W Rh+g==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=rFHMENwyF5iGvVBjDCi4LAKyyGKiSVcVvxnzgj15THM=; b=QlzDl980exCz8AiykE+962u/ExmUVZmxgQEwRWl7Yc02+mBXyZvQKLnK80IuSHVkz+ WycLO3hY1V91ZcOa6qVB6gIWITXhNjdZ8WLWCjp9SGeWfM8gXNoFyPaKMefdcDjQuHhE HuZGuC3Ney5WckM3T6lN9DTp9hQ3MtswRLSv0JUdktcKS7nL9w0B1fOdnk9XFi8IyPkO wbliC2X7R7r4ecUt1sU35fbp/YeT1xEdfv0Kw63iruXxudCAfOw2M0MCK9Elz3CCtDvb 0VxflHybt/+hAOv75ztg7DwxXv4OaFz9CZCE8tYWObUVPI/hTBsR8hqNh9qoq5968pnW kToA==
X-Gm-Message-State: AD7BkJKoCLwo52CmUjv+6MAuNngwFVq92OLnZ4U8Q/tgUolQRLfrrlBbe1Fg0IZWI43zGQ==
X-Received: by 10.98.11.78 with SMTP id t75mr24666829pfi.72.1458953312379; Fri, 25 Mar 2016 17:48:32 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id r65sm18875049pfa.27.2016.03.25.17.48.29 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 17:48:31 -0700 (PDT)
To: "Michael Behringer (mbehring)" , "anima@ietf.org" 
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F5DC69.7020005@gmail.com>
Date: Sat, 26 Mar 2016 13:48:41 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: [Anima] 4.1 Naming [was Reference Model Issues]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 00:48:34 -0000

On 25/03/2016 06:52, Michael Behringer (mbehring) wrote:

...
> 4.1 Naming: 
> - "It is recommended that the names are generated by the autonomic nodes
>    themselves." 

Just before that, it says
"Inside a domain, each autonomic device should be assigned a name."

I have a heretical question: why do autonomic nodes need names?

They clearly need ACP addresses and they need cryptographic identities.
But what is the purpose of a human-readable name?

The answer should tell us what sort of naming system we need.

     Brian


From nobody Fri Mar 25 17:51:06 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5311A12D0B4 for ; Fri, 25 Mar 2016 17:51:05 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0dhFrsL3vBcs for ; Fri, 25 Mar 2016 17:51:03 -0700 (PDT)
Received: from mail-pf0-x22a.google.com (mail-pf0-x22a.google.com [IPv6:2607:f8b0:400e:c00::22a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8277F12D099 for ; Fri, 25 Mar 2016 17:51:03 -0700 (PDT)
Received: by mail-pf0-x22a.google.com with SMTP id x3so92761149pfb.1 for ; Fri, 25 Mar 2016 17:51:03 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=/xksNVZ2fid/c0nFVM55mIn/ExHR58LIbBBHDnAlhF0=; b=BjdvlLa/5uMgmjEw4sURPiIZ6dl0KN7JjrcmNSPLUrkdeimZ3UNGOLYT6ld59L+Jeh GzQlK9YVhmuJZ0wSa9dktdVtorRKFB9RCeKP0OVwSNvjsxvDEf/FJWKqf/n5oFvIzcHq SCtulZmPwv3R3RvRfIjBf69eqJEkCLYo+LR22Mpzv1T8W8k4I/qJlQclybEawlB76Q0Z BnROBmSLHpkkwnp4/ahg2atZepaqdTLG01OnYczeht1pQdr0r2hZNXDneJBM/Pplu0vH dEciexkB3zT2QhEayvL7cqce1TSFOzPBZuO/TqGlXj2VuLaHcOYrOSZorAdtUrTmJMIm UfFw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=/xksNVZ2fid/c0nFVM55mIn/ExHR58LIbBBHDnAlhF0=; b=ExUrXJT1dZMikjwqGlOrXDzg61Wmt2T+3Ev5IQ6RGa8zl8RV0Qhj+wcwYXU4ulIEkg iXSIITwHiDR78L8aIuycKB8cKx3JUkVHN6Y19b3DW/SG0byT5aqCJABy9H6A+wvByNNd uBsygw08yf5uZYUqbeqUg9M5h4oxmlQipr1A6AnjzaCoSzJz7TsdMFbBR4gR/u/p8cQt XVMgH+NZO0INLmLn4f1xs+XIruL8e0uEmT7GoloejPjOAZlMp3x4FOknQsFYKK/Firqa B3mtTqgnR9SlXUhCDclpxz8abEZpunwzQGd+wyykO036DXegLPkBUt7AK1QByXz1TFwJ uA2g==
X-Gm-Message-State: AD7BkJL5sR/TX19KOnBqlATbXGY9FMAX0mAcEJvsRQ7STjWpvbR+q5C/3mGWOx48H1HNQg==
X-Received: by 10.98.19.82 with SMTP id b79mr24952256pfj.17.1458953463164; Fri, 25 Mar 2016 17:51:03 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id fk9sm18929914pad.9.2016.03.25.17.51.00 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 17:51:02 -0700 (PDT)
To: "Michael Behringer (mbehring)" , "anima@ietf.org" 
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F5DD00.9090809@gmail.com>
Date: Sat, 26 Mar 2016 13:51:12 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: [Anima] 5. Functional Overview [was Reference Model Issues]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 00:51:05 -0000

On 25/03/2016 06:52, Michael Behringer (mbehring) wrote:
...
> 5. Functional Overview
> - title is not good. This section really describes how an autonomic node behaves. maybe call it "Behaviour of an autonomic node"? 
> - we need to describe the bring-up better, specifically insecure discovery, 
>   ACP negotiation, ACP bringup, and subsequent operations, and which protocols to use
>   where. Or should this go into the ACP draft?? 

I think the full description belongs in the ACP draft,and when that is reasonably
stable, there should be a short summary here.

   Brian


From nobody Fri Mar 25 17:57:47 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ABACF12D105 for ; Fri, 25 Mar 2016 17:57:45 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ax9GDcJUdHtY for ; Fri, 25 Mar 2016 17:57:44 -0700 (PDT)
Received: from mail-pf0-x234.google.com (mail-pf0-x234.google.com [IPv6:2607:f8b0:400e:c00::234]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1A8F212D0E5 for ; Fri, 25 Mar 2016 17:57:44 -0700 (PDT)
Received: by mail-pf0-x234.google.com with SMTP id u190so93067003pfb.3 for ; Fri, 25 Mar 2016 17:57:44 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=HDV6J8qj9bufuTOmKp0g79LCoB5DweYKFY29VmSTPtg=; b=HDF8aUJjkNWpqJa+G8LVA3oatggQRoVEacBk1zP6Bm3UOutXi4GewYqNv73msZ43Us +bAh+aSdpzolYDHapiW4yHzSgdpYWwGGUTp1WfLXJQADJtqhnthT7AIy8+Tt7v5c/N/M Tu3P50Ie2JpBuU7wFpX2p/cx/7SbivWnab+vUz5rHl7jEvHujJ4iFrNRJWXic7dQg/Ol xHpedExYTyniKDRQUihctIUeXyZdVPBYzfIyOgN9uUPaE47zFtgRXLpZ7Gfkb4JWjhLL gASY9l1ewNnluNv9WFRSq4ds74B9spI4q+muA8j4G5WjUFmeCFPF159Vxb9dMlvA0iEE cC8w==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=HDV6J8qj9bufuTOmKp0g79LCoB5DweYKFY29VmSTPtg=; b=GWh69BbZYnxfgf2e6HgZ01RJA6PTTHy+Bvlo6y/dR9weWJ45fmfotkAr1oq++N13lC 6IQkJaoXpkGO47q+Ve0g1sYA96IYeqA04yrYjSD6JmexvxpaEc0kja827agWvyfWh9au cSDsCg1Y5y6qsbVTQHCbECxvB+E7at18hNcSJAb65RFXXDTSEMGSatcJy9WlMD+qBLKa UMzKMcqyO+Orx6nYfKkiEVsjCym+Ro+Xc3RgHlXfR0gcVfnPGdPeqjWY38ipaBKK6cjs +3InnzykgbM8pfNR8G6fT+DI4Cx/fcv56DjGyDfj76Es2DYrw8kIPQEwx0uXO/Tu6NwH IPow==
X-Gm-Message-State: AD7BkJJ21cLGJUhI3oWUA/1FtYRACngQAHW4CUFhky+xFZmjks8ED+re71O91GAb/acfoQ==
X-Received: by 10.98.74.17 with SMTP id x17mr25076168pfa.14.1458953863711; Fri, 25 Mar 2016 17:57:43 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id e12sm18925601pat.19.2016.03.25.17.57.41 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 17:57:42 -0700 (PDT)
To: "Michael Behringer (mbehring)" , "anima@ietf.org" 
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F5DE90.6010509@gmail.com>
Date: Sat, 26 Mar 2016 13:57:52 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: [Anima] 7.2 Specific ASAs [was Reference Model Issues]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 00:57:45 -0000

On 25/03/2016 06:52, Michael Behringer (mbehring) wrote:
...
> 7.2 Specific ASAs [for the Enrolment Process]
> - should explain MUST implement

And that they MUST be present in all nodes by default. In particular
the Proxy ASA MUST discover whether it needs to be active (see discussion
yesterday). I suppose the same applies to the Registrar ASA but
possibly that needs to be configured "on" by the NOC?

   Brian


From nobody Fri Mar 25 17:59:29 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B592D12D83D for ; Fri, 25 Mar 2016 17:59:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id r3TNVyKN-NuE for ; Fri, 25 Mar 2016 17:59:26 -0700 (PDT)
Received: from mail-pf0-x234.google.com (mail-pf0-x234.google.com [IPv6:2607:f8b0:400e:c00::234]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3A35112D0E5 for ; Fri, 25 Mar 2016 17:59:26 -0700 (PDT)
Received: by mail-pf0-x234.google.com with SMTP id u190so93086872pfb.3 for ; Fri, 25 Mar 2016 17:59:26 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=f1I2jnU0iyvt54IULkbUBaapuopD8BGNcPVLaXMtcDU=; b=IV/03OmViEfcucsQbFiMXYO8es8vXKb/STMAwmApcxljO2XWDbhPPx81ZiSOXqMafD daD3aXEn93B+M+fu/urBA3y+5BbtQTgKUdO3S3xl+D2fBJL97OVw2RhUs0RFuGGqn3kW rA9sV8ZBYw6AxPPmHe5NznULUQIVBgVkGwqsTlDv3gxCPQ4yYHbu6a0oGAna66JO2lfW 1XGdK8XF7aR0dcG7C8JWEckwjSArn4lhnmkjzYFioKV8Z/q2LibtGICjwClA37TIlwvx 7G/V6IJD7PxLG/lwlwJRWxec2SoMbD11Hb18GcHBvFEm1cKoUb+dgOPZe89c5hjWiwli ZfOw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=f1I2jnU0iyvt54IULkbUBaapuopD8BGNcPVLaXMtcDU=; b=EbxhC9VuPyHwDaD5pTLXEw4j9pUYkeb5T8/y+OLbwxyaihZ+6xJI53KIxN63z9wk2D 4nCdmIGbkGj7xNoIPUTPYihrPOownYqaqZ14pO8Ge0yTYkolGu6XTmksVgE0UPnlH8wN njF8zQY1XvIFv+mLzDtZ4Fz3yEikruWN4QQop2gv+VCBx0gu8nieqzN5gEOV27QT8aMs oAl3r6e3XI1Onx0xDPi11+yUQJF77FFVXB2DXUmt2cmds9kHsMNy2BxMGjIsdpFR20C9 Xye6cl7nmz4G6p95CoMyXle2G6QkgUxhUJHYprb1/KQh7HSsAD9oq0zCGtlN1ELVZRBm Okag==
X-Gm-Message-State: AD7BkJL7M73lPswBDvu/XKIgcwsKUMxmSKY7CmT6xDP/A3Zu4Hgstd5mXgQIYN3MMImSow==
X-Received: by 10.98.16.210 with SMTP id 79mr24938180pfq.69.1458953965897; Fri, 25 Mar 2016 17:59:25 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id fa3sm18914653pab.45.2016.03.25.17.59.23 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 17:59:24 -0700 (PDT)
To: "Michael Behringer (mbehring)" , "anima@ietf.org" 
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F5DEF6.4080005@gmail.com>
Date: Sat, 26 Mar 2016 13:59:34 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: [Anima] Future Topics [was Reference Model Issues]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 00:59:27 -0000

On 25/03/2016 06:52, Michael Behringer (mbehring) wrote:
...
> General: 
> - should we combine all items that are not part of the current charter into a separate section "Future Topics" or something like that? Then all (*) topic would be nicely contained. Yes, I think I'd prefer that to taking them out completely, or shortening them to next-to-nothing.

Agreed. Could even be an Appendix but we should not lose this material.

   Brian


From nobody Fri Mar 25 18:03:18 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 482F812D0E5 for ; Fri, 25 Mar 2016 18:03:15 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IfRWxEEH_URH for ; Fri, 25 Mar 2016 18:03:13 -0700 (PDT)
Received: from mail-pa0-x233.google.com (mail-pa0-x233.google.com [IPv6:2607:f8b0:400e:c03::233]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3F8C212D5C3 for ; Fri, 25 Mar 2016 18:03:12 -0700 (PDT)
Received: by mail-pa0-x233.google.com with SMTP id fe3so56156738pab.1 for ; Fri, 25 Mar 2016 18:03:12 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=Md+x0ayPNl3w0gRrXZEfQcZttLWSMEldCnd05GTwnTI=; b=S0bl2Tgo0Z7opHcG8kqJrLfU0Pn83Nuyq6syJHW3otj0dc8/VuaFb0/txTY4Xmkf7p 1uTnhw77ev196ikY7P/44qRkhrHcvaAGB03DwWkISF3OhhUBNqB5lsB89tKe4+ER7Co9 TBinEPA8cIuDdxRGwgNFoSCdOB1NFEkU9opZW4lwlZ7LEC7JeTtZJK0fP96Se+Y9Y4dL LS5w4Tvd79NUB442HJ+OhPHLZNpvCtDQ4xDkZfYDEGCYugRkOyS/cyi7DidpWZpc7Twq c3VoIZvMs1D3P01U0Zo3f69jyFUv5jBaeFloLWTETdoqaNxSDJQtU9E77pVomHxk3w/8 mqXw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=Md+x0ayPNl3w0gRrXZEfQcZttLWSMEldCnd05GTwnTI=; b=Ur1KMIOHWduEDhlxyMIDCgr8SR1lz95FFZbJeJUCnryXgyMN2llS95Ty3eJwXp07z2 CUa5mdv3tKid3xVT15MyXZG0NWEczKU0HEyfPG+hlWWR8mHD1rbRh5P7GYoh84HC8psq NotZsNp4Vd5akApiUdNZnbC5uDjyb9PJY5tRlc4eA8dnLmV4Yz15lLsENQwZTuAUn/E2 JdqYK0urA/+xiRXNi+Ml9HnCAklY5pOoHkNb9UP4iwf2/wPs+90Co4LbLMOfxDL36HRk D56RaLlnmzcgrwy3fSY/7jlYIVryu6+2hsb5/q+J5CnuocpN+M29cdnTbW3m3JBIHrSg 4dYQ==
X-Gm-Message-State: AD7BkJJ+jIcMWfuwepLl0/65Y7wmqnoGnmiOQJm/HMgYzMldrYJe15GEnOmnRIh0X9Vh5g==
X-Received: by 10.66.171.169 with SMTP id av9mr24986183pac.36.1458954191880; Fri, 25 Mar 2016 18:03:11 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id v14sm18961555pas.10.2016.03.25.18.03.07 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 18:03:10 -0700 (PDT)
To: "Michael Behringer (mbehring)" , John Strassner , Duzongpeng 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com> 
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F5DFD6.90000@gmail.com>
Date: Sat, 26 Mar 2016 14:03:18 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 01:03:15 -0000

On 26/03/2016 02:40, Michael Behringer (mbehring) wrote:
>> -----Original Message-----
>> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
>> Sent: 24 March 2016 20:33
>> To: Michael Behringer (mbehring) ; John Strassner
>> ; Duzongpeng 
>> Cc: Laurent Ciavaglia ; J=C3=A9ferson Cam=
pos Nobre
>> ; anima@ietf.org; Sheng Jiang
>> 
>> Subject: Re: Another use case for Intent: "Freeze network enrolment"
>>
>> On 25/03/2016 02:49, Michael Behringer (mbehring) wrote:
>>> Maybe we should combine two threads here. I think there is a perfect =
use
>> case for Intent I  mentioned in another mail, which would be even appl=
icable
>> to what we do today:
>>>
>>> "Freeze network enrolment".
>>>
>>> Since any node can be a proxy node for bootstrap, it makes perfect se=
nse
>> to flood this to all nodes.
>>>
>>> Local override possible via traditional config / netconf / etc.
>>>
>>> No parameters needed. Can't get any simpler.
>>
>> Right, but that one sure sounds like a partial update. The same would =
apply
>> to any other form of emergency signal.
>>
>>    Brian
>=20
> Sorry - can you expand? I'm not sure I understand what you're trying to=
 say...
> Michael

I just thought that if the NOC wants to send out some sort of urgent inst=
ruction
to all nodes, it is by its nature a partial update of Intent. Of course a=

partial update can be achieved by sending everything. That's a trade-off =
of
resource vs complexity, I suppose.

    Brian




From nobody Fri Mar 25 18:04:43 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EDEBC12D5C3 for ; Fri, 25 Mar 2016 18:04:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0tDyhVyZgqnl for ; Fri, 25 Mar 2016 18:04:38 -0700 (PDT)
Received: from mail-lf0-x231.google.com (mail-lf0-x231.google.com [IPv6:2a00:1450:4010:c07::231]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CF5DC12D0E5 for ; Fri, 25 Mar 2016 18:04:37 -0700 (PDT)
Received: by mail-lf0-x231.google.com with SMTP id f67so8691574lfb.0 for ; Fri, 25 Mar 2016 18:04:37 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=5QGTcHkMmVb/+al9zi1wdw4FlSy/4fBvXmNT6Kyo2OA=; b=VN87ZDDZh8wDkikV+tRohz/1cKoK+w3FN6Niq+1hf0l/O+zRT0uP3RrLez0vN1Aq4F A3Mn2rAjeP42s4txMuGvQ7pGRRz/dGEzkTPv7Je8ECQgZcxa/ncXPsyu+Sfxk3+W0iz7 4/MhMbcZAwuT/mJ1kfVDQDvaI5x9toSPMYa6TxINOnCB8BGAn0NJ3ZEVW/IzVh153bAE H/QvqCLff3YSmRJAR20k3LeD50afSKns0lvLsGi/jACBk4cgo1fJb5/KsgP/xcNNbKOQ 6Ie/YHwMK4PlLjodaINvFe8FiONNqE93oNLUEUj9ExOeUNoT5s47H2J6bo8zI5g4t0Eo m+sQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=5QGTcHkMmVb/+al9zi1wdw4FlSy/4fBvXmNT6Kyo2OA=; b=DhV+Wd1XqLhYThev5vEQLEJvK0mn+WRQMQ5CMQ8CWcLlEI11Bcg4uE/HVhDdaa8LNh fyO5x215CYXw1/TU0I/08JFNI3VTvpnHOUotc9H07PO5yoJGLizmYiFZIbdBrEiP8c6S JqHJ8wzxFAVsqR+tx3nSsuY/bohRQxETphoHqZ+8DLTqdW4A6gU40uIsPnS5Ttr4RBxM 1N7tsYjhHdCbrFWh/FuZLwwU0krRXq+/Mns/9l3tMe2nLW976BL9taDXv+iD1yge/9pV KUCobuWhOuLCoQwhOxxQpnFbe8f7Cv2sGjQuUsGiAcamdSxLtsuL5i+95K4JpZ4qkmG1 Lc2g==
X-Gm-Message-State: AD7BkJKxbt7HvIvaTHGBZHxRuGsvvxAScxYTtl+q6E8+e4pIS2ZyAp4K2Dau1iHgqmA0612+xPoxS6qIE2awhA==
MIME-Version: 1.0
X-Received: by 10.25.205.7 with SMTP id d7mr6557421lfg.70.1458954275970; Fri, 25 Mar 2016 18:04:35 -0700 (PDT)
Received: by 10.25.22.19 with HTTP; Fri, 25 Mar 2016 18:04:35 -0700 (PDT)
In-Reply-To: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  
Date: Fri, 25 Mar 2016 18:04:35 -0700
Message-ID: 
From: John Strassner 
To: Duzongpeng , John Strassner 
Content-Type: multipart/alternative; boundary=001a1142091a5ce41a052ee943b7
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , "Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 01:04:42 -0000

--001a1142091a5ce41a052ee943b7
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Zongpeng,

Let me try to rephrase. It isn't a question of top-down or bottom-up, but
rather, how complex the user model is. I am recommending a simple user
model. This means that the user should not have to know in advance the list
of specific autonomic functions that are present.

I do agree that we should start simple. One autonomic function is fine.
However, I don't understand why you say that "it is not practical that an
intent writer in ANIMA network do not know the information about autonomic
functions in the autonomic domain". Intent shouldn't be a function call or
an API. It should be expressed in a form natural to the intent writer, and
then translated to a form that a device can consume.

regards,
John

On Fri, Mar 25, 2016 at 1:30 AM, Duzongpeng  wrote:

> Hi, John
>
>
>
> Some personal understandings for discussion.
>
>
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
>
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
>
>
> I think I can understand your idea. However, it looks like a top-down
> method.
>
>
>
> It has been suggested that ANIMA can work by using a bottom-up method.
>
>
>
> I mean perhaps we can make a minimum set for intent right now, and make i=
t
> work first.
>
>
>
> It has been suggested that ANIMA ASAs will be developed in a one-by-one
> procedure, so that currently there is not too much information that neede=
d
> to be carried in the intent.
>
>
>
> I think that is why Michael has said intent should be a small one.
>
>
>
> In the ideal, AFA will finally be transparent for intent. Perhaps, it is
> because most of the need of operator can be expressed using abstract
> languages.
>
>
>
> But in current status, it is not practical that an intent writer in ANIMA
> network do not know the information about autonomic functions in the
> autonomic domain.
>
>
>
> Best Regards
>
> Zongpeng Du
>
>
>
>
>
> *From:* John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* Wednesday, March 23, 2016 4:32 AM
> *To:* Michael Behringer (mbehring); John Strassner
> *Cc:* Duzongpeng; anima@ietf.org; Laurent Ciavaglia; J=C3=A9ferson Campos
> Nobre; Sheng Jiang
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> In general, I agree with Michael. I'd like to emphasize and expand on a
> couple of key points:
>
>
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
>
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J
>
>
>
> 
>
> Agreed. Although to me, this means that we need to consider the types of
> actors that are using intent. For example, most of my CCIE and JCIE frien=
ds
> have no interest in using an intent language for pushing configuration. O=
n
> the other hand, when I talk to operators and application developers, they
> would love to be able to use a high-level intent and have the intent engi=
ne
> take care of the details for them.
>
> 
>
>
>
> It has become clear that in today=E2=80=99s networking, people will want =
to push
> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I th=
ink we=E2=80=99ll just have to acknowledge
> that.
>
>
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
>
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> 
>
> Agreed.
>
> 
>
>
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
>
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
>
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do.
>
>
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
>
>
>
>
> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=
=80=9Cnode 17: here is
> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CLI=
=E2=80=9D. But what
> I describe above COULD be abused in this way. RFC7575 says this is NOT
> Intent.
>
>
>
> 
>
> Absolutely agree!
>
> 
>
>
>
>  regards,
>
> John
>
>
>
>
>
>
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
> mbehring@cisco.com> wrote:
>
> Hi Zongpeng,
>
>
>
> Yes, this is a recurring question and topic. Let me give my view on two
> levels:
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J  It has=
 become clear
> that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writing =
Intent like =E2=80=9Cnode
> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here i=
s your
> CLI=E2=80=9D. But what I describe above COULD be abused in this way. RFC7=
575 says
> this is NOT Intent.
>
>
>
> In other words: Let=E2=80=99s not be too prescriptive, and hope that folk=
s will do
> more or less reasonable things.
>
>
>
> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly co=
ntribute,
> I=E2=80=99m just struggling with priorities at the moment (as you can see=
 by my
> relative silence over the past weeks).
>
>
>
> Michael
>
>
>
>
>
>
>
> *From:* Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Duzongpeng
> *Sent:* 16 March 2016 02:38
> *To:* anima@ietf.org
> *Cc:* Laurent Ciavaglia ; J=C3=A9ferson Camp=
os
> Nobre ; Sheng Jiang 
> *Subject:* [Anima] Does Autonomic network need some intervention beyond
> intent
>
>
>
> Hello, everyone
>
>
>
>          I understand that Autonomic network should be able to work with
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
>
>
>
>          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
>
>
>
>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
>
>
>
>          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
>
>
>
> Is =E2=80=9CASA parameters" ok here?
>
>
>
> I think some characteristics for this kind of low-level policies are
>
> 1.       Network level parameters, need to be distributed by GRASP and AC=
P
>
> 2.       Relatively static compared to intent, perhaps only needed to be
> configured once
>
> 3.       Unavoidable in current network environment, mostly for
> establishing the network infrastructure
>
> 4.       Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
>
>
>
> Welcome for comment. Your suggestions will be appreciated.
>
>
>
>
>
> Best Regards
>
> Zongpeng Du
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> --
>
> regards,
>
> John
>



--=20
regards,
John

--001a1142091a5ce41a052ee943b7
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi Zongpeng,

Let me try to r=
ephrase. It isn't a question of top-down or bottom-up, but rather, how =
complex the user model is. I am recommending a simple user model. This mean=
s that the user should not have to know in advance the list of specific aut=
onomic functions that are present.

I do agree that=
 we should start simple. One autonomic function is fine. However, I don'=
;t understand why you say that "it is not practical that an intent writer in ANIMA network do=
 not know the information about autonomic functions in the autonomic domain=
". Intent shouldn't be a function call or an API. It should=
 be expressed in a form natural to the intent writer, and then translated t=
o a form that a device can consume.

regards,
=

John

On Fri, Mar 25, 2016 at 1:30 AM, Duzongpeng <duzongpeng@huawei.com=
> wrote:











Hi,=
 John


=
=C2=A0


Som=
e personal understandings for discussion.


=
=C2=A0


&l=
t;jcs>


I&=
#39;m not sure what you mean by "function owner". (Sorry to be=



pi=
cky, I'm likely under-caffeinated.)


=C2=A0


Th=
e above seemed like an autonomous decision, made by the parser.


I =
don't think that the intent writer should be cognizant of how many or


wh=
at types of autonomic functions are present.


&l=
t;/jcs>


=
=C2=A0


I t=
hink I can understand your idea. However, it looks like a top-down method.<=
u>


=
=C2=A0


It =
has been suggested that ANIMA can work by using a bottom-up method.<=
u>


=
=C2=A0


I m=
ean perhaps we can make a minimum set for intent right now, and make it wor=
k first.



=
=C2=A0


It =
has been suggested that ANIMA ASAs will be developed in a one-by-one proced=
ure, so that currently there is not too much information that needed
 to be carried in the intent.


=
=C2=A0


I t=
hink that is why Michael has said intent should be a small one.


=
=C2=A0


In =
the ideal, AFA will finally be transparent for intent. Perhaps, it is becau=
se most of the need of operator can be expressed using abstract languages.<=
u>


=
=C2=A0


But=
 in current status, it is not practical that an intent writer in ANIMA netw=
ork do not know the information about autonomic functions in the
 autonomic domain.


=
=C2=A0


Bes=
t Regards


Zon=
gpeng Du


=
=C2=A0


=
=C2=A0




From: John Strassner [mailto:strazpdj@gmail.com]


Sent: Wednesday, March 23, 2016 4:32 AM

To: Michael Behringer (mbehring); John Strassner

Cc: Duzongpeng; =
anima@ietf.org; Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Ji=
ang

Subject: Re: [Anima] Does Autonomic network need some intervention b=
eyond intent




=C2=A0






In=
 general, I agree with Michael. I'd like to emphasize and expand on a c=
ouple of key points:






=C2=A0






=C2=A0






First=
, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =E2=
=80=9Csteering=E2=80=9D a network, at least not
 initially. There is still traditional config, plus all the existing and em=
erging SDN models. My view has always been that ANIMA sets a =E2=80=9Cdefau=
lt=E2=80=9D policy in a network, and all other means are more specific and =
can override intent. This is described in RFC7575.
 Main point here: ANIMA Intent isn=E2=80=99t the only tool in a network. 


=C2=A0






&l=
t;jcs>






Ab=
solutely agree. Note that even in the Congress case I mentioned






ea=
rlier, it is NOT pushing config, and it is NOT viewed as the<=
/span>






&q=
uot;only" policy








&l=
t;/jcs>








=C2=A0






=C2=A0






Inten=
t could perfectly well cover a high level policy such as =E2=80=9Call nodes=
 of type x do
 this; type y does that=E2=80=9D. (Which I think is what you mention below.=
) When we started with Intent, the vision was to keep it very high level, m=
ore along the line =E2=80=9Cdo the right thing=E2=80=9D
J=C2=A0


=C2=A0








&l=
t;jcs>






Ag=
reed. Although to me, this means that we need to consider the types of acto=
rs that are using intent. For example, most of my CCIE and JCIE friends hav=
e no interest
 in using an intent language for pushing configuration. On the other hand, =
when I talk to operators and application developers, they would love to be =
able to use a high-level intent and have the intent engine take care of the=
 details for them.






&l=
t;/jcs>






=C2=A0










It has become clear that in today=E2=80=99s ne=
tworking, people will want to push more =E2=80=9Cparameters=E2=80=9D than =
=E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowledg=
e
 that.






=C2=A0






&l=
t;jcs>






So=
rry Michael, I don't follow here. What do you mean=







by=
 "parameters"?






&l=
t;/jcs>






=C2=A0








=C2=A0






Perso=
nally I think it=E2=80=99ll pan out like this: Intent will have sections pe=
r autonomic
 function. An Intent parser will validate overall correctness (signature, e=
tc), then segment the Intent into its bits: One part will be the autonomic =
fabric itself, then there will be bits for each autonomic function.



=C2=A0






&l=
t;jcs>






Do=
 you mean that the intent writer will **know the difference**=







be=
tween the fabric and each autonomic function? I hope not,






bu=
t perhaps I'm misunderstanding you...






&l=
t;/jcs>






=C2=A0






=C2=A0






So th=
e Intent parser on a node will grab intent, see there is some Intent for au=
tonomic
 function A, will see whether this is running locally, and if so, it=E2=80=
=99ll just pass that piece of Intent to that function.









=C2=A0






&l=
t;jcs>






Ag=
reed.






&l=
t;/jcs>






=C2=A0








=C2=A0






In th=
is way, a function owner defines both the intent itself, and how it is cons=
umed
 by his/her function. 


=C2=A0






&l=
t;jcs>






I&=
#39;m not sure what you mean by "function owner". (Sorry to be=







pi=
cky, I'm likely under-caffeinated.)






=C2=A0






Th=
e above seemed like an autonomous decision, made by the parser.






I =
don't think that the intent writer should be cognizant of how many or






wh=
at types of autonomic functions are present.






&l=
t;/jcs>






=C2=A0






=C2=A0








The downside is that=
 there is no =E2=80=9Ccontrol=E2=80=9D and people can do what they want, in=
 the worst case even push
 config, which is clearly NOT what Intent is meant to do. 


=C2=A0






&l=
t;jcs>






An=
d hence, the "HAL 2000" scenario. :-)






&l=
t;/jcs>






=C2=A0








=C2=A0






I=E2=80=99d hope that people don=E2=80=99t start writing Intent li=
ke =E2=80=9Cnode 17: here is your CLI; node 23: here is your CLI=E2=80=9D;
 node 37: here is your CLI=E2=80=9D. But what I describe above COULD be abu=
sed in this way. RFC7575 says this is NOT Intent.







=C2=A0






&l=
t;jcs>






Ab=
solutely agree!






&l=
t;/jcs>






=C2=A0


















=C2=
=A0regards,


John<=
/span>


=
=C2=A0


=C2=A0








=C2=A0




On=
 Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <mbehring@cisco.com> wro=
te:






Hi Zo=
ngpeng,



=C2=
=A0


Yes, =
this is a recurring question and topic. Let me give my view on two levels:



=C2=
=A0


First=
, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =E2=
=80=9Csteering=E2=80=9D a network, at least not
 initially. There is still traditional config, plus all the existing and em=
erging SDN models. My view has always been that ANIMA sets a =E2=80=9Cdefau=
lt=E2=80=9D policy in a network, and all other means are more specific and =
can override intent. This is described in RFC7575.
 Main point here: ANIMA Intent isn=E2=80=99t the only tool in a network. 


=C2=
=A0


Inten=
t could perfectly well cover a high level policy such as =E2=80=9Call nodes=
 of type x do
 this; type y does that=E2=80=9D. (Which I think is what you mention below.=
) When we started with Intent, the vision was to keep it very high level, m=
ore along the line =E2=80=9Cdo the right thing=E2=80=9D
J=C2=A0 It has become clear that in today=E2=80=99s net=
working, people will want to push more =E2=80=9Cparameters=E2=80=9D than =
=E2=80=9Cpolicy=E2=80=9D,
 and I think we=E2=80=99ll just have to acknowledge that. 


=C2=
=A0


Perso=
nally I think it=E2=80=99ll pan out like this: Intent will have sections pe=
r autonomic
 function. An Intent parser will validate overall correctness (signature, e=
tc), then segment the Intent into its bits: One part will be the autonomic =
fabric itself, then there will be bits for each autonomic function.



=C2=
=A0


So th=
e Intent parser on a node will grab intent, see there is some Intent for au=
tonomic
 function A, will see whether this is running locally, and if so, it=E2=80=
=99ll just pass that piece of Intent to that function.



=C2=
=A0


In th=
is way, a function owner defines both the intent itself, and how it is cons=
umed
 by his/her function. 


=C2=
=A0


The d=
ownside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can do wha=
t they want, in the
 worst case even push config, which is clearly NOT what Intent is meant to =
do. I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =
=E2=80=9Cnode 17: here is your CLI; node 23: here is your CLI=E2=80=9D; nod=
e 37: here is your CLI=E2=80=9D. But what I describe above COULD be abused
 in this way. RFC7575 says this is NOT Intent. =



=C2=
=A0


In ot=
her words: Let=E2=80=99s not be too prescriptive, and hope that folks will =
do more or less
 reasonable things. 


=C2=
=A0


I=E2=
=80=99ve been meaning to comment on the Intent drafts and possibly contribu=
te, I=E2=80=99m just
 struggling with priorities at the moment (as you can see by my relative si=
lence over the past weeks).



=C2=
=A0


Micha=
el


=C2=
=A0


=C2=
=A0


=C2=
=A0








From: Anima [mailto:anima-bounces@ietf.org]
On Behalf Of Duzongpeng

Sent: 16 March 2016 02:38

To: anima@ietf.o=
rg

Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; J=C3=A9ferson C=
ampos Nobre <j=
cnobre@inf.ufrgs.br>; Sheng Jiang <jiangsheng@huawei.com>

Subject: [Anima] Does Autonomic network need some intervention beyon=
d intent






=C2=A0


Hello, everyone


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 I understand that Autonomic network should be able=
 to work with as less configuration as
 possible. In the ideal case, an Autonomic network should be able to work w=
ell while only intent is needed from human operators.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 However, I wonder whether Autonomic network may ne=
ed some intervention from human operators
 through some low-level policies beyond intents.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 For example, in the

https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it i=
s suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in IP RAN) can be assigned as an "intent". These configuration pa=
rameters need to be distributed in the autonomic
 domain to influence the detail configurations on each autonomic node.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 As intent is described as an abstract, declarative=
, high-level policy used to operate an
 autonomic domain, such as an enterprise network. Perhaps we can rename the=
 =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix-management ID=E2=80=9D to =
some other words.


=C2=A0



Is =E2=80=9CASA parameters" ok here?




=C2=A0



I think some characteristics for this kind of l=
ow-level policies are


1.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Network level parameters, need to be distribute=
d by GRASP and ACP


2.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Relatively static compared to intent, perhaps o=
nly needed to be configured once<=
/span>


3.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Unavoidable in current network environment, mos=
tly for establishing the network infrastructure=



4.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Rarely need coordinations with others parameter=
s, usually is closed related to one specific autonomic function


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0




Welcome for comment. Your suggestions will be a=
ppreciated.



=C2=A0



=C2=A0


Best Regards


Zongpeng Du











_______________________________________________

Anima mailing list

Anima@ietf.org

h=
ttps://www.ietf.org/mailman/listinfo/anima









-- 






re=
gards,






Jo=
hn
















-- 
regards,
John




--001a1142091a5ce41a052ee943b7--


From nobody Fri Mar 25 18:08:31 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6732012D83D for ; Fri, 25 Mar 2016 18:08:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gIKb33sqgq6e for ; Fri, 25 Mar 2016 18:08:28 -0700 (PDT)
Received: from mail-pa0-x231.google.com (mail-pa0-x231.google.com [IPv6:2607:f8b0:400e:c03::231]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4D3DD12D5C3 for ; Fri, 25 Mar 2016 18:08:28 -0700 (PDT)
Received: by mail-pa0-x231.google.com with SMTP id td3so55952658pab.2 for ; Fri, 25 Mar 2016 18:08:28 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=W9os8gWCwXF9HjxTQUd2w9N28BHkslKl02HGOvXdlYI=; b=A9muOS5ddrEmj0mdnfd3zWNwWJJ2GIZvmdrYnsHJUAO4ENx8aSdtV5ruN8D7Shg/IO RBnPm1gVrUynRaKRPe0P3d7gcmPqbHjiyZcc3hhaOV4o1PDHBdUnvCEift56AjBGWSg1 UQfqs8ZZaPpFPAUHWxEqkWOE9ZaK4ZRmVPxEn+sRFruF6TRlQVSseIEeb/a7JuGXCRtJ 3vZXJusNguBhLBeGZf1Rl0bHs7arZyiln4BijP/BInNTfzZrKsuRsiexh1uMh9QiUxnY 38gjFKC1NYqZ39WN4UcnhRHQZZYTsgROo3LioxgrE0gJ+V+WpTJUvKGU0D+JTtNmlEB9 uMBw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=W9os8gWCwXF9HjxTQUd2w9N28BHkslKl02HGOvXdlYI=; b=lWryIbTn/hPdi/Z8MWpHAMErQccmf8jTZ9O/j8/vs2H4NuHauPO4bRRx+4g6prz9y7 f71Pu9OpjxIDpoPx0PRgXta/ms/XX9sU11CpVOFFvtt/rnXyZXkVO2D7w0OqD3sNmgb1 cZPUSzD3VcwkVqkhSGef1HqgMuHjaybtdmdvkZtOHntgu8bwrIeXodXSgjSMxrMEvblC UyJLNF7TMQ7IBbG+iXDZ++hYFVfdfOTCF/OT1eGtlYClnLnHHQ2FbXznF0HB3fqGhiwq 0JRunrQejuGoHyVFmQD8SP0MxLjZNk8yL8rZljsod3uEESm67YYQxyBWWuoDDHDPQhKu 4s+w==
X-Gm-Message-State: AD7BkJIbZUyX8UJd1+sZqykxM87kM6O+IddKTMMDJ9F9uN64Mwj/Pk5K6W+MQCFbdVAeZQ==
X-Received: by 10.66.132.73 with SMTP id os9mr25051331pab.27.1458954507881; Fri, 25 Mar 2016 18:08:27 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id 19sm18913461pfp.66.2016.03.25.18.08.24 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 18:08:26 -0700 (PDT)
To: Michael Richardson , "anima@ietf.org" 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <56E9A8B2.8070504@gmail.com>  <56EEF72A.5050309@gmail.com> <485.1458564774@obiwan.sandelman.ca> <56F0BDB4.1040904@gmail.com> <56F35918.4080701@gmail.com> <11715.1458937069@obiwan.sandelman.ca>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F5E114.4000400@gmail.com>
Date: Sat, 26 Mar 2016 14:08:36 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <11715.1458937069@obiwan.sandelman.ca>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: Re: [Anima] Who's a proxy? [was: need for some ASA definitions definitions to help bootstrap]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 01:08:30 -0000

Ack to all that. One minor comment below...
On 26/03/2016 09:17, Michael Richardson wrote:
> 
> Brian E Carpenter  wrote:
>     >> That doesn't really seem right, because it assumes that the NOC has
>     >> perfect knowledge of the topology. Potentially, any node with multiple
>     >> link interfaces could be a proxy, and if the topology is
>     >> non-redundant, they would all need to be proxies.
> 
>     > This has been niggling at me.
> 
>     > 1. Does it matter if there is more than one proxy on a given LAN?
> 
> No, it's fine.
> What matters is that there is at least one *legitimate* proxy on a given LAN.
> 
> That means that a potential proxy can not just look for the *insecure* proxy
> announcement (it could be an attacker), it has to communicate securely to
> indicate to other proxies to indicate it is legit.  Or, just always become a
> proxy.
> 
>     >    1. Do I have more than one LAN interface?  If no, I cannot be a
>     > proxy; exit 2.
> 
> All machines that participate in the ACP have more than one interface!
> Okay, change that to more than one physical interface...
> 
> I can see your point that if a device has only one physical interface
> (accounting for multiple wifi channels and multiple lambdas...) that being a
> proxy is pointless... the other device at the other end of the near ACP hop
> ought to be legitimate and can operate a proxy.

Yes, exactly. It's exactly the same for the question whether a node
acts as a GRASP relay: only if it has multiple physical links.

Now, I just need a socket property that tells me "this interface is
a real one." Seriously, that's a bit of an issue, although Michael B's
proposed adjacency table for the ACP should fix it.

    Brian


From nobody Fri Mar 25 18:41:18 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 08BC212D09C for ; Fri, 25 Mar 2016 18:41:18 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xgm7LHW8ruyv for ; Fri, 25 Mar 2016 18:41:14 -0700 (PDT)
Received: from mail-lb0-x22c.google.com (mail-lb0-x22c.google.com [IPv6:2a00:1450:4010:c04::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id DB36012D138 for ; Fri, 25 Mar 2016 18:41:13 -0700 (PDT)
Received: by mail-lb0-x22c.google.com with SMTP id vo2so3166697lbb.1 for ; Fri, 25 Mar 2016 18:41:13 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=42hAIqQLR4MbEkIDcHKxGrcvdEWbsbipxoFTexo+cRw=; b=YQgiOKBsfNx/4GV8+85rifiCSQCBL7BoJe3ZDEwycrz4BGxK5Vbw8F/59ImSCEnqi+ swSgYGyDRc9uLO0jqT4N2Li3qI3XAjveJppEcgXc1U4xaFFNtlhH+HGasA5euMQ/g+g3 jGxtBbOIyiNl/ct7LIWY6dPhezy/Iw4lhwtKQt35Yl+9wRm5aP3jlNhFfAMRGqFepB5g Ddk+1DAOcY7Rd+LWuDfT6F1mioQOcGOVpCfM6zxBzIYEvZO7jX2cuYVSH9+CsR1lsQIr 1jTK3KoTBlqCGPjJV058ADTLX34aTtltfzgoehrWypL3UmRPiGPAVCU7xNJ77yP/vwom kAqQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=42hAIqQLR4MbEkIDcHKxGrcvdEWbsbipxoFTexo+cRw=; b=LtdoaB/ady2PLRUf8iZIuHnnuPPwLQJSutudiSTANoKjKx9sf7Ty+NWxI/+F5l/MnD rX4s8bBxFPzSN2bV+L+67mQXJnwAbOcqglrpUp3paAFzbtowkadkiPM5dvaOOLgduzyU HFd63ldO+sQjQ1RzVZqebQZjVpD7C1VA5vJ7f6NVzQjf3QECUJy0AL7HXfiG5/eCbQ5e 8KE+XYqZS3/BzsNHAt7pbbEMDql+nhPt+A6Tds31siIH1oAYJ5tUiTs5c7PXzH7/FJV8 Gp1WCb02nHvMmZSsdP7rN5WrfgRQZ3zfVwdR1luU5jl6iQ4N3M/QsxqFIilK+yk/YoY0 Leog==
X-Gm-Message-State: AD7BkJJU4uZm8zOKw2C7C0YYIzD8Q19DpIB84ihwcFwG1yYwxurGi71+Q9xYgqzXVAgHxdY3gpR5MfMy53mb0Q==
MIME-Version: 1.0
X-Received: by 10.112.147.101 with SMTP id tj5mr5284436lbb.119.1458956471815;  Fri, 25 Mar 2016 18:41:11 -0700 (PDT)
Received: by 10.25.22.19 with HTTP; Fri, 25 Mar 2016 18:41:11 -0700 (PDT)
In-Reply-To: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>  
Date: Fri, 25 Mar 2016 18:41:11 -0700
Message-ID: 
From: John Strassner 
To: Duzongpeng , John Strassner 
Content-Type: multipart/alternative; boundary=047d7b3a8bd23ed46c052ee9c6fa
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , "Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 01:41:18 -0000

--047d7b3a8bd23ed46c052ee9c6fa
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Zongpeng,

I think we have a misunderstanding here.


But I do not support that the intent writer should look every detail in the
catalog.


Most catalogs that I have seen have very few details. They instead tell the
user what services or products are available.


The intent writer may mainly care about what they want, but how to do it
can be left to the autonomic network to deal with.


I agree completely.


Perhaps several solutions exist, and autonomic management system choose one
according to the abilities collected.

This is one of the reasons why I would like to see capabilities (and needs)
advertised.


...

I mean in current status, some of the Autonomic Functions should be exposed
to the network operator, and some of the Autonomic Functions can be hidden
to the network operator

...


How is this different than running a script, or invoking an API? There are
plenty of southbound APIs (from a management function to a device). There
are very few northbound APIs. That's what the operators that I have talked
to want.

regards,
John

On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng  wrote:

> Hi John
>
>
>
> Some personal understanding here. If any misunderstanding, please correct
> me. Thanks.
>
>
>
> 
> I would suggest a slight modification, to enable users of
> intent (and even intent writers) to NOT have to have
> detailed implementation information.
>
> What if all of the Autonomic Functions, after they joined
> the network, advertised both their capabilities as well as
> their needs? We can put aside the latter (needs) as this
> is more complex, but if the capabilities are advertised,
> then the autonomic management system can assemble
> those needs into a catalog. Then, the intent writer can
> look at the catalog and write a declarative request to
> use the functionality of the catalog.
>
> I will look at your YANG model in a bit (sorry, I'm woefully
>
> behind...).
> < /jcs2>
>
>
>
> I agree that autonomic node should have that self-advertisement ability,
> and autonomic management system should have this catalog.
>
>
>
> But I do not support that the intent writer should look every detail in
> the catalog.
>
>
>
> The intent writer may mainly care about what they want, but how to do it
> can be left to the autonomic network to deal with.
>
> Perhaps several solutions exist, and autonomic management system choose
> one according to the abilities collected.
>
>
>
> In the ideal, the network operator may communicate with the autonomic
> network using network-level level intent and network-level report.
>
> I mean in current status, some of the Autonomic Functions should be
> exposed to the network operator, and some of the Autonomic Functions can =
be
> hidden to the network operator
>
> Perhaps in future, when the network nodes become more intelligent, the
> hidden part will increase.
>
>
>
>
>
> Best regards
>
> Zongpeng Du
>
>
>
> *From:* Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *John
> Strassner
> *Sent:* Thursday, March 24, 2016 10:41 AM
> *To:* Michael Behringer (mbehring); John Strassner
> *Cc:* Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org;
> J=C3=A9ferson Campos Nobre
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> Hi Michael,
>
>
>
> thanks for your reply. I do think that we largely agree, and would enjoy
> chatting during or (hopefully before) BA. Please see ... Als=
o
> trying to unclutter, so I am eliding points that we agree on.
>
>
>
>
>
> While *CIEs don=E2=80=99t like magic, they do understand the concept of a=
 default
> behaviour which they can influence if needed. Any reasonable network
> management approach tries to define =E2=80=9Cdefault=E2=80=9D behaviour a=
nd =E2=80=9Cexceptions=E2=80=9D.
> The AN story fills the =E2=80=9Cdefault=E2=80=9D part very nicely, and *C=
IEs get that.
>
> 
>
> Agreed!
>
> 
>
>
>
> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussio=
n: My view is that we shouldn=E2=80=99t
> try to define exactly **what** intent allows and doesn=E2=80=99t allow, s=
ince
> we=E2=80=99ve heard from many sides that the boundary is not clear, and i=
t may be
> impossible to have a really clear, usable definition. So my point was:
> Let=E2=80=99s focus on the distribution model: Intent is flooded to all n=
odes.
> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific nodes=
, etc,
> because that would blur the boundary even worse.
> 
> I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1 wa=
s
> 2006, almost 10 years ago), we used a message bus and pub-sub. In v2, we
> flooded; in v3, we went back to pub-sub because we wanted to do some
> optimization and auditing. But I would be happy to try some type of
> controlled flooding again.
> 
>
> My suggestion: Intent is flooded to all nodes in a domain. The ASA owners
> define the content of what they want to push in their bit of the Intent. =
If
> we insist of full flooding, we encourage the right behaviour. And if
> someone really puts into his Intent =E2=80=9Cnode A: do this; node B: do =
the other;
> node C: do the third=E2=80=9D it is awkward enough for people to take a s=
tep back
> and think whether they=E2=80=99re doing the right thing J
> 
> I agree in principle with what you said. Plus, see above.
> < /jcs2>
>
> 
> Do you mean that the intent writer will **know the difference**
> between the fabric and each autonomic function? I hope not,
> but perhaps I'm misunderstanding you...
> < /jcs>
> Not sure I understand your concern. Yes, in my model there is a section
> for the fabric, and sections for the various functions. Seems a natural
> structure?!? 
> The intent **writer** would have to know implementation
> details in order to place everything in a single file. I would
> rather not require this, and instead, have the intent engine
> (e.g., compiler and associated logic) do this.
> < /jcs2>
>
> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structur=
e Intent around Autonomic
> Functions. This is not optimal from a scientific point of view, but align=
s
> very well with the way how things are organised. Probably we need to
> discuss that in BA. But refer to my YANG model sent a few days ago.
> 
> I would suggest a slight modification, to enable users of
> intent (and even intent writers) to NOT have to have
> detailed implementation information.
>
> What if all of the Autonomic Functions, after they joined
> the network, advertised both their capabilities as well as
> their needs? We can put aside the latter (needs) as this
> is more complex, but if the capabilities are advertised,
> then the autonomic management system can assemble
> those needs into a catalog. Then, the intent writer can
> look at the catalog and write a declarative request to
> use the functionality of the catalog.
>
> I will look at your YANG model in a bit (sorry, I'm woefully
>
> behind...).
> < /jcs2>
>
>
>
>
>
> regards,
>
> John
>
>
>
> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <
> mbehring@cisco.com> wrote:
>
> John,
>
>
>
> I think we=E2=80=99re on the same page. Couple of points, top posting her=
e,
> otherwise it gets too messy.
>
>
>
> While *CIEs don=E2=80=99t like magic, they do understand the concept of a=
 default
> behaviour which they can influence if needed. Any reasonable network
> management approach tries to define =E2=80=9Cdefault=E2=80=9D behaviour a=
nd =E2=80=9Cexceptions=E2=80=9D.
> The AN story fills the =E2=80=9Cdefault=E2=80=9D part very nicely, and *C=
IEs get that.
>
>
>
> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussio=
n: My view is that we shouldn=E2=80=99t
> try to define exactly **what** intent allows and doesn=E2=80=99t allow, s=
ince
> we=E2=80=99ve heard from many sides that the boundary is not clear, and i=
t may be
> impossible to have a really clear, usable definition. So my point was:
> Let=E2=80=99s focus on the distribution model: Intent is flooded to all n=
odes.
> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific nodes=
, etc,
> because that would blur the boundary even worse.
>
>
>
> My suggestion: Intent is flooded to all nodes in a domain. The ASA owners
> define the content of what they want to push in their bit of the Intent. =
If
> we insist of full flooding, we encourage the right behaviour. And if
> someone really puts into his Intent =E2=80=9Cnode A: do this; node B: do =
the other;
> node C: do the third=E2=80=9D it is awkward enough for people to take a s=
tep back
> and think whether they=E2=80=99re doing the right thing J
>
>
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
> Not sure I understand your concern. Yes, in my model there is a section
> for the fabric, and sections for the various functions. Seems a natural
> structure?!?
>
>
>
> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structur=
e Intent around Autonomic
> Functions. This is not optimal from a scientific point of view, but align=
s
> very well with the way how things are organised. Probably we need to
> discuss that in BA. But refer to my YANG model sent a few days ago.
>
>
>
> We should schedule some discussion time outside the ANIMA meeting window.=
..
>
>
>
> Michael
>
>
>
>
>
> *From:* John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* 22 March 2016 21:32
> *To:* Michael Behringer (mbehring) ; John Strassner <
> strazpdj@gmail.com>
> *Cc:* Duzongpeng ; anima@ietf.org; Laurent
> Ciavaglia ; J=C3=A9ferson Campos Nobre <
> jcnobre@inf.ufrgs.br>; Sheng Jiang 
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> In general, I agree with Michael. I'd like to emphasize and expand on a
> couple of key points:
>
>
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
>
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J
>
>
>
> 
>
> Agreed. Although to me, this means that we need to consider the types of
> actors that are using intent. For example, most of my CCIE and JCIE frien=
ds
> have no interest in using an intent language for pushing configuration. O=
n
> the other hand, when I talk to operators and application developers, they
> would love to be able to use a high-level intent and have the intent engi=
ne
> take care of the details for them.
>
> 
>
>
>
> It has become clear that in today=E2=80=99s networking, people will want =
to push
> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I th=
ink we=E2=80=99ll just have to acknowledge
> that.
>
>
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
>
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> 
>
> Agreed.
>
> 
>
>
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
>
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
>
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do.
>
>
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
>
>
>
>
> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=
=80=9Cnode 17: here is
> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CLI=
=E2=80=9D. But what
> I describe above COULD be abused in this way. RFC7575 says this is NOT
> Intent.
>
>
>
> 
>
> Absolutely agree!
>
> 
>
>
>
>  regards,
>
> John
>
>
>
>
>
>
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
> mbehring@cisco.com> wrote:
>
> Hi Zongpeng,
>
>
>
> Yes, this is a recurring question and topic. Let me give my view on two
> levels:
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J  It has=
 become clear
> that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writing =
Intent like =E2=80=9Cnode
> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here i=
s your
> CLI=E2=80=9D. But what I describe above COULD be abused in this way. RFC7=
575 says
> this is NOT Intent.
>
>
>
> In other words: Let=E2=80=99s not be too prescriptive, and hope that folk=
s will do
> more or less reasonable things.
>
>
>
> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly co=
ntribute,
> I=E2=80=99m just struggling with priorities at the moment (as you can see=
 by my
> relative silence over the past weeks).
>
>
>
> Michael
>
>
>
>
>
>
>
> *From:* Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Duzongpeng
> *Sent:* 16 March 2016 02:38
> *To:* anima@ietf.org
> *Cc:* Laurent Ciavaglia ; J=C3=A9ferson Camp=
os
> Nobre ; Sheng Jiang 
> *Subject:* [Anima] Does Autonomic network need some intervention beyond
> intent
>
>
>
> Hello, everyone
>
>
>
>          I understand that Autonomic network should be able to work with
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
>
>
>
>          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
>
>
>
>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
>
>
>
>          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
>
>
>
> Is =E2=80=9CASA parameters" ok here?
>
>
>
> I think some characteristics for this kind of low-level policies are
>
> 1.       Network level parameters, need to be distributed by GRASP and AC=
P
>
> 2.       Relatively static compared to intent, perhaps only needed to be
> configured once
>
> 3.       Unavoidable in current network environment, mostly for
> establishing the network infrastructure
>
> 4.       Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
>
>
>
> Welcome for comment. Your suggestions will be appreciated.
>
>
>
>
>
> Best Regards
>
> Zongpeng Du
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> --
>
> regards,
>
> John
>
>
>
>
> --
>
> regards,
>
> John
>



--=20
regards,
John

--047d7b3a8bd23ed46c052ee9c6fa
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi Zongpeng,

I think we have=
 a misunderstanding here.


But I do no=
t support that the intent writer should look every detail in the catalog.
<jcs>
Most catalogs tha=
t I have seen have very few details. They instead tell the user what servic=
es or products are available.
</jcs>

=
The intent writer may mainly care about what they want, but how to do it ca=
n be left to the autonomic network to deal with. 
<jcs>
I agree completely.
&l=
t;/jcs>

Perhaps several solutions exist, and auton=
omic management system choose one according to the abilities collected.<=
/u>
<jcs>
This is one of t=
he reasons why I would like to see capabilities (and needs) advertised.
=

</jcs>

...
I mean in current status, some of the Auto=
nomic Functions should be exposed to the network operator, and some of the =
Autonomic Functions can be hidden to the network operator
...
<jcs>
How is this =
different than running a script, or invoking an API? There are plenty of so=
uthbound APIs (from a management function to a device). There are very few =
northbound APIs. That's what the operators that I have talked to want.<=
/div>

regards,
John

On Fri, Mar 25, 2016 at 2:31 AM,=
 Duzongpeng <duzongpeng@huawei.com> wrote:











Hi =
John


=
=C2=A0


Som=
e personal understanding here. If any misunderstanding, please correct me. =
Thanks.


=
=C2=A0


&l=
t;jcs2>

I would suggest a slight modification, to enable users of

intent (and even intent writers) to NOT have to have

detailed implementation information.



What if all of the Autonomic Functions, after they joined

the network, advertised both their capabilities as well as

their needs? We can put aside the latter (needs) as this

is more complex, but if the capabilities are advertised,

then the autonomic management system can assemble

those needs into a catalog. Then, the intent writer can

look at the catalog and write a declarative request to

use the functionality of the catalog.



I will look at your YANG model in a bit (sorry, I'm woefully<=
/u>


be=
hind...).

< /jcs2>


=
=C2=A0


I a=
gree that autonomic node should have that self-advertisement ability, and a=
utonomic management system should have this catalog.

=
=C2=A0


But=
 I do not support that the intent writer should look every detail in the ca=
talog.


=
=C2=A0


The=
 intent writer may mainly care about what they want, but how to do it can b=
e left to the autonomic network to deal with.



Per=
haps several solutions exist, and autonomic management system choose one ac=
cording to the abilities collected.


=
=C2=A0


In =
the ideal, the network operator may communicate with the autonomic network =
using network-level level intent and network-level report.


I m=
ean in current status, some of the Autonomic Functions should be exposed to=
 the network operator, and some of the Autonomic Functions can be
 hidden to the network operator


Per=
haps in future, when the network nodes become more intelligent, the hidden =
part will increase.


=
=C2=A0


=
=C2=A0


Bes=
t regards


Zon=
gpeng Du


=
=C2=A0




From: Anima [mailto:anima-bounces@ietf.org]
On Behalf Of John Strassner

Sent: Thursday, March 24, 2016 10:41 AM

To: Michael Behringer (mbehring); John Strassner

Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org; J=C3=A9ferson Campos No=
bre

Subject: Re: [Anima] Does Autonomic network need some intervention b=
eyond intent




=C2=A0




Hi Michael,


=C2=A0


thanks for your reply. I do thi=
nk that we largely agree, and would enjoy chatting during or (hopefully bef=
ore) BA. Please see <jcs2>..</jcs2>. Also trying to
 unclutter, so I am eliding points that we agree on.

=C2=A0


=C2=A0


=
While *CIEs don=E2=80=99t like magic, they do understand the concept of a d=
efault behaviour which they can influence if needed. Any reasonable network
 management approach tries to define =E2=80=9Cdefault=E2=80=9D behaviour an=
d =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=E2=
=80=9D part very nicely, and *CIEs get that.=



<jcs2>


Agreed!

</jcs2>


=C2=A0


The =E2=80=9Cparameter=E2=80=9D versus =E2=80=
=9Cpolicy=E2=80=9D discussion: My view is that we shouldn=E2=80=99t try to =
define exactly *what* intent allows and doesn=E2=80=99t allow,
 since we=E2=80=99ve heard from many sides that the boundary is not clear, =
and it may be impossible to have a really clear, usable definition. So my p=
oint was: Let=E2=80=99s focus on the distribution model: Intent is flooded =
to all nodes. Let=E2=80=99s NOT do selective flooding, unicast
 Intent to specific nodes, etc, because that would blur the boundary even w=
orse.

<jcs2>

I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1 was =
2006, almost 10 years ago), we used a message bus and pub-sub. In v2, we fl=
ooded; in v3, we went back to pub-sub because we wanted to do some optimiza=
tion and auditing. But I would be
 happy to try some type of controlled flooding again.

</jcs2>


My suggestion: Intent is flooded to all nodes =
in a domain. The ASA owners define the content of what they want to push in=
 their bit of the Intent.
 If we insist of full flooding, we encourage the right behaviour. And if so=
meone really puts into his Intent =E2=80=9Cnode A: do this; node B: do the =
other; node C: do the third=E2=80=9D it is awkward enough for people to tak=
e a step back and think whether they=E2=80=99re doing the
 right thing J


<jcs2>

I agree in principle with what you said. Plus, see above.

< /jcs2>


<jcs>

Do you mean that the intent writer will **know the difference**

between the fabric and each autonomic function? I hope not,

but perhaps I'm misunderstanding you...

< /jcs>

Not sure I understand your concern. Yes, in my model there is a section for=
 the fabric, and sections for the various functions. Seems a natural struct=
ure?!?
<jcs2>

The intent **writer** would have to know implementation

details in order to place everything in a single file. I would

rather not require this, and instead, have the intent engine

(e.g., compiler and associated logic) do this.

< /jcs2>




=E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=
=99m suggesting to structure Intent around Autonomic Functions. This is not=
 optimal from a scientific point of view,
 but aligns very well with the way how things are organised. Probably we ne=
ed to discuss that in BA. But refer to my YANG model sent a few days ago.


<jcs2>

I would suggest a slight modification, to enable users of

intent (and even intent writers) to NOT have to have

detailed implementation information.



What if all of the Autonomic Functions, after they joined

the network, advertised both their capabilities as well as

their needs? We can put aside the latter (needs) as this

is more complex, but if the capabilities are advertised,

then the autonomic management system can assemble

those needs into a catalog. Then, the intent writer can

look at the catalog and write a declarative request to

use the functionality of the catalog.



I will look at your YANG model in a bit (sorry, I'm woefully<=
/u>






be=
hind...).

< /jcs2>




=C2=A0


=C2=A0


regards,

John






=C2=A0




On=
 Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <mbehring@cisco.com> wro=
te:






John,



=C2=
=A0


I thi=
nk we=E2=80=99re on the same page. Couple of points, top posting here, othe=
rwise it gets
 too messy.


=C2=
=A0


While=
 *CIEs don=E2=80=99t like magic, they do understand the concept of a defaul=
t behaviour
 which they can influence if needed. Any reasonable network management appr=
oach tries to define =E2=80=9Cdefault=E2=80=9D behaviour and =E2=80=9Cexcep=
tions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=E2=80=9D part very =
nicely, and *CIEs get that.



=C2=
=A0


The =
=E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discussion: My =
view is that we shouldn=E2=80=99t try to define
 exactly *what* intent allows and doesn=E2=80=99t allow, since we=E2=
=80=99ve heard from many sides that the boundary is not clear, and it may b=
e impossible to have a really clear, usable definition. So my point was: Le=
t=E2=80=99s focus on the distribution model: Intent is flooded
 to all nodes. Let=E2=80=99s NOT do selective flooding, unicast Intent to s=
pecific nodes, etc, because that would blur the boundary even worse.



=C2=
=A0


My su=
ggestion: Intent is flooded to all nodes in a domain. The ASA owners define=
 the
 content of what they want to push in their bit of the Intent. If we insist=
 of full flooding, we encourage the right behaviour. And if someone really =
puts into his Intent =E2=80=9Cnode A: do this; node B: do the other; node C=
: do the third=E2=80=9D it is awkward enough for
 people to take a step back and think whether they=E2=80=99re doing the rig=
ht thing 
J



=C2=
=A0


<j=
cs>


Do yo=
u mean that the intent writer will **know the difference**


betwe=
en the fabric and each autonomic function? I hope not,


but p=
erhaps I'm misunderstanding you...


</=
jcs>


=C2=
=A0


Not s=
ure I understand your concern. Yes, in my model there is a section for the =
fabric,
 and sections for the various functions. Seems a natural structure?!? 


=C2=
=A0


=E2=
=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to structure Int=
ent around Autonomic Functions.
 This is not optimal from a scientific point of view, but aligns very well =
with the way how things are organised. Probably we need to discuss that in =
BA. But refer to my YANG model sent a few days ago.



=C2=
=A0


We sh=
ould schedule some discussion time outside the ANIMA meeting window...


=C2=
=A0


Micha=
el


=C2=
=A0


=C2=
=A0








From:
 John Strassner [mailto:strazpdj@gmail.com]


Sent: 22 March 2016 21:32

To: Michael Behringer (mbehring) <mbehring@cisco.com>; John Strassner <strazpdj@gmail.com=
>

Cc: Duzongpeng <duzongpeng@huawei.com>;
anima@ietf.org; Lau=
rent Ciavaglia <laurent.ciavaglia@nokia.com>; J=C3=A9ferson Campos Nobre &l=
t;jcnobre@inf.ufr=
gs.br>;
 Sheng Jiang <jiangsheng@huawei.com>

Subject: Re: [Anima] Does Autonomic network need some intervention b=
eyond intent






=C2=A0<=
/u>






In general, I agree with Michael. I=
9;d like to emphasize and expand on a couple of key points:






=C2=A0<=
/u>






=C2=A0<=
/u>






First=
, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =E2=
=80=9Csteering=E2=80=9D a network, at least not
 initially. There is still traditional config, plus all the existing and em=
erging SDN models. My view has always been that ANIMA sets a =E2=80=9Cdefau=
lt=E2=80=9D policy in a network, and all other means are more specific and =
can override intent. This is described in RFC7575.
 Main point here: ANIMA Intent isn=E2=80=99t the only tool in a network. 


=C2=A0<=
/u>






<jcs>






Absolutely agree. Note that even in th=
e Congress case I mentioned






earlier, it is NOT pushing config, and=
 it is NOT viewed as the





"only" policy








</jcs>








=C2=A0<=
/u>






=C2=A0<=
/u>






Inten=
t could perfectly well cover a high level policy such as =E2=80=9Call nodes=
 of type x do
 this; type y does that=E2=80=9D. (Which I think is what you mention below.=
) When we started with Intent, the vision was to keep it very high level, m=
ore along the line =E2=80=9Cdo the right thing=E2=80=9D
J=C2=A0


=C2=A0<=
/u>








<jcs>






Agreed. Although to me, this means tha=
t we need to consider the types of actors that are using intent. For exampl=
e,
 most of my CCIE and JCIE friends have no interest in using an intent langu=
age for pushing configuration. On the other hand, when I talk to operators =
and application developers, they would love to be able to use a high-level =
intent and have the intent engine
 take care of the details for them.






</jcs>






=C2=A0<=
/u>










It has become cle=
ar that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D,
 and I think we=E2=80=99ll just have to acknowledge that.






=C2=A0<=
/u>






<jcs>






Sorry Michael, I don't follow here=
. What do you mean






by "parameters"?






</jcs>






=C2=A0<=
/u>








=C2=A0<=
/u>






Perso=
nally I think it=E2=80=99ll pan out like this: Intent will have sections pe=
r autonomic
 function. An Intent parser will validate overall correctness (signature, e=
tc), then segment the Intent into its bits: One part will be the autonomic =
fabric itself, then there will be bits for each autonomic function.



=C2=A0<=
/u>






<jcs>






Do you mean that the intent writer wil=
l **know the difference**<=
/p>





between the fabric and each autonomic =
function? I hope not,






but perhaps I'm misunderstanding y=
ou...






</jcs>






=C2=A0<=
/u>






=C2=A0<=
/u>






So th=
e Intent parser on a node will grab intent, see there is some Intent for au=
tonomic
 function A, will see whether this is running locally, and if so, it=E2=80=
=99ll just pass that piece of Intent to that function.









=C2=A0<=
/u>






<jcs>






Agreed.=







</jcs>






=C2=A0<=
/u>








=C2=A0<=
/u>






In th=
is way, a function owner defines both the intent itself, and how it is cons=
umed
 by his/her function. 


=C2=A0<=
/u>






<jcs>






I'm not sure what you mean by &quo=
t;function owner". (Sorry to be<=
/u>






picky, I'm likely under-caffeinate=
d.)






=C2=A0<=
/u>






The above seemed like an autonomous de=
cision, made by the parser.






I don't think that the intent writ=
er should be cognizant of how many or=







what types of autonomic functions are =
present.






</jcs>






=C2=A0<=
/u>






=C2=A0<=
/u>








The downside is that=
 there is no =E2=80=9Ccontrol=E2=80=9D and people can do what they want, in=
 the worst case even push
 config, which is clearly NOT what Intent is meant to do. 


=C2=A0<=
/u>






<jcs>






And hence, the "HAL 2000" sc=
enario. :-)






</jcs>






=C2=A0<=
/u>








=C2=A0<=
/u>






I=E2=80=99d hope tha=
t people don=E2=80=99t start writing Intent like =E2=80=9Cnode 17: here is =
your CLI; node 23: here is your
 CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D. But what I describe abov=
e COULD be abused in this way. RFC7575 says this is NOT Intent.







=C2=A0<=
/u>






<jcs>






Absolutely agree!






</jcs>






=C2=A0<=
/u>


















=C2=
=A0regards,


John<=
/span>


=C2=A0


=C2=A0<=
/u>








=C2=A0<=
/u>




On Wed, Mar 16, 2016 at 1:16 AM, Micha=
el Behringer (mbehring) <mbehring@cisco.com>
 wrote:








Hi Zo=
ngpeng,



=C2=
=A0


Yes, =
this is a recurring question and topic. Let me give my view on two levels:



=C2=
=A0


First=
, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =E2=
=80=9Csteering=E2=80=9D a network, at least not
 initially. There is still traditional config, plus all the existing and em=
erging SDN models. My view has always been that ANIMA sets a =E2=80=9Cdefau=
lt=E2=80=9D policy in a network, and all other means are more specific and =
can override intent. This is described in RFC7575.
 Main point here: ANIMA Intent isn=E2=80=99t the only tool in a network. 


=C2=
=A0


Inten=
t could perfectly well cover a high level policy such as =E2=80=9Call nodes=
 of type x do
 this; type y does that=E2=80=9D. (Which I think is what you mention below.=
) When we started with Intent, the vision was to keep it very high level, m=
ore along the line =E2=80=9Cdo the right thing=E2=80=9D
J=C2=A0 It has become clear that in today=E2=80=99s net=
working, people will want to push more =E2=80=9Cparameters=E2=80=9D than =
=E2=80=9Cpolicy=E2=80=9D,
 and I think we=E2=80=99ll just have to acknowledge that. 


=C2=
=A0


Perso=
nally I think it=E2=80=99ll pan out like this: Intent will have sections pe=
r autonomic
 function. An Intent parser will validate overall correctness (signature, e=
tc), then segment the Intent into its bits: One part will be the autonomic =
fabric itself, then there will be bits for each autonomic function.



=C2=
=A0


So th=
e Intent parser on a node will grab intent, see there is some Intent for au=
tonomic
 function A, will see whether this is running locally, and if so, it=E2=80=
=99ll just pass that piece of Intent to that function.



=C2=
=A0


In th=
is way, a function owner defines both the intent itself, and how it is cons=
umed
 by his/her function. 


=C2=
=A0


The d=
ownside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can do wha=
t they want, in the
 worst case even push config, which is clearly NOT what Intent is meant to =
do. I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =
=E2=80=9Cnode 17: here is your CLI; node 23: here is your CLI=E2=80=9D; nod=
e 37: here is your CLI=E2=80=9D. But what I describe above COULD be abused
 in this way. RFC7575 says this is NOT Intent. =



=C2=
=A0


In ot=
her words: Let=E2=80=99s not be too prescriptive, and hope that folks will =
do more or less
 reasonable things. 


=C2=
=A0


I=E2=
=80=99ve been meaning to comment on the Intent drafts and possibly contribu=
te, I=E2=80=99m just
 struggling with priorities at the moment (as you can see by my relative si=
lence over the past weeks).



=C2=
=A0


Micha=
el


=C2=
=A0


=C2=
=A0


=C2=
=A0








From: Anima [mailto:anima-bounces@ietf.org]
On Behalf Of Duzongpeng

Sent: 16 March 2016 02:38

To: anima@ietf.or=
g

Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; J=C3=A9ferson
 Campos Nobre <=
jcnobre@inf.ufrgs.br>; Sheng Jiang <jiangsheng@huawei.com>

Subject: [Anima] Does Autonomic network need some intervention beyon=
d intent






=C2=A0


Hello, everyone


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 I understand that Autonomic network should be able=
 to work with as less configuration as
 possible. In the ideal case, an Autonomic network should be able to work w=
ell while only intent is needed from human operators.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 However, I wonder whether Autonomic network may ne=
ed some intervention from human operators
 through some low-level policies beyond intents.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 For example, in the
https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00<=
/a>,
 it is suggested that the prefix lengths for the CSG, ASG, RSG (different r=
oles in IP RAN) can be assigned as an "intent". These configurati=
on parameters need to be distributed in the autonomic domain to influence t=
he detail configurations on each autonomic
 node.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 As intent is described as an abstract, declarative=
, high-level policy used to operate an
 autonomic domain, such as an enterprise network. Perhaps we can rename the=
 =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix-management ID=E2=80=9D to =
some other words.


=C2=A0



Is =E2=80=9CASA parameters" ok here?




=C2=A0



I think some characteristics for this kind of l=
ow-level policies are


1.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Network level parameters, need to be distribute=
d by GRASP and ACP


2.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Relatively static compared to intent, perhaps o=
nly needed to be configured once<=
/span>


3.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Unavoidable in current network environment, mos=
tly for establishing the network infrastructure=



4.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Rarely need coordinations with others parameter=
s, usually is closed related to one specific autonomic function


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0




Welcome for comment. Your suggestions will be a=
ppreciated.



=C2=A0



=C2=A0


Best Regards


Zongpeng Du










_______________________________________________

Anima mailing list

Anima@ietf.org

h=
ttps://www.ietf.org/mailman/listinfo/anima<=
u>












-- 






regards,






John





















-- 






re=
gards,






Jo=
hn
















-- 
regards,
John




--047d7b3a8bd23ed46c052ee9c6fa--


From nobody Fri Mar 25 18:41:57 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1D45312D57A for ; Fri, 25 Mar 2016 18:41:56 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GtufB-66-bJo for ; Fri, 25 Mar 2016 18:41:54 -0700 (PDT)
Received: from mail-pf0-x22c.google.com (mail-pf0-x22c.google.com [IPv6:2607:f8b0:400e:c00::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 94D6A12D10E for ; Fri, 25 Mar 2016 18:41:54 -0700 (PDT)
Received: by mail-pf0-x22c.google.com with SMTP id 4so93554326pfd.0 for ; Fri, 25 Mar 2016 18:41:54 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=wJIVbSBBOwtd+RAzX1RySBNmKbHBHYHJMQyabSIq1wA=; b=bwKHTBWMTkAIqWMm+ipZdUQsIGxwrT5jhfCdPVur+n3BReVfLzIJjYb3tn9iQaCNkn Qr1/IhBNWDTmUZm3xIY0Z0DWRuqSluz9lNvYoEX+xjgonnNUGBPXTJEFKVM/z8cd0I71 0xqH9U3KoAi9IC0Wj/AYjO6wJ+fKpEEIZoCcP7Rj5dkv9YbOlSum+OZYuvuQPxh7+wtn Es3ZOtiF11vilY6WjFwE2FVOK4aJmMn6EkT+of8vXYCyvI44dHu/DqT7V7g3JZRzuKzc HDjMDo5rkBzk6KenxzjNDltmdYGhMOpRpbOW9kieHyi9UC7VNZ+hO2irrhNq4nIzSud6 3w+g==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=wJIVbSBBOwtd+RAzX1RySBNmKbHBHYHJMQyabSIq1wA=; b=AUhjS/QQeR3dvmJnXXxypGB6wQ+cGZZFZ2Pixh6YHoOL0AIdPSsH1dD5ejERAM/dUv YfBbTiCoTHQoCKinY9NCiLtAFM6BQ5omg0F13LTRTJjm2vKs8zCqZIRTOpAm/qOJ40QI v8Kz0qySKDAWJUn5cmjCpmLIXz4vR7afPG52mg6MOwNrBlKKDlMUDplGDtOlhY8eqk97 jGxgRyUMoBAgqIejVB4saemwHqLzqb/VsxK3YcGln9bHTUTcHH4+XNEeg7IerS0Hg8Ga umV+qgG9CerqJ3ABOVk5D3N7kc1PqErwwcsI9/Sp9yqoUNQthPwMYH9/Qec1o/5ag3lE /+HQ==
X-Gm-Message-State: AD7BkJIojQ7oLg+3Rn6l9mV3ZXVeLu1N4vGmI3h3f0ZdCHJe0D0Ixa7alKOlqRwCQsdILA==
X-Received: by 10.98.8.219 with SMTP id 88mr25146773pfi.51.1458956514130; Fri, 25 Mar 2016 18:41:54 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id y23sm19016292pfi.22.2016.03.25.18.41.51 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 18:41:52 -0700 (PDT)
To: "Michael Behringer (mbehring)" , "anima@ietf.org" 
References: <90c950d5911040b28da16e33e2ca4d05@XCH-RCD-006.cisco.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F5E8E7.3060906@gmail.com>
Date: Sat, 26 Mar 2016 14:41:59 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <90c950d5911040b28da16e33e2ca4d05@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: Re: [Anima] Content Distribution
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 01:41:56 -0000

On 25/03/2016 21:55, Michael Behringer (mbehring) wrote:
> Just writing down how I see Intent distribution as a whole process, for discussion. And, I want to abstract from distributing "Intent" to distributing "content", where "Intent" is one type only. 
> 
> Starting point: a node is booted; say it has no Intent initially; 
>    $content_type = "Intent"
>    $my_version = 0;
>    it consults the AN adjacency table (see section 5 of the reference model draft).
>    it has n neighbours in the same domain --> n ACP connections.

They may be connections in the routing protocol, but they aren't connections
in the general sense, as far as I can see.

> Run every t sec, or when I just received new content: 
> - For each neighbour in the adjacency table with an ACP connection:  
>    - send $content_type; $my_version

How is that different from link-local multicast to ALL_GRASP_NEIGHBOR_6 ?
(That is, I think this is simply GRASP flooding with a loop count of 1.)

>    - receive $content_type; $remote _version

So are you asking each neighbor to send you a unicast, or are
you simply assuming that they will flood at the same frequency?

>    - if  $my_version = $remote _version
>            exit;
>    - if  $my_version < $remote_ version
>            pull $content (unicast from that neighbour)

OK, that's standard synchronization.

>    - if  $my _version > $remote_version
>            push $content (unicast to that neighbour)

We have no mechanism for that. You can flood it out or you can
wait to be asked. But anyway, if all nodes execute the same algorithm
at the same frequency, you don't need to push the content; the neighbor
will be asking for it soon enough.

> All of the above happens in the ANI; ASAs are not involved in this operation.

If you want to use GRASP, there will be ad hoc ASAs and objectives that exist
only for this purpose.

> Probably GRASP does the version comparison; 

Nope, that's exactly what the ad hoc ASA would do. It would in fact just
be the container for the above logic (except for the push, unless we add
a protocol feature, which is of course possible). Of course, it's an ASA
that is provided as part of the ANI (like those for enrollment).

> not sure about the file transfer.

If the file can be sent by TCP there's no problem that I can see; it will be
a separate thread anyway. If it really needs something more complex, that
would have to be explicit code in the ad hoc ASA. That's not a problem
either except more code to write.

> The $content_type allows us to use the same mechanism to flood other things; this gives us a lot of flexibility to distribute other content types later on; and, if we find out that we *really* need to separate Intent for different autonomic functions, we could then say $content_type = "Intent for function a" for example (of course this is schematic only; we may want sub-fields). 
> 
> The Intent is stored as a file. it is that file that gets re-distributed. This file is locally validated (signature) and parsed, and then passed to the ASAs. And a portion of Intent will affect the ANI itself. 
> 
> How far off am I with this thinking? I think this is very simple and robust. 
> 
> draft-liu-anima-grasp-distribution-01 has I think the same fundamental thinking, but specified a lot of things that are not needed, IMO: 
> - I do not think we need selective distribution (2.2)

I don't know yet. But I don't think it's a decision we have to take yet.
Maybe there's one objective called "Intent". Maybe there are multiple
objectives called things like "Intent.General", "Intent.PrefixManager",
"Intent.MyStupidIdea". Your idea that $content_type is a parameter is
exactly right, IMHO. That allows for selective distribution later, if
we need it.

> - I do not think we need incremental distribution (2.3)

Again, I don't know yet, but it's a lot simpler if we don't.

> - 3.1 is a bit redundant, if we assume distribution in the ACP. I would rather state
>   as above "to all ACP neighbors", then we have the boundary detection implicit.

It's implicit anyway if we use GRASP discovery. It can't discover anything
outside the AN.

> - I think the above algorithm avoids loops automatically. 

I agree. It's probably not immune to a bad actor that keeps incrementing
the version number, however.

> - I don't think we need selective flooding (3.4)

And again, I don't know yet, but it's a lot simpler if we don't.

    Brian


From nobody Fri Mar 25 20:08:57 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BD23E12D613 for ; Fri, 25 Mar 2016 20:08:54 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.23
X-Spam-Level: 
X-Spam-Status: No, score=-4.23 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id v51oLvudvJJq for ; Fri, 25 Mar 2016 20:08:52 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1AC4712D558 for ; Fri, 25 Mar 2016 20:08:50 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml702-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGO94398; Sat, 26 Mar 2016 03:08:48 +0000 (GMT)
Received: from NKGEML411-HUB.china.huawei.com (10.98.56.70) by lhreml702-cah.china.huawei.com (10.201.5.99) with Microsoft SMTP Server (TLS) id 14.3.235.1; Sat, 26 Mar 2016 03:08:46 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml411-hub.china.huawei.com ([10.98.56.70]) with mapi id 14.03.0235.001; Sat, 26 Mar 2016 11:08:38 +0800
From: Duzongpeng 
To: John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwATdGeIAAjeb4MAASeeeAABRYkEA=
Date: Sat, 26 Mar 2016 03:08:38 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>   
In-Reply-To: 
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.149.226]
Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFCD33nkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A090203.56F5FD41.0047, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 75564e734a57cfeb0e99f48b48eaa534
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , "Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 03:08:55 -0000

--_000_BAFEC9523F57BC48A51C20226A5589575FDFCD33nkgeml514mbxchi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFCD33nkgeml514mbxchi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFCD33nkgeml514mbxchi_--



From nobody Fri Mar 25 20:46:35 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6BD1812D170 for ; Fri, 25 Mar 2016 20:46:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.23
X-Spam-Level: 
X-Spam-Status: No, score=-4.23 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0bQF0a3WrGTT for ; Fri, 25 Mar 2016 20:46:29 -0700 (PDT)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 82F9212D169 for ; Fri, 25 Mar 2016 20:46:27 -0700 (PDT)
Received: from 172.18.7.190 (EHLO lhreml703-cah.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CGO95960; Sat, 26 Mar 2016 03:46:24 +0000 (GMT)
Received: from NKGEML408-HUB.china.huawei.com (10.98.56.39) by lhreml703-cah.china.huawei.com (10.201.5.104) with Microsoft SMTP Server (TLS) id 14.3.235.1; Sat, 26 Mar 2016 03:46:22 +0000
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml408-hub.china.huawei.com ([10.98.56.39]) with mapi id 14.03.0235.001; Sat, 26 Mar 2016 11:46:11 +0800
From: Duzongpeng 
To: John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AdF/JG1hKQbSxNs2S22ajBGI+Y2RPgANV1rwATdGeIAAHjH6AAAg/AuAAFA8wmAAEj1TgAAUYyTg
Date: Sat, 26 Mar 2016 03:46:11 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>   
In-Reply-To: 
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.149.226]
Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFCD67nkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A090203.56F60611.0097, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 75564e734a57cfeb0e99f48b48eaa534
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , "Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 03:46:34 -0000

--_000_BAFEC9523F57BC48A51C20226A5589575FDFCD67nkgeml514mbxchi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFCD67nkgeml514mbxchi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFCD67nkgeml514mbxchi_--


From nobody Fri Mar 25 21:15:31 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6F3FD12D1A5 for ; Fri, 25 Mar 2016 21:15:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4x1qDOygNYMe for ; Fri, 25 Mar 2016 21:15:26 -0700 (PDT)
Received: from mail-pf0-x230.google.com (mail-pf0-x230.google.com [IPv6:2607:f8b0:400e:c00::230]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 4CB9012D192 for ; Fri, 25 Mar 2016 21:15:26 -0700 (PDT)
Received: by mail-pf0-x230.google.com with SMTP id n5so95007226pfn.2 for ; Fri, 25 Mar 2016 21:15:26 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=0mri5m63iwYHbSQHN7LWCFnbrBeM5TGC/JZihHaS6+4=; b=ne2QGmkWugbkPQJ3GP9VZN7PeJgCMpdn001HpGlm3rZpfN2slwzmqUfT6tO3vf3pr5 PDeemK/xmfMctY0eYGm/GOr8SUDxcoUtmYz7JzlQQhL2J5oTEChPfk5eMzGN5FZ1X9Ix oZLT87kKW/AKIJ2OszHXUfPL8iMWdjqQGKKxv4jz6FLM3VuUuK++vUipwTiApOBNqpeY 4KOkM3gtxzFpXpxsrJMYGlqFzfyO7P3zSTmsuS44MazvKr3e5/VFpkjeZWJEV3XoiS4k XkvSMLXkctYQSVYdMjihGH/5XLp8Q7EiQlggsA8AKRHkucWNvRcqreh3Azm0jee/0M1z 3EOA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=0mri5m63iwYHbSQHN7LWCFnbrBeM5TGC/JZihHaS6+4=; b=VcRz2o63hOxRuWS44KciVNr5jkYAxc9qtd/yrGnKHi0OUHkvncgLmLx7CV7z4YkO75 7crWse5/C3f3oVHEslcYHmLuRnHHaRXZqabSsTOPBQmL5gZKz9w7lXPNiNONyT30eDIi imS4ep4yo7tZZA2IDRDaJplHsRUHO7X8LWyNGA5ILnMv9XBKBXjdRMowUMdsj95O0ovF fJJquBYJFI3uEmM9cT8wQ6RZ6XJMkYsrH8EyaWvwblevr6c2VK+rChKwpnlNgQoY356z 7717fwdd6Fxy1pdr7TAl5dWWz318SA1A3c8xKirVT2SOl16dSn2scdUOZgeDyqpPWkQi n7pw==
X-Gm-Message-State: AD7BkJLTP2TYURIuRmVxvL+9YZHkdU28z66fJ1IuB22C42SG1SNCR5odOprue2MJ/3beGg==
X-Received: by 10.98.89.22 with SMTP id n22mr25814323pfb.55.1458965725833; Fri, 25 Mar 2016 21:15:25 -0700 (PDT)
Received: from ?IPv6:2406:e007:65bb:1:28cc:dc4c:9703:6781? ([2406:e007:65bb:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id p26sm19503159pfi.84.2016.03.25.21.15.20 (version=TLSv1/SSLv3 cipher=OTHER); Fri, 25 Mar 2016 21:15:24 -0700 (PDT)
To: Duzongpeng , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>    
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F60CE4.9040901@gmail.com>
Date: Sat, 26 Mar 2016 17:15:32 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "Michael Behringer \(mbehring\)" , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 04:15:30 -0000

> However, in ANIMA network, perhaps we do not have a centralized parsing=
 node (controller), which can replace almost all the communications betwe=
en nodes.

Correct, and there is also no presumption that individual nodes have perf=
ect
knowledge of the topology. Of course the ACP routing protocol will have p=
erfect
knowledge of the ACP topology, but individual ASAs only know what they ha=
ve
gleaned from discovery and what they have been told by Intent.

It is, I think, a very different world view than the SDN or SUPA or NVO4 =
or NFV view.

Regards
   Brian

On 26/03/2016 16:46, Duzongpeng wrote:
> Hi John:
>=20
> Thanks for your reply.
>=20
> 
> How is this different than running a script, or invoking an API? There =
are plenty of southbound APIs (from a management function to a device). T=
here are very few northbound APIs. That's what the operators that I have =
talked to want.
> 
>=20
> I agree that intent belongs to northbound interface in the SDN network.=
 And ANIMA network perhaps can share the same northbound interface.
>=20
> But, my understanding is that ANIMA network is a little different from =
SDN network.
>=20
> SDN controller will parse the intent, and transform to detailed configu=
rations of every device. It is a good model, and little communications be=
tween nodes are needed.
>=20
> Some network level parameters can be handled in SDN controller.
>=20
> However, in ANIMA network, perhaps we do not have a centralized parsing=
 node (controller), which can replace almost all the communications betwe=
en nodes.
>=20
> So when a network operator is operating a ANIMA network, the network le=
vel parameters are still needed to be flooded.
>=20
> I mean the ANIMA interface is perhaps something like =E2=80=9Cnorthboun=
d + part of southbound=E2=80=9D.
>=20
> Best Regards
> Zongpeng Du
>=20
>=20
> From: John Strassner [mailto:strazpdj@gmail.com]
> Sent: Saturday, March 26, 2016 9:41 AM
> To: Duzongpeng; John Strassner
> Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang; anima=
@ietf.org; J=C3=A9ferson Campos Nobre
> Subject: Re: [Anima] Does Autonomic network need some intervention beyo=
nd intent
>=20
> Hi Zongpeng,
>=20
> I think we have a misunderstanding here.
>=20
>=20
> But I do not support that the intent writer should look every detail in=
 the catalog.
> 
> Most catalogs that I have seen have very few details. They instead tell=
 the user what services or products are available.
> 
>=20
> The intent writer may mainly care about what they want, but how to do i=
t can be left to the autonomic network to deal with.
> 
> I agree completely.
> 
>=20
> Perhaps several solutions exist, and autonomic management system choose=
 one according to the abilities collected.
> 
> This is one of the reasons why I would like to see capabilities (and ne=
eds) advertised.
> 
>=20
> ...
> I mean in current status, some of the Autonomic Functions should be exp=
osed to the network operator, and some of the Autonomic Functions can be =
hidden to the network operator
> ...
> 
> How is this different than running a script, or invoking an API? There =
are plenty of southbound APIs (from a management function to a device). T=
here are very few northbound APIs. That's what the operators that I have =
talked to want.
>=20
> regards,
> John
>=20
> On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng > wrote:
> Hi John
>=20
> Some personal understanding here. If any misunderstanding, please corre=
ct me. Thanks.
>=20
> 
> I would suggest a slight modification, to enable users of
> intent (and even intent writers) to NOT have to have
> detailed implementation information.
>=20
> What if all of the Autonomic Functions, after they joined
> the network, advertised both their capabilities as well as
> their needs? We can put aside the latter (needs) as this
> is more complex, but if the capabilities are advertised,
> then the autonomic management system can assemble
> those needs into a catalog. Then, the intent writer can
> look at the catalog and write a declarative request to
> use the functionality of the catalog.
>=20
> I will look at your YANG model in a bit (sorry, I'm woefully
> behind...).
> < /jcs2>
>=20
> I agree that autonomic node should have that self-advertisement ability=
, and autonomic management system should have this catalog.
>=20
> But I do not support that the intent writer should look every detail in=
 the catalog.
>=20
> The intent writer may mainly care about what they want, but how to do i=
t can be left to the autonomic network to deal with.
> Perhaps several solutions exist, and autonomic management system choose=
 one according to the abilities collected.
>=20
> In the ideal, the network operator may communicate with the autonomic n=
etwork using network-level level intent and network-level report.
> I mean in current status, some of the Autonomic Functions should be exp=
osed to the network operator, and some of the Autonomic Functions can be =
hidden to the network operator
> Perhaps in future, when the network nodes become more intelligent, the =
hidden part will increase.
>=20
>=20
> Best regards
> Zongpeng Du
>=20
> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of John Strassner
> Sent: Thursday, March 24, 2016 10:41 AM
> To: Michael Behringer (mbehring); John Strassner
> Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org; J=C3=A9ferson Campos Nobre
> Subject: Re: [Anima] Does Autonomic network need some intervention beyo=
nd intent
>=20
> Hi Michael,
>=20
> thanks for your reply. I do think that we largely agree, and would enjo=
y chatting during or (hopefully before) BA. Please see ... A=
lso trying to unclutter, so I am eliding points that we agree on.
>=20
>=20
> While *CIEs don=E2=80=99t like magic, they do understand the concept of=
 a default behaviour which they can influence if needed. Any reasonable n=
etwork management approach tries to define =E2=80=9Cdefault=E2=80=9D beha=
viour and =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cd=
efault=E2=80=9D part very nicely, and *CIEs get that.
> 
> Agreed!
> 
>=20
> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discuss=
ion: My view is that we shouldn=E2=80=99t try to define exactly *what* in=
tent allows and doesn=E2=80=99t allow, since we=E2=80=99ve heard from man=
y sides that the boundary is not clear, and it may be impossible to have =
a really clear, usable definition. So my point was: Let=E2=80=99s focus o=
n the distribution model: Intent is flooded to all nodes. Let=E2=80=99s N=
OT do selective flooding, unicast Intent to specific nodes, etc, because =
that would blur the boundary even worse.
> 
> I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1 =
was 2006, almost 10 years ago), we used a message bus and pub-sub. In v2,=
 we flooded; in v3, we went back to pub-sub because we wanted to do some =
optimization and auditing. But I would be happy to try some type of contr=
olled flooding again.
> 
> My suggestion: Intent is flooded to all nodes in a domain. The ASA owne=
rs define the content of what they want to push in their bit of the Inten=
t. If we insist of full flooding, we encourage the right behaviour. And i=
f someone really puts into his Intent =E2=80=9Cnode A: do this; node B: d=
o the other; node C: do the third=E2=80=9D it is awkward enough for peopl=
e to take a step back and think whether they=E2=80=99re doing the right t=
hing =E2=98=BA
> 
> I agree in principle with what you said. Plus, see above.
> < /jcs2>
> 
> Do you mean that the intent writer will **know the difference**
> between the fabric and each autonomic function? I hope not,
> but perhaps I'm misunderstanding you...
> < /jcs>
> Not sure I understand your concern. Yes, in my model there is a section=
 for the fabric, and sections for the various functions. Seems a natural =
structure?!? 
> The intent **writer** would have to know implementation
> details in order to place everything in a single file. I would
> rather not require this, and instead, have the intent engine
> (e.g., compiler and associated logic) do this.
> < /jcs2>
> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to struct=
ure Intent around Autonomic Functions. This is not optimal from a scienti=
fic point of view, but aligns very well with the way how things are organ=
ised. Probably we need to discuss that in BA. But refer to my YANG model =
sent a few days ago.
> 
> I would suggest a slight modification, to enable users of
> intent (and even intent writers) to NOT have to have
> detailed implementation information.
>=20
> What if all of the Autonomic Functions, after they joined
> the network, advertised both their capabilities as well as
> their needs? We can put aside the latter (needs) as this
> is more complex, but if the capabilities are advertised,
> then the autonomic management system can assemble
> those needs into a catalog. Then, the intent writer can
> look at the catalog and write a declarative request to
> use the functionality of the catalog.
>=20
> I will look at your YANG model in a bit (sorry, I'm woefully
> behind...).
> < /jcs2>
>=20
>=20
> regards,
> John
>=20
> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) > wrote:
> John,
>=20
> I think we=E2=80=99re on the same page. Couple of points, top posting h=
ere, otherwise it gets too messy.
>=20
> While *CIEs don=E2=80=99t like magic, they do understand the concept of=
 a default behaviour which they can influence if needed. Any reasonable n=
etwork management approach tries to define =E2=80=9Cdefault=E2=80=9D beha=
viour and =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cd=
efault=E2=80=9D part very nicely, and *CIEs get that.
>=20
> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discuss=
ion: My view is that we shouldn=E2=80=99t try to define exactly *what* in=
tent allows and doesn=E2=80=99t allow, since we=E2=80=99ve heard from man=
y sides that the boundary is not clear, and it may be impossible to have =
a really clear, usable definition. So my point was: Let=E2=80=99s focus o=
n the distribution model: Intent is flooded to all nodes. Let=E2=80=99s N=
OT do selective flooding, unicast Intent to specific nodes, etc, because =
that would blur the boundary even worse.
>=20
> My suggestion: Intent is flooded to all nodes in a domain. The ASA owne=
rs define the content of what they want to push in their bit of the Inten=
t. If we insist of full flooding, we encourage the right behaviour. And i=
f someone really puts into his Intent =E2=80=9Cnode A: do this; node B: d=
o the other; node C: do the third=E2=80=9D it is awkward enough for peopl=
e to take a step back and think whether they=E2=80=99re doing the right t=
hing =E2=98=BA
>=20
> 
> Do you mean that the intent writer will **know the difference**
> between the fabric and each autonomic function? I hope not,
> but perhaps I'm misunderstanding you...
> 
>=20
> Not sure I understand your concern. Yes, in my model there is a section=
 for the fabric, and sections for the various functions. Seems a natural =
structure?!?
>=20
> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to struct=
ure Intent around Autonomic Functions. This is not optimal from a scienti=
fic point of view, but aligns very well with the way how things are organ=
ised. Probably we need to discuss that in BA. But refer to my YANG model =
sent a few days ago.
>=20
> We should schedule some discussion time outside the ANIMA meeting windo=
w...
>=20
> Michael
>=20
>=20
> From: John Strassner [mailto:strazpdj@gmail.com]
> Sent: 22 March 2016 21:32
> To: Michael Behringer (mbehring) >; John Strassner >=

> Cc: Duzongpeng >; a=
nima@ietf.org; Laurent Ciavaglia >; J=C3=A9ferson Campos No=
bre >; Sheng Jiang >
> Subject: Re: [Anima] Does Autonomic network need some intervention beyo=
nd intent
>=20
> In general, I agree with Michael. I'd like to emphasize and expand on a=
 couple of key points:
>=20
>=20
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D=
 or =E2=80=9Csteering=E2=80=9D a network, at least not initially. There i=
s still traditional config, plus all the existing and emerging SDN models=
=2E My view has always been that ANIMA sets a =E2=80=9Cdefault=E2=80=9D p=
olicy in a network, and all other means are more specific and can overrid=
e intent. This is described in RFC7575. Main point here: ANIMA Intent isn=
=E2=80=99t the only tool in a network.
>=20
> 
> Absolutely agree. Note that even in the Congress case I mentioned
> earlier, it is NOT pushing config, and it is NOT viewed as the
> "only" policy
> 
>=20
>=20
> Intent could perfectly well cover a high level policy such as =E2=80=9C=
all nodes of type x do this; type y does that=E2=80=9D. (Which I think is=
 what you mention below.) When we started with Intent, the vision was to =
keep it very high level, more along the line =E2=80=9Cdo the right thing=E2=
=80=9D =E2=98=BA
>=20
> 
> Agreed. Although to me, this means that we need to consider the types o=
f actors that are using intent. For example, most of my CCIE and JCIE fri=
ends have no interest in using an intent language for pushing configurati=
on. On the other hand, when I talk to operators and application developer=
s, they would love to be able to use a high-level intent and have the int=
ent engine take care of the details for them.
> 
>=20
> It has become clear that in today=E2=80=99s networking, people will wan=
t to push more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D=
, and I think we=E2=80=99ll just have to acknowledge that.
>=20
> 
> Sorry Michael, I don't follow here. What do you mean
> by "parameters"?
> 
>=20
>=20
> Personally I think it=E2=80=99ll pan out like this: Intent will have se=
ctions per autonomic function. An Intent parser will validate overall cor=
rectness (signature, etc), then segment the Intent into its bits: One par=
t will be the autonomic fabric itself, then there will be bits for each a=
utonomic function.
>=20
> 
> Do you mean that the intent writer will **know the difference**
> between the fabric and each autonomic function? I hope not,
> but perhaps I'm misunderstanding you...
> 
>=20
>=20
> So the Intent parser on a node will grab intent, see there is some Inte=
nt for autonomic function A, will see whether this is running locally, an=
d if so, it=E2=80=99ll just pass that piece of Intent to that function.
>=20
> 
> Agreed.
> 
>=20
>=20
> In this way, a function owner defines both the intent itself, and how i=
t is consumed by his/her function.
>=20
> 
> I'm not sure what you mean by "function owner". (Sorry to be
> picky, I'm likely under-caffeinated.)
>=20
> The above seemed like an autonomous decision, made by the parser.
> I don't think that the intent writer should be cognizant of how many or=

> what types of autonomic functions are present.
> 
>=20
>=20
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people c=
an do what they want, in the worst case even push config, which is clearl=
y NOT what Intent is meant to do.
>=20
> 
> And hence, the "HAL 2000" scenario. :-)
> 
>=20
>=20
> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=
=80=9Cnode 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node=
 37: here is your CLI=E2=80=9D. But what I describe above COULD be abused=
 in this way. RFC7575 says this is NOT Intent.
>=20
> 
> Absolutely agree!
> 
>=20
>  regards,
> John
>=20
>=20
>=20
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) > wrote:
> Hi Zongpeng,
>=20
> Yes, this is a recurring question and topic. Let me give my view on two=
 levels:
>=20
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D=
 or =E2=80=9Csteering=E2=80=9D a network, at least not initially. There i=
s still traditional config, plus all the existing and emerging SDN models=
=2E My view has always been that ANIMA sets a =E2=80=9Cdefault=E2=80=9D p=
olicy in a network, and all other means are more specific and can overrid=
e intent. This is described in RFC7575. Main point here: ANIMA Intent isn=
=E2=80=99t the only tool in a network.
>=20
> Intent could perfectly well cover a high level policy such as =E2=80=9C=
all nodes of type x do this; type y does that=E2=80=9D. (Which I think is=
 what you mention below.) When we started with Intent, the vision was to =
keep it very high level, more along the line =E2=80=9Cdo the right thing=E2=
=80=9D =E2=98=BA  It has become clear that in today=E2=80=99s networking,=
 people will want to push more =E2=80=9Cparameters=E2=80=9D than =E2=80=9C=
policy=E2=80=9D, and I think we=E2=80=99ll just have to acknowledge that.=

>=20
> Personally I think it=E2=80=99ll pan out like this: Intent will have se=
ctions per autonomic function. An Intent parser will validate overall cor=
rectness (signature, etc), then segment the Intent into its bits: One par=
t will be the autonomic fabric itself, then there will be bits for each a=
utonomic function.
>=20
> So the Intent parser on a node will grab intent, see there is some Inte=
nt for autonomic function A, will see whether this is running locally, an=
d if so, it=E2=80=99ll just pass that piece of Intent to that function.
>=20
> In this way, a function owner defines both the intent itself, and how i=
t is consumed by his/her function.
>=20
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people c=
an do what they want, in the worst case even push config, which is clearl=
y NOT what Intent is meant to do. I=E2=80=99d hope that people don=E2=80=99=
t start writing Intent like =E2=80=9Cnode 17: here is your CLI; node 23: =
here is your CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D. But what I=
 describe above COULD be abused in this way. RFC7575 says this is NOT Int=
ent.
>=20
> In other words: Let=E2=80=99s not be too prescriptive, and hope that fo=
lks will do more or less reasonable things.
>=20
> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly =
contribute, I=E2=80=99m just struggling with priorities at the moment (as=
 you can see by my relative silence over the past weeks).
>=20
> Michael
>=20
>=20
>=20
> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Duzongpeng
> Sent: 16 March 2016 02:38
> To: anima@ietf.org
> Cc: Laurent Ciavaglia >; J=C3=A9ferson Campos Nobre >; Sheng Jiang >
> Subject: [Anima] Does Autonomic network need some intervention beyond i=
ntent
>=20
> Hello, everyone
>=20
>          I understand that Autonomic network should be able to work wit=
h as less configuration as possible. In the ideal case, an Autonomic netw=
ork should be able to work well while only intent is needed from human op=
erators.
>=20
>          However, I wonder whether Autonomic network may need some inte=
rvention from human operators through some low-level policies beyond inte=
nts.
>=20
>          For example, in the https://tools.ietf.org/html/draft-ietf-ani=
ma-prefix-management-00, it is suggested that the prefix lengths for the =
CSG, ASG, RSG (different roles in IP RAN) can be assigned as an "intent".=
 These configuration parameters need to be distributed in the autonomic d=
omain to influence the detail configurations on each autonomic node.
>=20
>          As intent is described as an abstract, declarative, high-level=
 policy used to operate an autonomic domain, such as an enterprise networ=
k. Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cpre=
fix-management ID=E2=80=9D to some other words.
>=20
> Is =E2=80=9CASA parameters" ok here?
>=20
> I think some characteristics for this kind of low-level policies are
>=20
> 1.       Network level parameters, need to be distributed by GRASP and =
ACP
>=20
> 2.       Relatively static compared to intent, perhaps only needed to b=
e configured once
>=20
> 3.       Unavoidable in current network environment, mostly for establi=
shing the network infrastructure
>=20
> 4.       Rarely need coordinations with others parameters, usually is c=
losed related to one specific autonomic function
>=20
> Welcome for comment. Your suggestions will be appreciated.
>=20
>=20
> Best Regards
> Zongpeng Du
>=20
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>=20
>=20
>=20
> --
> regards,
> John
>=20
>=20
>=20
> --
> regards,
> John
>=20
>=20
>=20
> --
> regards,
> John
>=20
>=20
>=20
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>=20


From nobody Sat Mar 26 08:43:51 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F1EDB12D670 for ; Sat, 26 Mar 2016 08:43:49 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id GZSDempS-3El for ; Sat, 26 Mar 2016 08:43:46 -0700 (PDT)
Received: from mail-lb0-x236.google.com (mail-lb0-x236.google.com [IPv6:2a00:1450:4010:c04::236]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 71BF112D675 for ; Sat, 26 Mar 2016 08:43:42 -0700 (PDT)
Received: by mail-lb0-x236.google.com with SMTP id bc4so61344921lbc.2 for ; Sat, 26 Mar 2016 08:43:42 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=woRMlLE9tJ/ctPxFPDi5vKNO30NXz7zkltSbuLPHyjQ=; b=oaNp5RqXJVJOxLFOXeqwG1ypnKag/d3IFVA4Kpg4/x1HSWkyP/QjKNOCoLTzXVVa7O pJpTE8QowlPOMdbjfH1EHUhhydj+kDMPqqKRkrdM+93yfYzetD48/EhleYNpCoVJTd8J kgnOg6I9eawO3+B9tWy0vfEvbVyrN8X6xjPqnMV9jq0FNmpz6TO/AtAGbMzDzKM54MVY JEyARolKKln9Vs6QkeMQ97PJFW6kA8KIktg2aD+TgHGLbDjE1upkCBQ6RraIQKi2B4fy Y73lG7aArrZIO9Nh/3I3w3KugDJ9tny3IYZyXC59yl1awbO/4G/keVdOR0fWJraZQt2I MPbQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=woRMlLE9tJ/ctPxFPDi5vKNO30NXz7zkltSbuLPHyjQ=; b=G1f5qBomnpCzPCF252a3oRjjXUdZQt2Fni53TTSpSRYiGOtebV39ih64IiVlqWK4I3 rLDUnG+I1RbWZ81E7jfMGYduooytL5A8nSX3pG37GOKz0ET2LVjS3Jz9MN/44wgXn7PL g/cuzeSk8cRxHCpru+NS7T9Ym3dz204TVnQ3JiaCrC0Lrbzk7wBSZIFHB+o7h2j2Ccel SzToay5sHnC5RwUdOkuc1PXJnkSub3aKxbPSjjLnoyues338McM4yxMFUzMtyZQzfhKa OnAyRX9kAlOpWGDrk/JQONqPd2ZJDhl4tmmJ55Ok5umogdo6eU2xObTA40xphzHWiYQ1 Xt4Q==
X-Gm-Message-State: AD7BkJJM9njNlYga4QZt+d6Ag1JossCpSM5AT8cxTksgz2VXl9jqxOB1dDGd4nCxmuqYX2pXPWP4NWrQa7g3ew==
MIME-Version: 1.0
X-Received: by 10.112.170.68 with SMTP id ak4mr5988027lbc.94.1459007020599; Sat, 26 Mar 2016 08:43:40 -0700 (PDT)
Received: by 10.25.22.19 with HTTP; Sat, 26 Mar 2016 08:43:40 -0700 (PDT)
In-Reply-To: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>    
Date: Sat, 26 Mar 2016 08:43:40 -0700
Message-ID: 
From: John Strassner 
To: Duzongpeng , John Strassner 
Content-Type: multipart/alternative; boundary=001a11c259a83011b9052ef58b11
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "Michael Behringer \(mbehring\)" , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 15:43:50 -0000

--001a11c259a83011b9052ef58b11
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Zongpeng,

I agree that both approaches can exist concurrently. This is a good
compromise.

regards,
John

On Fri, Mar 25, 2016 at 8:08 PM, Duzongpeng  wrote:

> Hi John
>
>
>
> Thanks for your reply.
>
>
>
> I know that you have been working on a policy model, In which intents are
> translated to low-level policies (with parameters) by a policy continuum.
>
>
>
> I think it is valuable and should be considered in the development of
> ANIMA network.
>
>
>
> However, it does not mean in ANIMA network, we can not operator ASA
> parameters directly using some simple mechanism.
>
>
>
> IMO, which mechanism should be used is decided by the ASA, and it should
> be implementation-specific.
>
>
>
> I means two mechanisms can both exist, and ASA can choose which to use
> according to its requirements and context.
>
>
>
> If some ASAs need that the network operator to management directly, inten=
t
> writer will need to know about that autonomic function, such as the purpo=
se
> of the function, how to use, and parameters needed.
>
>
>
> I agree that is kindly like a function call or an API, but it is network
> level. If you choose it is not an intent, while I think that perhaps some
> ASA need that, my suggestion is that we can choose another name.
>
>
>
> Or we call it a certain kind of ANIMA intent, and declare that ANIMA
> intent has a larger scope than a pure intent.
>
>
>
> Best Regards
>
> Zongpeng Du
>
>
>
> *From:* John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* Saturday, March 26, 2016 9:05 AM
> *To:* Duzongpeng; John Strassner
> *Cc:* Michael Behringer (mbehring); anima@ietf.org; Laurent Ciavaglia;
> J=C3=A9ferson Campos Nobre; Sheng Jiang
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> Hi Zongpeng,
>
>
>
> Let me try to rephrase. It isn't a question of top-down or bottom-up, but
> rather, how complex the user model is. I am recommending a simple user
> model. This means that the user should not have to know in advance the li=
st
> of specific autonomic functions that are present.
>
>
>
> I do agree that we should start simple. One autonomic function is fine.
> However, I don't understand why you say that "it is not practical that an
> intent writer in ANIMA network do not know the information about autonomi=
c
> functions in the autonomic domain". Intent shouldn't be a function call
> or an API. It should be expressed in a form natural to the intent writer,
> and then translated to a form that a device can consume.
>
>
>
> regards,
>
> John
>
>
>
> On Fri, Mar 25, 2016 at 1:30 AM, Duzongpeng  wrote=
:
>
> Hi, John
>
>
>
> Some personal understandings for discussion.
>
>
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
>
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
>
>
> I think I can understand your idea. However, it looks like a top-down
> method.
>
>
>
> It has been suggested that ANIMA can work by using a bottom-up method.
>
>
>
> I mean perhaps we can make a minimum set for intent right now, and make i=
t
> work first.
>
>
>
> It has been suggested that ANIMA ASAs will be developed in a one-by-one
> procedure, so that currently there is not too much information that neede=
d
> to be carried in the intent.
>
>
>
> I think that is why Michael has said intent should be a small one.
>
>
>
> In the ideal, AFA will finally be transparent for intent. Perhaps, it is
> because most of the need of operator can be expressed using abstract
> languages.
>
>
>
> But in current status, it is not practical that an intent writer in ANIMA
> network do not know the information about autonomic functions in the
> autonomic domain.
>
>
>
> Best Regards
>
> Zongpeng Du
>
>
>
>
>
> *From:* John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* Wednesday, March 23, 2016 4:32 AM
> *To:* Michael Behringer (mbehring); John Strassner
> *Cc:* Duzongpeng; anima@ietf.org; Laurent Ciavaglia; J=C3=A9ferson Campos
> Nobre; Sheng Jiang
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
>
>
> In general, I agree with Michael. I'd like to emphasize and expand on a
> couple of key points:
>
>
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
>
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J
>
>
>
> 
>
> Agreed. Although to me, this means that we need to consider the types of
> actors that are using intent. For example, most of my CCIE and JCIE frien=
ds
> have no interest in using an intent language for pushing configuration. O=
n
> the other hand, when I talk to operators and application developers, they
> would love to be able to use a high-level intent and have the intent engi=
ne
> take care of the details for them.
>
> 
>
>
>
> It has become clear that in today=E2=80=99s networking, people will want =
to push
> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I th=
ink we=E2=80=99ll just have to acknowledge
> that.
>
>
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
>
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
>
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> 
>
> Agreed.
>
> 
>
>
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
>
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
>
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do.
>
>
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
>
>
>
>
> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=
=80=9Cnode 17: here is
> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CLI=
=E2=80=9D. But what
> I describe above COULD be abused in this way. RFC7575 says this is NOT
> Intent.
>
>
>
> 
>
> Absolutely agree!
>
> 
>
>
>
>  regards,
>
> John
>
>
>
>
>
>
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
> mbehring@cisco.com> wrote:
>
> Hi Zongpeng,
>
>
>
> Yes, this is a recurring question and topic. Let me give my view on two
> levels:
>
>
>
> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D o=
r =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
>
>
>
> Intent could perfectly well cover a high level policy such as =E2=80=9Cal=
l nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J  It has=
 become clear
> that in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.
>
>
>
> Personally I think it=E2=80=99ll pan out like this: Intent will have sect=
ions per
> autonomic function. An Intent parser will validate overall correctness
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
>
>
>
> So the Intent parser on a node will grab intent, see there is some Intent
> for autonomic function A, will see whether this is running locally, and i=
f
> so, it=E2=80=99ll just pass that piece of Intent to that function.
>
>
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
>
>
> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can=
 do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writing =
Intent like =E2=80=9Cnode
> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here i=
s your
> CLI=E2=80=9D. But what I describe above COULD be abused in this way. RFC7=
575 says
> this is NOT Intent.
>
>
>
> In other words: Let=E2=80=99s not be too prescriptive, and hope that folk=
s will do
> more or less reasonable things.
>
>
>
> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly co=
ntribute,
> I=E2=80=99m just struggling with priorities at the moment (as you can see=
 by my
> relative silence over the past weeks).
>
>
>
> Michael
>
>
>
>
>
>
>
> *From:* Anima [mailto:anima-bounces@ietf.org] *On Behalf Of *Duzongpeng
> *Sent:* 16 March 2016 02:38
> *To:* anima@ietf.org
> *Cc:* Laurent Ciavaglia ; J=C3=A9ferson Camp=
os
> Nobre ; Sheng Jiang 
> *Subject:* [Anima] Does Autonomic network need some intervention beyond
> intent
>
>
>
> Hello, everyone
>
>
>
>          I understand that Autonomic network should be able to work with
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
>
>
>
>          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
>
>
>
>          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
>
>
>
>          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
>
>
>
> Is =E2=80=9CASA parameters" ok here?
>
>
>
> I think some characteristics for this kind of low-level policies are
>
> 1.       Network level parameters, need to be distributed by GRASP and AC=
P
>
> 2.       Relatively static compared to intent, perhaps only needed to be
> configured once
>
> 3.       Unavoidable in current network environment, mostly for
> establishing the network infrastructure
>
> 4.       Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
>
>
>
> Welcome for comment. Your suggestions will be appreciated.
>
>
>
>
>
> Best Regards
>
> Zongpeng Du
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> --
>
> regards,
>
> John
>
>
>
>
> --
>
> regards,
>
> John
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>


--=20
regards,
John

--001a11c259a83011b9052ef58b11
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi Zongpeng,

I agree that bo=
th approaches can exist concurrently. This is a good compromise.
=

regards,
John

On Fri, Mar 25, 2016 at 8:08 PM, Duzongpen=
g <duzongpeng@huawei.com> wrote:











Hi =
John


=
=C2=A0


Tha=
nks for your reply.



=
=C2=A0


I k=
now that you have been working on a policy model, In which intents are tran=
slated to low-level policies (with parameters) by a policy continuum.



=
=C2=A0


I t=
hink it is valuable and should be considered in the development of ANIMA ne=
twork.



=
=C2=A0


How=
ever, it does not mean in ANIMA network, we can not operator ASA parameters=
 directly using some simple mechanism.


=
=C2=A0


IMO=
, which mechanism should be used is decided by the ASA, and it should be im=
plementation-specific.


=
=C2=A0


I m=
eans two mechanisms can both exist, and ASA can choose which to use accordi=
ng to its requirements and context.


=
=C2=A0


If =
some ASAs need that the network operator to management directly, intent wri=
ter will need to know about that autonomic function, such as the
 purpose of the function, how to use, and parameters needed.<=
/span>


=
=C2=A0


I a=
gree that is kindly like a function call or an API, but it is network level=
. If you choose it is not an intent, while I think that perhaps some
 ASA need that, my suggestion is that we can choose another name.=



=
=C2=A0


Or =
we call it a certain kind of ANIMA intent, and declare that ANIMA intent ha=
s a larger scope than a pure intent.


=
=C2=A0


Bes=
t Regards


Zon=
gpeng Du


=
=C2=A0




From: John Strassner [mailto:strazpdj@gmail.com]


Sent: Saturday, March 26, 2016 9:05 AM

To: Duzongpeng; John Strassner

Cc: Michael Behringer (mbehring); anima@ietf.org; Laurent Ciavaglia; J=C3=A9ferson Camp=
os Nobre; Sheng Jiang

Subject: Re: [Anima] Does Autonomic network need some intervention b=
eyond intent




=C2=A0






Hi=
 Zongpeng,






=C2=A0






Le=
t me try to rephrase. It isn't a question of top-down or bottom-up, but=
 rather, how complex the user model is. I am recommending a simple user mod=
el. This means
 that the user should not have to know in advance the list of specific auto=
nomic functions that are present.






=C2=A0






I =
do agree that we should start simple. One autonomic function is fine. Howev=
er, I don't understand why you say that "it
 is not practical that an intent writer in ANIMA network do not know the in=
formation about autonomic functions in the autonomic domain". Intent shouldn't be a function call or an API. It s=
hould be expressed in a form natural to the
 intent writer, and then translated to a form that a device can consume.=







=C2=A0






re=
gards,






Jo=
hn








=C2=A0




On=
 Fri, Mar 25, 2016 at 1:30 AM, Duzongpeng <duzongpeng@huawei.com> wrote:






Hi,=
 John


=C2=
=A0


Som=
e personal understandings for discussion.


=C2=
=A0



<jcs>



I'm not sure what you mean by "function owner=
". (Sorry to be



picky, I'm likely under-caffeinated.)



=C2=A0



The above seemed like an autonomous decision, made by =
the parser.



I don't think that the intent writer should be cog=
nizant of how many or



what types of autonomic functions are present.<=
u>



</jcs>


=C2=
=A0


I t=
hink I can understand your idea. However, it looks like a top-down method.<=
/span>


=C2=
=A0


It =
has been suggested that ANIMA can work by using a bottom-up method.<=
span lang=3D"EN-US">


=C2=
=A0


I m=
ean perhaps we can make a minimum set for intent right now, and make it wor=
k first.



=C2=
=A0


It =
has been suggested that ANIMA ASAs will be developed in a one-by-one proced=
ure, so
 that currently there is not too much information that needed to be carried=
 in the intent.


=C2=
=A0


I t=
hink that is why Michael has said intent should be a small one.


=C2=
=A0


In =
the ideal, AFA will finally be transparent for intent. Perhaps, it is becau=
se most
 of the need of operator can be expressed using abstract languages.<=
span lang=3D"EN-US">


=C2=
=A0


But=
 in current status, it is not practical that an intent writer in ANIMA netw=
ork do
 not know the information about autonomic functions in the autonomic domain=
.


=C2=
=A0


Bes=
t Regards


Zon=
gpeng Du


=C2=
=A0


=C2=
=A0




From: John
 Strassner [mailto:=
strazpdj@gmail.com]


Sent: Wednesday, March 23, 2016 4:32 AM

To: Michael Behringer (mbehring); John Strassner

Cc: Duzongpeng; =
anima@ietf.org; Laurent Ciavaglia; J=C3=A9ferson Campos Nobre; Sheng Ji=
ang

Subject: Re: [Anima] Does Autonomic network need some intervention b=
eyond intent




=C2=A0







In general, I agree with Michael. I'd like to emph=
asize and expand on a couple of key points:







=C2=A0







=C2=A0






First=
, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =E2=
=80=9Csteering=E2=80=9D a network, at least not
 initially. There is still traditional config, plus all the existing and em=
erging SDN models. My view has always been that ANIMA sets a =E2=80=9Cdefau=
lt=E2=80=9D policy in a network, and all other means are more specific and =
can override intent. This is described in RFC7575.
 Main point here: ANIMA Intent isn=E2=80=99t the only tool in a network. 


=C2=A0







<jcs>







Absolutely agree. Note that even in the Congress case =
I mentioned







earlier, it is NOT pushing config, and it is NOT viewe=
d as the







"only" policy









</jcs>









=C2=A0







=C2=A0






Inten=
t could perfectly well cover a high level policy such as =E2=80=9Call nodes=
 of type x do
 this; type y does that=E2=80=9D. (Which I think is what you mention below.=
) When we started with Intent, the vision was to keep it very high level, m=
ore along the line =E2=80=9Cdo the right thing=E2=80=9D
J=C2=A0


=C2=A0









<jcs>







Agreed. Although to me, this means that we need to con=
sider the types of actors that are using intent. For example, most of my CC=
IE and JCIE friends have no interest in using an intent language for pushin=
g configuration. On the other hand,
 when I talk to operators and application developers, they would love to be=
 able to use a high-level intent and have the intent engine take care of th=
e details for them.







</jcs>







=C2=A0











It has become clear tha=
t in today=E2=80=99s networking, people will want to push more =E2=80=9Cpar=
ameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll j=
ust have to acknowledge that.







=C2=A0







<jcs>







Sorry Michael, I don't follow here. What do you me=
an







by "parameters"?







</jcs>







=C2=A0









=C2=A0






Perso=
nally I think it=E2=80=99ll pan out like this: Intent will have sections pe=
r autonomic
 function. An Intent parser will validate overall correctness (signature, e=
tc), then segment the Intent into its bits: One part will be the autonomic =
fabric itself, then there will be bits for each autonomic function.




=C2=A0







<jcs>







Do you mean that the intent writer will **know the dif=
ference**







between the fabric and each autonomic function? I hope=
 not,







but perhaps I'm misunderstanding you...<=
/u>







</jcs>







=C2=A0







=C2=A0






So th=
e Intent parser on a node will grab intent, see there is some Intent for au=
tonomic
 function A, will see whether this is running locally, and if so, it=E2=80=
=99ll just pass that piece of Intent to that function.










=C2=A0







<jcs>







Agreed.







</jcs>







=C2=A0









=C2=A0






In th=
is way, a function owner defines both the intent itself, and how it is cons=
umed
 by his/her function. 



=C2=A0







<jcs>







I'm not sure what you mean by "function owner=
". (Sorry to be







picky, I'm likely under-caffeinated.)







=C2=A0







The above seemed like an autonomous decision, made by =
the parser.







I don't think that the intent writer should be cog=
nizant of how many or







what types of autonomic functions are present.<=
u>







</jcs>







=C2=A0







=C2=A0








The downside is that=
 there is no =E2=80=9Ccontrol=E2=80=9D and people can do what they want, in=
 the worst case even push
 config, which is clearly NOT what Intent is meant to do. 



=C2=A0







<jcs>







And hence, the "HAL 2000" scenario. :-)







</jcs>







=C2=A0









=C2=A0







I=E2=80=99d hope that people don=E2=80=99t =
start writing Intent like =E2=80=9Cnode 17: here is your CLI; node 23: here=
 is your CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D. But what I descr=
ibe above COULD be abused in
 this way. RFC7575 says this is NOT Intent. =








=C2=A0







<jcs>







Absolutely agree!







</jcs>







=C2=A0


















=C2=
=A0regards,


John<=
/span>


=
=C2=A0


=C2=A0









=C2=A0





On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mb=
ehring) <mbehrin=
g@cisco.com> wrote:






Hi Zo=
ngpeng,



=C2=
=A0


Yes, =
this is a recurring question and topic. Let me give my view on two levels:



=C2=
=A0


First=
, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D or =E2=
=80=9Csteering=E2=80=9D a network, at least not
 initially. There is still traditional config, plus all the existing and em=
erging SDN models. My view has always been that ANIMA sets a =E2=80=9Cdefau=
lt=E2=80=9D policy in a network, and all other means are more specific and =
can override intent. This is described in RFC7575.
 Main point here: ANIMA Intent isn=E2=80=99t the only tool in a network. 


=C2=
=A0


Inten=
t could perfectly well cover a high level policy such as =E2=80=9Call nodes=
 of type x do
 this; type y does that=E2=80=9D. (Which I think is what you mention below.=
) When we started with Intent, the vision was to keep it very high level, m=
ore along the line =E2=80=9Cdo the right thing=E2=80=9D
J=C2=A0 It has become clear that in today=E2=80=99s net=
working, people will want to push more =E2=80=9Cparameters=E2=80=9D than =
=E2=80=9Cpolicy=E2=80=9D,
 and I think we=E2=80=99ll just have to acknowledge that. 


=C2=
=A0


Perso=
nally I think it=E2=80=99ll pan out like this: Intent will have sections pe=
r autonomic
 function. An Intent parser will validate overall correctness (signature, e=
tc), then segment the Intent into its bits: One part will be the autonomic =
fabric itself, then there will be bits for each autonomic function.



=C2=
=A0


So th=
e Intent parser on a node will grab intent, see there is some Intent for au=
tonomic
 function A, will see whether this is running locally, and if so, it=E2=80=
=99ll just pass that piece of Intent to that function.



=C2=
=A0


In th=
is way, a function owner defines both the intent itself, and how it is cons=
umed
 by his/her function. 


=C2=
=A0


The d=
ownside is that there is no =E2=80=9Ccontrol=E2=80=9D and people can do wha=
t they want, in the
 worst case even push config, which is clearly NOT what Intent is meant to =
do. I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =
=E2=80=9Cnode 17: here is your CLI; node 23: here is your CLI=E2=80=9D; nod=
e 37: here is your CLI=E2=80=9D. But what I describe above COULD be abused
 in this way. RFC7575 says this is NOT Intent. =



=C2=
=A0


In ot=
her words: Let=E2=80=99s not be too prescriptive, and hope that folks will =
do more or less
 reasonable things. 


=C2=
=A0


I=E2=
=80=99ve been meaning to comment on the Intent drafts and possibly contribu=
te, I=E2=80=99m just
 struggling with priorities at the moment (as you can see by my relative si=
lence over the past weeks).



=C2=
=A0


Micha=
el


=C2=
=A0


=C2=
=A0


=C2=
=A0








From: Anima [mailto:anima-bounces@ietf.org]
On Behalf Of Duzongpeng

Sent: 16 March 2016 02:38

To: anima@ietf.o=
rg

Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; J=C3=A9ferson C=
ampos Nobre <j=
cnobre@inf.ufrgs.br>; Sheng Jiang <jiangsheng@huawei.com>

Subject: [Anima] Does Autonomic network need some intervention beyon=
d intent






=C2=A0


Hello, everyone


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 I understand that Autonomic network should be able=
 to work with as less configuration as
 possible. In the ideal case, an Autonomic network should be able to work w=
ell while only intent is needed from human operators.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 However, I wonder whether Autonomic network may ne=
ed some intervention from human operators
 through some low-level policies beyond intents.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 For example, in the

https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it i=
s suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in IP RAN) can be assigned as an "intent". These configuration pa=
rameters need to be distributed in the autonomic
 domain to influence the detail configurations on each autonomic node.


=C2=A0


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0 As intent is described as an abstract, declarative=
, high-level policy used to operate an
 autonomic domain, such as an enterprise network. Perhaps we can rename the=
 =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix-management ID=E2=80=9D to =
some other words.


=C2=A0



Is =E2=80=9CASA parameters" ok here?




=C2=A0



I think some characteristics for this kind of l=
ow-level policies are


1.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Network level parameters, need to be distribute=
d by GRASP and ACP


2.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Relatively static compared to intent, perhaps o=
nly needed to be configured once


3.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Unavoidable in current network environment, mos=
tly for establishing the network infrastructure


4.=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Rarely need coordinations with others parameter=
s, usually is closed related to one specific autonomic function


=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0=C2=A0=C2=A0=C2=A0




Welcome for comment. Your suggestions will be a=
ppreciated.



=C2=A0



=C2=A0


Best Regards


Zongpeng Du











_______________________________________________

Anima mailing list

Anima@ietf.org

h=
ttps://www.ietf.org/mailman/listinfo/anima











-- 







regards,







John



















-- 






re=
gards,






Jo=
hn













_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima





-- 
regards,
John




--001a11c259a83011b9052ef58b11--


From nobody Sat Mar 26 08:54:10 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7383E12D55A for ; Sat, 26 Mar 2016 08:54:08 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EgYUWrzf9qYr for ; Sat, 26 Mar 2016 08:54:04 -0700 (PDT)
Received: from mail-lf0-x22d.google.com (mail-lf0-x22d.google.com [IPv6:2a00:1450:4010:c07::22d]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 859D212D1E4 for ; Sat, 26 Mar 2016 08:54:03 -0700 (PDT)
Received: by mail-lf0-x22d.google.com with SMTP id q73so67432141lfe.2 for ; Sat, 26 Mar 2016 08:54:03 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=lmoavbn8DuPWCzdd3ZJiCXdkw7QnyM5XHCUZyrMi758=; b=XehMsK3Uug3HYPS4MO6FIAJWoXPKjFof8Q9/UO6F9FgdkKJuX0uTA10iMTDofnZclE nVpxYiYnNjMq9lHgECG4KXTqnUqKUHiU1BuFCNlp5oB91sIM99McX2sKd9RpxBW4g3Ih DOYwdsRDYgE0l2LPAA8hKsApryk2cQ1GGfO8OVBkUjKOJmjbs6foB2wUCA1fQSS/UIuB 24nxi7ULsKoxipix4T+tJuDlQ/sC4Y+kwblsU4ZJ4Fz9SPqkxAOHqoPkSqx4SSkWC3Bk 5jNTWzAfNFHIbwZR6NapdSuK73YBJ5chpCFfXvrmAd7IO+1sycP6FOopd0mXOlxhHQRr CT0Q==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=lmoavbn8DuPWCzdd3ZJiCXdkw7QnyM5XHCUZyrMi758=; b=Hqikvj3NJACIRL4grCdvtI5NZr9iZK2ZF2WHnESCL42Xn086MYcz9dm4qepLqDdsnZ x76WVlDtSTKel9zVZsrCfgZ5n/b1KvMI+4qENTqtJAlSZ1XT7PARdP2m1f2VD9DEXpN+ Nfn3jc+JLSNA851CEEJ3oXE1VXgDs2xZdk12sqLymCWWyPYQP709arFy77y1fQifpx9r bXKYqBxHaEqqX1rrPbr1ZTkI/YMLzgrZtdlCbjt7R9HnXhkLFRHPYmWTHx9I1wLM9P6U j/BYCAgGTZWaYSS4MeUjda6wW8CLKawM9564IGpjAc3Amdbm0YihMvmFROrfUjB/w7dT Q2gA==
X-Gm-Message-State: AD7BkJIHGUZvQ4zx5dpSfoDfPql4CG3xyt94ZvQLsixFTgvCpAbZlKkx7cphPx2NJIayons/Sc30ucq+4e9+Pg==
MIME-Version: 1.0
X-Received: by 10.25.143.141 with SMTP id r135mr8049024lfd.100.1459007641663;  Sat, 26 Mar 2016 08:54:01 -0700 (PDT)
Received: by 10.25.22.19 with HTTP; Sat, 26 Mar 2016 08:54:01 -0700 (PDT)
In-Reply-To: <56F60CE4.9040901@gmail.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>     <56F60CE4.9040901@gmail.com>
Date: Sat, 26 Mar 2016 08:54:01 -0700
Message-ID: 
From: John Strassner 
To: Brian E Carpenter , John Strassner 
Content-Type: multipart/alternative; boundary=001a114016e434bca5052ef5b022
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 15:54:08 -0000

--001a114016e434bca5052ef5b022
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

> Correct, and there is also no presumption that individual nodes
> have perfect knowledge of the topology. Of course the ACP
> routing protocol will have perfect knowledge of the ACP topology,
> but individual ASAs only know what they have gleaned from
> discovery and what they have been told by Intent.
>
> It is, I think, a very different world view than the SDN or SUPA or
> NVO4 or NFV view.

Please note that I did NOT say anything about SDN, NFV, or SUPA
here. In fact, I'm confused as to why you both think I did.

I specifically said "There are plenty of southbound APIs (from a
management function to a device). There are very few northbound APIs."

Note the word "management function".

For the record, I don't think that either Phase 2 of NFV or the SDN 1.1
architecture have treated policy in any detail. Both architectures lack a
policy management component. I hope that ANIMA doesn't make the
same mistake.


>> However, in ANIMA network, perhaps we do not have a centralized
>> parsing node (controller), which can replace almost all the
>> communications between nodes.

Earlier I wrote about intent requiring a translation function. I never said
that it was centralized or distributed. In the autonomic architectures I
have built, it was in fact a decentralized (agent-driven) function.
Furthermore, it did not "replace almost all the communications" - it
was used for compilation. There was, in fact, a separate distribution
mechanism (which I have also described), but that was limited to
"just" distribution.

I really don't understand what "replacing communications" means.

regards,
John

On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <
brian.e.carpenter@gmail.com> wrote:

> > However, in ANIMA network, perhaps we do not have a centralized parsing
> node (controller), which can replace almost all the communications betwee=
n
> nodes.
>
> Correct, and there is also no presumption that individual nodes have
> perfect
> knowledge of the topology. Of course the ACP routing protocol will have
> perfect
> knowledge of the ACP topology, but individual ASAs only know what they ha=
ve
> gleaned from discovery and what they have been told by Intent.
>
> It is, I think, a very different world view than the SDN or SUPA or NVO4
> or NFV view.
>
> Regards
>    Brian
>
> On 26/03/2016 16:46, Duzongpeng wrote:
> > Hi John:
> >
> > Thanks for your reply.
> >
> > 
> > How is this different than running a script, or invoking an API? There
> are plenty of southbound APIs (from a management function to a device).
> There are very few northbound APIs. That's what the operators that I have
> talked to want.
> > 
> >
> > I agree that intent belongs to northbound interface in the SDN network.
> And ANIMA network perhaps can share the same northbound interface.
> >
> > But, my understanding is that ANIMA network is a little different from
> SDN network.
> >
> > SDN controller will parse the intent, and transform to detailed
> configurations of every device. It is a good model, and little
> communications between nodes are needed.
> >
> > Some network level parameters can be handled in SDN controller.
> >
> > However, in ANIMA network, perhaps we do not have a centralized parsing
> node (controller), which can replace almost all the communications betwee=
n
> nodes.
> >
> > So when a network operator is operating a ANIMA network, the network
> level parameters are still needed to be flooded.
> >
> > I mean the ANIMA interface is perhaps something like =E2=80=9Cnorthboun=
d + part
> of southbound=E2=80=9D.
> >
> > Best Regards
> > Zongpeng Du
> >
> >
> > From: John Strassner [mailto:strazpdj@gmail.com]
> > Sent: Saturday, March 26, 2016 9:41 AM
> > To: Duzongpeng; John Strassner
> > Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang;
> anima@ietf.org; J=C3=A9ferson Campos Nobre
> > Subject: Re: [Anima] Does Autonomic network need some intervention
> beyond intent
> >
> > Hi Zongpeng,
> >
> > I think we have a misunderstanding here.
> >
> >
> > But I do not support that the intent writer should look every detail in
> the catalog.
> > 
> > Most catalogs that I have seen have very few details. They instead tell
> the user what services or products are available.
> > 
> >
> > The intent writer may mainly care about what they want, but how to do i=
t
> can be left to the autonomic network to deal with.
> > 
> > I agree completely.
> > 
> >
> > Perhaps several solutions exist, and autonomic management system choose
> one according to the abilities collected.
> > 
> > This is one of the reasons why I would like to see capabilities (and
> needs) advertised.
> > 
> >
> > ...
> > I mean in current status, some of the Autonomic Functions should be
> exposed to the network operator, and some of the Autonomic Functions can =
be
> hidden to the network operator
> > ...
> > 
> > How is this different than running a script, or invoking an API? There
> are plenty of southbound APIs (from a management function to a device).
> There are very few northbound APIs. That's what the operators that I have
> talked to want.
> >
> > regards,
> > John
> >
> > On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng  > wrote:
> > Hi John
> >
> > Some personal understanding here. If any misunderstanding, please
> correct me. Thanks.
> >
> > 
> > I would suggest a slight modification, to enable users of
> > intent (and even intent writers) to NOT have to have
> > detailed implementation information.
> >
> > What if all of the Autonomic Functions, after they joined
> > the network, advertised both their capabilities as well as
> > their needs? We can put aside the latter (needs) as this
> > is more complex, but if the capabilities are advertised,
> > then the autonomic management system can assemble
> > those needs into a catalog. Then, the intent writer can
> > look at the catalog and write a declarative request to
> > use the functionality of the catalog.
> >
> > I will look at your YANG model in a bit (sorry, I'm woefully
> > behind...).
> > < /jcs2>
> >
> > I agree that autonomic node should have that self-advertisement ability=
,
> and autonomic management system should have this catalog.
> >
> > But I do not support that the intent writer should look every detail in
> the catalog.
> >
> > The intent writer may mainly care about what they want, but how to do i=
t
> can be left to the autonomic network to deal with.
> > Perhaps several solutions exist, and autonomic management system choose
> one according to the abilities collected.
> >
> > In the ideal, the network operator may communicate with the autonomic
> network using network-level level intent and network-level report.
> > I mean in current status, some of the Autonomic Functions should be
> exposed to the network operator, and some of the Autonomic Functions can =
be
> hidden to the network operator
> > Perhaps in future, when the network nodes become more intelligent, the
> hidden part will increase.
> >
> >
> > Best regards
> > Zongpeng Du
> >
> > From: Anima [mailto:anima-bounces@ietf.org]
> On Behalf Of John Strassner
> > Sent: Thursday, March 24, 2016 10:41 AM
> > To: Michael Behringer (mbehring); John Strassner
> > Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org anima@ietf.org>; J=C3=A9ferson Campos Nobre
> > Subject: Re: [Anima] Does Autonomic network need some intervention
> beyond intent
> >
> > Hi Michael,
> >
> > thanks for your reply. I do think that we largely agree, and would enjo=
y
> chatting during or (hopefully before) BA. Please see ... Als=
o
> trying to unclutter, so I am eliding points that we agree on.
> >
> >
> > While *CIEs don=E2=80=99t like magic, they do understand the concept of=
 a
> default behaviour which they can influence if needed. Any reasonable
> network management approach tries to define =E2=80=9Cdefault=E2=80=9D beh=
aviour and
> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=E2=
=80=9D part very nicely, and *CIEs
> get that.
> > 
> > Agreed!
> > 
> >
> > The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discuss=
ion: My view is that we shouldn=E2=80=99t
> try to define exactly *what* intent allows and doesn=E2=80=99t allow, sin=
ce we=E2=80=99ve
> heard from many sides that the boundary is not clear, and it may be
> impossible to have a really clear, usable definition. So my point was:
> Let=E2=80=99s focus on the distribution model: Intent is flooded to all n=
odes.
> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific nodes=
, etc,
> because that would blur the boundary even worse.
> > 
> > I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1
> was 2006, almost 10 years ago), we used a message bus and pub-sub. In v2,
> we flooded; in v3, we went back to pub-sub because we wanted to do some
> optimization and auditing. But I would be happy to try some type of
> controlled flooding again.
> > 
> > My suggestion: Intent is flooded to all nodes in a domain. The ASA
> owners define the content of what they want to push in their bit of the
> Intent. If we insist of full flooding, we encourage the right behaviour.
> And if someone really puts into his Intent =E2=80=9Cnode A: do this; node=
 B: do the
> other; node C: do the third=E2=80=9D it is awkward enough for people to t=
ake a step
> back and think whether they=E2=80=99re doing the right thing =E2=98=BA
> > 
> > I agree in principle with what you said. Plus, see above.
> > < /jcs2>
> > 
> > Do you mean that the intent writer will **know the difference**
> > between the fabric and each autonomic function? I hope not,
> > but perhaps I'm misunderstanding you...
> > < /jcs>
> > Not sure I understand your concern. Yes, in my model there is a section
> for the fabric, and sections for the various functions. Seems a natural
> structure?!? 
> > The intent **writer** would have to know implementation
> > details in order to place everything in a single file. I would
> > rather not require this, and instead, have the intent engine
> > (e.g., compiler and associated logic) do this.
> > < /jcs2>
> > =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to struct=
ure Intent around
> Autonomic Functions. This is not optimal from a scientific point of view,
> but aligns very well with the way how things are organised. Probably we
> need to discuss that in BA. But refer to my YANG model sent a few days ag=
o.
> > 
> > I would suggest a slight modification, to enable users of
> > intent (and even intent writers) to NOT have to have
> > detailed implementation information.
> >
> > What if all of the Autonomic Functions, after they joined
> > the network, advertised both their capabilities as well as
> > their needs? We can put aside the latter (needs) as this
> > is more complex, but if the capabilities are advertised,
> > then the autonomic management system can assemble
> > those needs into a catalog. Then, the intent writer can
> > look at the catalog and write a declarative request to
> > use the functionality of the catalog.
> >
> > I will look at your YANG model in a bit (sorry, I'm woefully
> > behind...).
> > < /jcs2>
> >
> >
> > regards,
> > John
> >
> > On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <
> mbehring@cisco.com> wrote:
> > John,
> >
> > I think we=E2=80=99re on the same page. Couple of points, top posting h=
ere,
> otherwise it gets too messy.
> >
> > While *CIEs don=E2=80=99t like magic, they do understand the concept of=
 a
> default behaviour which they can influence if needed. Any reasonable
> network management approach tries to define =E2=80=9Cdefault=E2=80=9D beh=
aviour and
> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=E2=
=80=9D part very nicely, and *CIEs
> get that.
> >
> > The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discuss=
ion: My view is that we shouldn=E2=80=99t
> try to define exactly *what* intent allows and doesn=E2=80=99t allow, sin=
ce we=E2=80=99ve
> heard from many sides that the boundary is not clear, and it may be
> impossible to have a really clear, usable definition. So my point was:
> Let=E2=80=99s focus on the distribution model: Intent is flooded to all n=
odes.
> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific nodes=
, etc,
> because that would blur the boundary even worse.
> >
> > My suggestion: Intent is flooded to all nodes in a domain. The ASA
> owners define the content of what they want to push in their bit of the
> Intent. If we insist of full flooding, we encourage the right behaviour.
> And if someone really puts into his Intent =E2=80=9Cnode A: do this; node=
 B: do the
> other; node C: do the third=E2=80=9D it is awkward enough for people to t=
ake a step
> back and think whether they=E2=80=99re doing the right thing =E2=98=BA
> >
> > 
> > Do you mean that the intent writer will **know the difference**
> > between the fabric and each autonomic function? I hope not,
> > but perhaps I'm misunderstanding you...
> > 
> >
> > Not sure I understand your concern. Yes, in my model there is a section
> for the fabric, and sections for the various functions. Seems a natural
> structure?!?
> >
> > =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to struct=
ure Intent around
> Autonomic Functions. This is not optimal from a scientific point of view,
> but aligns very well with the way how things are organised. Probably we
> need to discuss that in BA. But refer to my YANG model sent a few days ag=
o.
> >
> > We should schedule some discussion time outside the ANIMA meeting
> window...
> >
> > Michael
> >
> >
> > From: John Strassner [mailto:strazpdj@gmail.com strazpdj@gmail.com>]
> > Sent: 22 March 2016 21:32
> > To: Michael Behringer (mbehring)  mbehring@cisco.com>>; John Strassner  strazpdj@gmail.com>>
> > Cc: Duzongpeng >;
> anima@ietf.org; Laurent Ciavaglia <
> laurent.ciavaglia@nokia.com>;
> J=C3=A9ferson Campos Nobre >;
> Sheng Jiang >
> > Subject: Re: [Anima] Does Autonomic network need some intervention
> beyond intent
> >
> > In general, I agree with Michael. I'd like to emphasize and expand on a
> couple of key points:
> >
> >
> > First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D=
 or =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
> >
> > 
> > Absolutely agree. Note that even in the Congress case I mentioned
> > earlier, it is NOT pushing config, and it is NOT viewed as the
> > "only" policy
> > 
> >
> >
> > Intent could perfectly well cover a high level policy such as =E2=80=9C=
all nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=98=BA
> >
> > 
> > Agreed. Although to me, this means that we need to consider the types o=
f
> actors that are using intent. For example, most of my CCIE and JCIE frien=
ds
> have no interest in using an intent language for pushing configuration. O=
n
> the other hand, when I talk to operators and application developers, they
> would love to be able to use a high-level intent and have the intent engi=
ne
> take care of the details for them.
> > 
> >
> > It has become clear that in today=E2=80=99s networking, people will wan=
t to push
> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I th=
ink we=E2=80=99ll just have to acknowledge
> that.
> >
> > 
> > Sorry Michael, I don't follow here. What do you mean
> > by "parameters"?
> > 
> >
> >
> > Personally I think it=E2=80=99ll pan out like this: Intent will have se=
ctions
> per autonomic function. An Intent parser will validate overall correctnes=
s
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
> >
> > 
> > Do you mean that the intent writer will **know the difference**
> > between the fabric and each autonomic function? I hope not,
> > but perhaps I'm misunderstanding you...
> > 
> >
> >
> > So the Intent parser on a node will grab intent, see there is some
> Intent for autonomic function A, will see whether this is running locally=
,
> and if so, it=E2=80=99ll just pass that piece of Intent to that function.
> >
> > 
> > Agreed.
> > 
> >
> >
> > In this way, a function owner defines both the intent itself, and how i=
t
> is consumed by his/her function.
> >
> > 
> > I'm not sure what you mean by "function owner". (Sorry to be
> > picky, I'm likely under-caffeinated.)
> >
> > The above seemed like an autonomous decision, made by the parser.
> > I don't think that the intent writer should be cognizant of how many or
> > what types of autonomic functions are present.
> > 
> >
> >
> > The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people c=
an do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do.
> >
> > 
> > And hence, the "HAL 2000" scenario. :-)
> > 
> >
> >
> > I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =
=E2=80=9Cnode 17: here is
> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CLI=
=E2=80=9D. But what
> I describe above COULD be abused in this way. RFC7575 says this is NOT
> Intent.
> >
> > 
> > Absolutely agree!
> > 
> >
> >  regards,
> > John
> >
> >
> >
> > On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
> mbehring@cisco.com> wrote:
> > Hi Zongpeng,
> >
> > Yes, this is a recurring question and topic. Let me give my view on two
> levels:
> >
> > First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D=
 or =E2=80=9Csteering=E2=80=9D a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA sets=
 a
> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are mo=
re specific and
> can override intent. This is described in RFC7575. Main point here: ANIMA
> Intent isn=E2=80=99t the only tool in a network.
> >
> > Intent could perfectly well cover a high level policy such as =E2=80=9C=
all nodes
> of type x do this; type y does that=E2=80=9D. (Which I think is what you =
mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=98=BA=
  It has become clear that
> in today=E2=80=99s networking, people will want to push more =E2=80=9Cpar=
ameters=E2=80=9D than
> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowle=
dge that.
> >
> > Personally I think it=E2=80=99ll pan out like this: Intent will have se=
ctions
> per autonomic function. An Intent parser will validate overall correctnes=
s
> (signature, etc), then segment the Intent into its bits: One part will be
> the autonomic fabric itself, then there will be bits for each autonomic
> function.
> >
> > So the Intent parser on a node will grab intent, see there is some
> Intent for autonomic function A, will see whether this is running locally=
,
> and if so, it=E2=80=99ll just pass that piece of Intent to that function.
> >
> > In this way, a function owner defines both the intent itself, and how i=
t
> is consumed by his/her function.
> >
> > The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people c=
an do what they
> want, in the worst case even push config, which is clearly NOT what Inten=
t
> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writing =
Intent like =E2=80=9Cnode
> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: here i=
s your
> CLI=E2=80=9D. But what I describe above COULD be abused in this way. RFC7=
575 says
> this is NOT Intent.
> >
> > In other words: Let=E2=80=99s not be too prescriptive, and hope that fo=
lks will
> do more or less reasonable things.
> >
> > I=E2=80=99ve been meaning to comment on the Intent drafts and possibly
> contribute, I=E2=80=99m just struggling with priorities at the moment (as=
 you can
> see by my relative silence over the past weeks).
> >
> > Michael
> >
> >
> >
> > From: Anima [mailto:anima-bounces@ietf.org]
> On Behalf Of Duzongpeng
> > Sent: 16 March 2016 02:38
> > To: anima@ietf.org
> > Cc: Laurent Ciavaglia  laurent.ciavaglia@nokia.com>>; J=C3=A9ferson Campos Nobre  >; Sheng Jiang  jiangsheng@huawei.com>>
> > Subject: [Anima] Does Autonomic network need some intervention beyond
> intent
> >
> > Hello, everyone
> >
> >          I understand that Autonomic network should be able to work wit=
h
> as less configuration as possible. In the ideal case, an Autonomic networ=
k
> should be able to work well while only intent is needed from human
> operators.
> >
> >          However, I wonder whether Autonomic network may need some
> intervention from human operators through some low-level policies beyond
> intents.
> >
> >          For example, in the
> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it is
> suggested that the prefix lengths for the CSG, ASG, RSG (different roles =
in
> IP RAN) can be assigned as an "intent". These configuration parameters ne=
ed
> to be distributed in the autonomic domain to influence the detail
> configurations on each autonomic node.
> >
> >          As intent is described as an abstract, declarative, high-level
> policy used to operate an autonomic domain, such as an enterprise network=
.
> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cprefix=
-management ID=E2=80=9D to some
> other words.
> >
> > Is =E2=80=9CASA parameters" ok here?
> >
> > I think some characteristics for this kind of low-level policies are
> >
> > 1.       Network level parameters, need to be distributed by GRASP and
> ACP
> >
> > 2.       Relatively static compared to intent, perhaps only needed to b=
e
> configured once
> >
> > 3.       Unavoidable in current network environment, mostly for
> establishing the network infrastructure
> >
> > 4.       Rarely need coordinations with others parameters, usually is
> closed related to one specific autonomic function
> >
> > Welcome for comment. Your suggestions will be appreciated.
> >
> >
> > Best Regards
> > Zongpeng Du
> >
> > _______________________________________________
> > Anima mailing list
> > Anima@ietf.org
> > https://www.ietf.org/mailman/listinfo/anima
> >
> >
> >
> > --
> > regards,
> > John
> >
> >
> >
> > --
> > regards,
> > John
> >
> >
> >
> > --
> > regards,
> > John
> >
> >
> >
> > _______________________________________________
> > Anima mailing list
> > Anima@ietf.org
> > https://www.ietf.org/mailman/listinfo/anima
> >
>
>


--=20
regards,
John

--001a114016e434bca5052ef5b022
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


> Correct, and there is also no presumption that i=
ndividual nodes
>=C2=A0have perfect knowledge of the topology.=
 Of course the ACP
> routing protocol will have perfect knowle=
dge of the ACP topology,
>=C2=A0but individual ASAs only know =
what they have gleaned from
>=C2=A0discovery and what they hav=
e been told by Intent.
>
> It is, I think, a very different wor=
ld view than the SDN or SUPA or
>=C2=A0NVO4 or NFV view.
=


Please note that I did NOT say anything about SDN, NFV,=
 or SUPA
here. In fact, I'm confused as to why you both think=
 I did.

I specifically said "There are plenty=
 of southbound APIs (from a
management function to a device). The=
re are very few northbound APIs."

Note the wo=
rd "management function".

For the record=
, I don't think that either Phase 2 of NFV or the SDN 1.1
arc=
hitecture have treated policy in any detail. Both architectures lack a
policy management component. I hope that ANIMA doesn't make the
same mistake.


>> How=
ever, in ANIMA network, perhaps we do not have a centralized
>=
>=C2=A0parsing node (controller), which can replace almost all the
=

>> communications between nodes.

E=
arlier I wrote about intent requiring a translation function. I never said<=
/div>
that it was centralized or distributed. In the autonomic architec=
tures I
have built, it was in fact a decentralized (agent-driven)=
 function.
Furthermore, it did not "replace almost all the c=
ommunications" - it
was used for compilation. There was, in =
fact, a separate distribution
mechanism (which I have also descri=
bed), but that was limited to
"just" distribution.

I really don't understand what "replacing com=
munications" means.

regards,
John

O=
n Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <brian.e.carpe=
nter@gmail.com> wrote:
>=
 However, in ANIMA network, perhaps we do not have a centralized parsing no=
de (controller), which can replace almost all the communications between no=
des.



Correct, and there is also no presumption that individual nodes have perfec=
t

knowledge of the topology. Of course the ACP routing protocol will have per=
fect

knowledge of the ACP topology, but individual ASAs only know what they have=


gleaned from discovery and what they have been told by Intent.



It is, I think, a very different world view than the SDN or SUPA or NVO4 or=
 NFV view.



Regards

=C2=A0 =C2=A0Brian



On 26/03/2016 16:46, Duzongpeng wrote:

> Hi John:

>

> Thanks for your reply.

>

> <jcs>

> How is this different than running a script, or invoking an API? There=
 are plenty of southbound APIs (from a management function to a device). Th=
ere are very few northbound APIs. That's what the operators that I have=
 talked to want.

> </jcs>

>

> I agree that intent belongs to northbound interface in the SDN network=
. And ANIMA network perhaps can share the same northbound interface.

>

> But, my understanding is that ANIMA network is a little different from=
 SDN network.

>

> SDN controller will parse the intent, and transform to detailed config=
urations of every device. It is a good model, and little communications bet=
ween nodes are needed.

>

> Some network level parameters can be handled in SDN controller.

>

> However, in ANIMA network, perhaps we do not have a centralized parsin=
g node (controller), which can replace almost all the communications betwee=
n nodes.

>

> So when a network operator is operating a ANIMA network, the network l=
evel parameters are still needed to be flooded.

>

> I mean the ANIMA interface is perhaps something like =E2=80=9Cnorthbou=
nd + part of southbound=E2=80=9D.

>

> Best Regards

> Zongpeng Du

>

>

> From: John Strassner [mailto:str=
azpdj@gmail.com]

> Sent: Saturday, March 26, 2016 9:41 AM

> To: Duzongpeng; John Strassner

> Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang; anima@ietf.org; J=C3=A9ferson Campos Nobr=
e

> Subject: Re: [Anima] Does Autonomic network need some intervention bey=
ond intent

>

> Hi Zongpeng,

>

> I think we have a misunderstanding here.

>

>

> But I do not support that the intent writer should look every detail i=
n the catalog.

> <jcs>

> Most catalogs that I have seen have very few details. They instead tel=
l the user what services or products are available.

> </jcs>

>

> The intent writer may mainly care about what they want, but how to do =
it can be left to the autonomic network to deal with.

> <jcs>

> I agree completely.

> </jcs>

>

> Perhaps several solutions exist, and autonomic management system choos=
e one according to the abilities collected.

> <jcs>

> This is one of the reasons why I would like to see capabilities (and n=
eeds) advertised.

> </jcs>

>

> ...

> I mean in current status, some of the Autonomic Functions should be ex=
posed to the network operator, and some of the Autonomic Functions can be h=
idden to the network operator

> ...

> <jcs>

> How is this different than running a script, or invoking an API? There=
 are plenty of southbound APIs (from a management function to a device). Th=
ere are very few northbound APIs. That's what the operators that I have=
 talked to want.

>

> regards,

> John

>

> On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng <duzongpeng@huawei.com<mailto:duzongpeng@huawei.com>> wrote:

> Hi John

>

> Some personal understanding here. If any misunderstanding, please corr=
ect me. Thanks.

>

> <jcs2>

> I would suggest a slight modification, to enable users of

> intent (and even intent writers) to NOT have to have

> detailed implementation information.

>

> What if all of the Autonomic Functions, after they joined

> the network, advertised both their capabilities as well as

> their needs? We can put aside the latter (needs) as this

> is more complex, but if the capabilities are advertised,

> then the autonomic management system can assemble

> those needs into a catalog. Then, the intent writer can

> look at the catalog and write a declarative request to

> use the functionality of the catalog.

>

> I will look at your YANG model in a bit (sorry, I'm woefully

> behind...).

> < /jcs2>

>

> I agree that autonomic node should have that self-advertisement abilit=
y, and autonomic management system should have this catalog.

>

> But I do not support that the intent writer should look every detail i=
n the catalog.

>

> The intent writer may mainly care about what they want, but how to do =
it can be left to the autonomic network to deal with.

> Perhaps several solutions exist, and autonomic management system choos=
e one according to the abilities collected.

>

> In the ideal, the network operator may communicate with the autonomic =
network using network-level level intent and network-level report.

> I mean in current status, some of the Autonomic Functions should be ex=
posed to the network operator, and some of the Autonomic Functions can be h=
idden to the network operator

> Perhaps in future, when the network nodes become more intelligent, the=
 hidden part will increase.

>

>

> Best regards

> Zongpeng Du

>

> From: Anima [mailto:anima-bo=
unces@ietf.org<mailto:anim=
a-bounces@ietf.org>] On Behalf Of John Strassner

> Sent: Thursday, March 24, 2016 10:41 AM

> To: Michael Behringer (mbehring); John Strassner

> Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org<mailto:=
anima@ietf.org>; J=C3=A9ferson Campos Nobre

> Subject: Re: [Anima] Does Autonomic network need some intervention bey=
ond intent

>

> Hi Michael,

>

> thanks for your reply. I do think that we largely agree, and would enj=
oy chatting during or (hopefully before) BA. Please see <jcs2>..</=
jcs2>. Also trying to unclutter, so I am eliding points that we agree on=
.

>

>

> While *CIEs don=E2=80=99t like magic, they do understand the concept o=
f a default behaviour which they can influence if needed. Any reasonable ne=
twork management approach tries to define =E2=80=9Cdefault=E2=80=9D behavio=
ur and =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefaul=
t=E2=80=9D part very nicely, and *CIEs get that.

> <jcs2>

> Agreed!

> </jcs2>

>

> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discus=
sion: My view is that we shouldn=E2=80=99t try to define exactly *what* int=
ent allows and doesn=E2=80=99t allow, since we=E2=80=99ve heard from many s=
ides that the boundary is not clear, and it may be impossible to have a rea=
lly clear, usable definition. So my point was: Let=E2=80=99s focus on the d=
istribution model: Intent is flooded to all nodes. Let=E2=80=99s NOT do sel=
ective flooding, unicast Intent to specific nodes, etc, because that would =
blur the boundary even worse.

> <jcs2>

> I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1=
 was 2006, almost 10 years ago), we used a message bus and pub-sub. In v2, =
we flooded; in v3, we went back to pub-sub because we wanted to do some opt=
imization and auditing. But I would be happy to try some type of controlled=
 flooding again.

> </jcs2>

> My suggestion: Intent is flooded to all nodes in a domain. The ASA own=
ers define the content of what they want to push in their bit of the Intent=
. If we insist of full flooding, we encourage the right behaviour. And if s=
omeone really puts into his Intent =E2=80=9Cnode A: do this; node B: do the=
 other; node C: do the third=E2=80=9D it is awkward enough for people to ta=
ke a step back and think whether they=E2=80=99re doing the right thing =E2=
=98=BA

> <jcs2>

> I agree in principle with what you said. Plus, see above.

> < /jcs2>

> <jcs>

> Do you mean that the intent writer will **know the difference**

> between the fabric and each autonomic function? I hope not,

> but perhaps I'm misunderstanding you...

> < /jcs>

> Not sure I understand your concern. Yes, in my model there is a sectio=
n for the fabric, and sections for the various functions. Seems a natural s=
tructure?!? <jcs2>

> The intent **writer** would have to know implementation

> details in order to place everything in a single file. I would

> rather not require this, and instead, have the intent engine

> (e.g., compiler and associated logic) do this.

> < /jcs2>

> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to struc=
ture Intent around Autonomic Functions. This is not optimal from a scientif=
ic point of view, but aligns very well with the way how things are organise=
d. Probably we need to discuss that in BA. But refer to my YANG model sent =
a few days ago.

> <jcs2>

> I would suggest a slight modification, to enable users of

> intent (and even intent writers) to NOT have to have

> detailed implementation information.

>

> What if all of the Autonomic Functions, after they joined

> the network, advertised both their capabilities as well as

> their needs? We can put aside the latter (needs) as this

> is more complex, but if the capabilities are advertised,

> then the autonomic management system can assemble

> those needs into a catalog. Then, the intent writer can

> look at the catalog and write a declarative request to

> use the functionality of the catalog.

>

> I will look at your YANG model in a bit (sorry, I'm woefully

> behind...).

> < /jcs2>

>

>

> regards,

> John

>

> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <mbehring@cisco.com<mailto:mbehring@cisco.com>> wrote:

> John,

>

> I think we=E2=80=99re on the same page. Couple of points, top posting =
here, otherwise it gets too messy.

>

> While *CIEs don=E2=80=99t like magic, they do understand the concept o=
f a default behaviour which they can influence if needed. Any reasonable ne=
twork management approach tries to define =E2=80=9Cdefault=E2=80=9D behavio=
ur and =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefaul=
t=E2=80=9D part very nicely, and *CIEs get that.

>

> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discus=
sion: My view is that we shouldn=E2=80=99t try to define exactly *what* int=
ent allows and doesn=E2=80=99t allow, since we=E2=80=99ve heard from many s=
ides that the boundary is not clear, and it may be impossible to have a rea=
lly clear, usable definition. So my point was: Let=E2=80=99s focus on the d=
istribution model: Intent is flooded to all nodes. Let=E2=80=99s NOT do sel=
ective flooding, unicast Intent to specific nodes, etc, because that would =
blur the boundary even worse.

>

> My suggestion: Intent is flooded to all nodes in a domain. The ASA own=
ers define the content of what they want to push in their bit of the Intent=
. If we insist of full flooding, we encourage the right behaviour. And if s=
omeone really puts into his Intent =E2=80=9Cnode A: do this; node B: do the=
 other; node C: do the third=E2=80=9D it is awkward enough for people to ta=
ke a step back and think whether they=E2=80=99re doing the right thing =E2=
=98=BA

>

> <jcs>

> Do you mean that the intent writer will **know the difference**

> between the fabric and each autonomic function? I hope not,

> but perhaps I'm misunderstanding you...

> </jcs>

>

> Not sure I understand your concern. Yes, in my model there is a sectio=
n for the fabric, and sections for the various functions. Seems a natural s=
tructure?!?

>

> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to struc=
ture Intent around Autonomic Functions. This is not optimal from a scientif=
ic point of view, but aligns very well with the way how things are organise=
d. Probably we need to discuss that in BA. But refer to my YANG model sent =
a few days ago.

>

> We should schedule some discussion time outside the ANIMA meeting wind=
ow...

>

> Michael

>

>

> From: John Strassner [mailto:str=
azpdj@gmail.com<mailto:strazpd=
j@gmail.com>]

> Sent: 22 March 2016 21:32

> To: Michael Behringer (mbehring) <mbehring@cisco.com<mailto:mbehring@cisco.com>>; John Strassner <strazpdj@gmail.com<mailto:strazpdj@gmail.com>>

> Cc: Duzongpeng <duzongpeng=
@huawei.com<mailto:duzongpe=
ng@huawei.com>>; anima@ietf.org=
<mailto:anima@ietf.org>; La=
urent Ciavaglia <laurent.=
ciavaglia@nokia.com<mailto:laurent.ciavaglia@nokia.com>>; J=C3=A9ferson Campos Nobre &=
lt;jcnobre@inf.ufrgs.br<mail=
to:jcnobre@inf.ufrgs.br>>=
; Sheng Jiang <jiangsheng@huawe=
i.com<mailto:jiangsheng@hua=
wei.com>>

> Subject: Re: [Anima] Does Autonomic network need some intervention bey=
ond intent

>

> In general, I agree with Michael. I'd like to emphasize and expand=
 on a couple of key points:

>

>

> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=
=9D or =E2=80=9Csteering=E2=80=9D a network, at least not initially. There =
is still traditional config, plus all the existing and emerging SDN models.=
 My view has always been that ANIMA sets a =E2=80=9Cdefault=E2=80=9D policy=
 in a network, and all other means are more specific and can override inten=
t. This is described in RFC7575. Main point here: ANIMA Intent isn=E2=80=99=
t the only tool in a network.

>

> <jcs>

> Absolutely agree. Note that even in the Congress case I mentioned

> earlier, it is NOT pushing config, and it is NOT viewed as the

> "only" policy

> </jcs>

>

>

> Intent could perfectly well cover a high level policy such as =E2=80=
=9Call nodes of type x do this; type y does that=E2=80=9D. (Which I think i=
s what you mention below.) When we started with Intent, the vision was to k=
eep it very high level, more along the line =E2=80=9Cdo the right thing=E2=
=80=9D =E2=98=BA

>

> <jcs>

> Agreed. Although to me, this means that we need to consider the types =
of actors that are using intent. For example, most of my CCIE and JCIE frie=
nds have no interest in using an intent language for pushing configuration.=
 On the other hand, when I talk to operators and application developers, th=
ey would love to be able to use a high-level intent and have the intent eng=
ine take care of the details for them.

> </jcs>

>

> It has become clear that in today=E2=80=99s networking, people will wa=
nt to push more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D,=
 and I think we=E2=80=99ll just have to acknowledge that.

>

> <jcs>

> Sorry Michael, I don't follow here. What do you mean

> by "parameters"?

> </jcs>

>

>

> Personally I think it=E2=80=99ll pan out like this: Intent will have s=
ections per autonomic function. An Intent parser will validate overall corr=
ectness (signature, etc), then segment the Intent into its bits: One part w=
ill be the autonomic fabric itself, then there will be bits for each autono=
mic function.

>

> <jcs>

> Do you mean that the intent writer will **know the difference**

> between the fabric and each autonomic function? I hope not,

> but perhaps I'm misunderstanding you...

> </jcs>

>

>

> So the Intent parser on a node will grab intent, see there is some Int=
ent for autonomic function A, will see whether this is running locally, and=
 if so, it=E2=80=99ll just pass that piece of Intent to that function.

>

> <jcs>

> Agreed.

> </jcs>

>

>

> In this way, a function owner defines both the intent itself, and how =
it is consumed by his/her function.

>

> <jcs>

> I'm not sure what you mean by "function owner". (Sorry t=
o be

> picky, I'm likely under-caffeinated.)

>

> The above seemed like an autonomous decision, made by the parser.

> I don't think that the intent writer should be cognizant of how ma=
ny or

> what types of autonomic functions are present.

> </jcs>

>

>

> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people =
can do what they want, in the worst case even push config, which is clearly=
 NOT what Intent is meant to do.

>

> <jcs>

> And hence, the "HAL 2000" scenario. :-)

> </jcs>

>

>

> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =
=E2=80=9Cnode 17: here is your CLI; node 23: here is your CLI=E2=80=9D; nod=
e 37: here is your CLI=E2=80=9D. But what I describe above COULD be abused =
in this way. RFC7575 says this is NOT Intent.

>

> <jcs>

> Absolutely agree!

> </jcs>

>

>=C2=A0 regards,

> John

>

>

>

> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <mbehring@cisco.com<mailto:mbehring@cisco.com>> wrote:

> Hi Zongpeng,

>

> Yes, this is a recurring question and topic. Let me give my view on tw=
o levels:

>

> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=
=9D or =E2=80=9Csteering=E2=80=9D a network, at least not initially. There =
is still traditional config, plus all the existing and emerging SDN models.=
 My view has always been that ANIMA sets a =E2=80=9Cdefault=E2=80=9D policy=
 in a network, and all other means are more specific and can override inten=
t. This is described in RFC7575. Main point here: ANIMA Intent isn=E2=80=99=
t the only tool in a network.

>

> Intent could perfectly well cover a high level policy such as =E2=80=
=9Call nodes of type x do this; type y does that=E2=80=9D. (Which I think i=
s what you mention below.) When we started with Intent, the vision was to k=
eep it very high level, more along the line =E2=80=9Cdo the right thing=E2=
=80=9D =E2=98=BA=C2=A0 It has become clear that in today=E2=80=99s networki=
ng, people will want to push more =E2=80=9Cparameters=E2=80=9D than =E2=80=
=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to acknowledge that=
.

>

> Personally I think it=E2=80=99ll pan out like this: Intent will have s=
ections per autonomic function. An Intent parser will validate overall corr=
ectness (signature, etc), then segment the Intent into its bits: One part w=
ill be the autonomic fabric itself, then there will be bits for each autono=
mic function.

>

> So the Intent parser on a node will grab intent, see there is some Int=
ent for autonomic function A, will see whether this is running locally, and=
 if so, it=E2=80=99ll just pass that piece of Intent to that function.

>

> In this way, a function owner defines both the intent itself, and how =
it is consumed by his/her function.

>

> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people =
can do what they want, in the worst case even push config, which is clearly=
 NOT what Intent is meant to do. I=E2=80=99d hope that people don=E2=80=99t=
 start writing Intent like =E2=80=9Cnode 17: here is your CLI; node 23: her=
e is your CLI=E2=80=9D; node 37: here is your CLI=E2=80=9D. But what I desc=
ribe above COULD be abused in this way. RFC7575 says this is NOT Intent.
>

> In other words: Let=E2=80=99s not be too prescriptive, and hope that f=
olks will do more or less reasonable things.

>

> I=E2=80=99ve been meaning to comment on the Intent drafts and possibly=
 contribute, I=E2=80=99m just struggling with priorities at the moment (as =
you can see by my relative silence over the past weeks).

>

> Michael

>

>

>

> From: Anima [mailto:anima-bo=
unces@ietf.org<mailto:anim=
a-bounces@ietf.org>] On Behalf Of Duzongpeng

> Sent: 16 March 2016 02:38

> To: anima@ietf.org<mailto:anima@ietf.org>

> Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com<mailto:laurent.ciavaglia@nokia.com>>; J=C3=A9ferson Cam=
pos Nobre <jcnobre@inf.ufrgs.br<=
/a><mailto:jcnobre@inf.ufrgs.br<=
/a>>>; Sheng Jiang <jiang=
sheng@huawei.com<mailto:jia=
ngsheng@huawei.com>>

> Subject: [Anima] Does Autonomic network need some intervention beyond =
intent

>

> Hello, everyone

>

>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 I understand that Autonomic network =
should be able to work with as less configuration as possible. In the ideal=
 case, an Autonomic network should be able to work well while only intent i=
s needed from human operators.

>

>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 However, I wonder whether Autonomic =
network may need some intervention from human operators through some low-le=
vel policies beyond intents.

>

>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 For example, in the https://tools.ietf.org/html/draft-ietf-anima-prefix=
-management-00, it is suggested that the prefix lengths for the CSG, AS=
G, RSG (different roles in IP RAN) can be assigned as an "intent"=
. These configuration parameters need to be distributed in the autonomic do=
main to influence the detail configurations on each autonomic node.

>

>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 As intent is described as an abstrac=
t, declarative, high-level policy used to operate an autonomic domain, such=
 as an enterprise network. Perhaps we can rename the =E2=80=9Cintent=E2=80=
=9D in the =E2=80=9Cprefix-management ID=E2=80=9D to some other words.

>

> Is =E2=80=9CASA parameters" ok here?

>

> I think some characteristics for this kind of low-level policies are
>

> 1.=C2=A0 =C2=A0 =C2=A0 =C2=A0Network level parameters, need to be dist=
ributed by GRASP and ACP

>

> 2.=C2=A0 =C2=A0 =C2=A0 =C2=A0Relatively static compared to intent, per=
haps only needed to be configured once

>

> 3.=C2=A0 =C2=A0 =C2=A0 =C2=A0Unavoidable in current network environmen=
t, mostly for establishing the network infrastructure

>

> 4.=C2=A0 =C2=A0 =C2=A0 =C2=A0Rarely need coordinations with others par=
ameters, usually is closed related to one specific autonomic function

>

> Welcome for comment. Your suggestions will be appreciated.

>

>

> Best Regards

> Zongpeng Du

>

> _______________________________________________

> Anima mailing list

> Anima@ietf.org<mailto:Anima@ietf.org>

> https://www.ietf.org/mailman/listinfo/anima

>

>

>

> --

> regards,

> John

>

>

>

> --

> regards,

> John

>

>

>

> --

> regards,

> John

>

>

>

> _______________________________________________

> Anima mailing list

> Anima@ietf.org

> https://www.ietf.org/mailman/listinfo/anima

>







-- 
regards,
John




--001a114016e434bca5052ef5b022--


From nobody Sat Mar 26 11:53:18 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A3DD512D547 for ; Sat, 26 Mar 2016 11:53:17 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id E5_hPt4RgqDa for ; Sat, 26 Mar 2016 11:53:14 -0700 (PDT)
Received: from mail-pf0-x22c.google.com (mail-pf0-x22c.google.com [IPv6:2607:f8b0:400e:c00::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9D10F12D0B8 for ; Sat, 26 Mar 2016 11:53:14 -0700 (PDT)
Received: by mail-pf0-x22c.google.com with SMTP id x3so105464133pfb.1 for ; Sat, 26 Mar 2016 11:53:14 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=SXcq4rlfbBdEjhhHhYL0csIqfzFi7z49TcYs7D3mQTk=; b=WZwENb7KwjJgp7CCuX1aaFKwMr/1IBvuOMQrWwhIjlO0iDxzU7REs6ji3WkTVipuMQ PCyTMZ96k4N9jp+SZfAyDZrb/ejKkLUhBr+ubv6mIZBcOVMEbDr+DSsn8BHwCWULttV6 d9udsJ/ReT9PgIQiPaVoqOMyilpGvldI58A4tTcPwvagtwwZXXrQNhFryT5GyuY4GOwM fiLpFo24akWQQWq2wzqfJL4wGA8qHdwBFrCiGIyA6B7/4BjiLSJe2EPp2nXOo+KMEjI2 kPrPE/kC6l9Wn7Ue3TSdNDJgfD2fo6iN8SE5Btm/3S8Hhgev1CnqHlLFj5nvAxiJt4J5 1spQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=SXcq4rlfbBdEjhhHhYL0csIqfzFi7z49TcYs7D3mQTk=; b=gAmrvKn+aXeq/dgsvWaZrQtTlYVV7+Z6wqywC5QZh5W35guNbvW28noFDUESBHV9Z7 ueV7qXoPvH/i7EPvMgITWWhYUHwDngpN/wMSMf8LnzCKLsdXrdKFlK9a2EOxjxeIG1m5 hqysMkVgL3MYwnCJrOIEE476pLdnmxFojWf6fkJGTvctKhAIPLTUeNr7q8pzxG9BB2Fo 1p2dJz8pcZC81gGyw7z5WEOywqWPC2N7AS0WLeMwl9kpZ7PNaTxdvMoEaUiIJrxpmgZb 1uLI2NQfMbvIH8XwzsS0inFPPZc4VyDXOFkWBdjf/oxTNhsdCK07yZhb/DuXTM0d02QE M4nA==
X-Gm-Message-State: AD7BkJIjg8/E5C7A8wtpbaH1P69yS3sEVGSBDHqtIhsYM6zK/bqro+LWPIxE5npBmRkjYw==
X-Received: by 10.98.17.78 with SMTP id z75mr30952656pfi.40.1459018393899; Sat, 26 Mar 2016 11:53:13 -0700 (PDT)
Received: from ?IPv6:2406:e007:738b:1:28cc:dc4c:9703:6781? ([2406:e007:738b:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id o71sm24654521pfj.68.2016.03.26.11.53.08 (version=TLSv1/SSLv3 cipher=OTHER); Sat, 26 Mar 2016 11:53:12 -0700 (PDT)
To: John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>     <56F60CE4.9040901@gmail.com> 
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F6DAA1.8000206@gmail.com>
Date: Sun, 27 Mar 2016 07:53:21 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 18:53:17 -0000

Hi John,
On 27/03/2016 04:54, John Strassner wrote:
>> Correct, and there is also no presumption that individual nodes
>> have perfect knowledge of the topology. Of course the ACP
>> routing protocol will have perfect knowledge of the ACP topology,
>> but individual ASAs only know what they have gleaned from
>> discovery and what they have been told by Intent.
>>
>> It is, I think, a very different world view than the SDN or SUPA or
>> NVO4 or NFV view.
>=20
> Please note that I did NOT say anything about SDN, NFV, or SUPA
> here. In fact, I'm confused as to why you both think I did.

I can't speak for Zongpeng, but I was triggered by the mention of
northbound and southbound APIs. There is no presumed direction in
Anima relationships, since in the limit a network is self-forming;
all relationships are peer-to-peer.

Of course, when we put Intent into the picture, there's a presumption
that Intent comes from some authority such as a NOC, but that's an
external semantic, not an assumption of the autonomic model.

> I specifically said "There are plenty of southbound APIs (from a
> management function to a device). There are very few northbound APIs."
>=20
> Note the word "management function".
>=20
> For the record, I don't think that either Phase 2 of NFV or the SDN 1.1=

> architecture have treated policy in any detail. Both architectures lack=
 a
> policy management component. I hope that ANIMA doesn't make the
> same mistake.

Certainly, Intent needs a number of properties such as self-consistency
and stability that imply it is managed in some way.

>>> However, in ANIMA network, perhaps we do not have a centralized
>>> parsing node (controller), which can replace almost all the
>>> communications between nodes.
>=20
> Earlier I wrote about intent requiring a translation function. I never =
said
> that it was centralized or distributed. In the autonomic architectures =
I
> have built, it was in fact a decentralized (agent-driven) function.
> Furthermore, it did not "replace almost all the communications" - it
> was used for compilation. There was, in fact, a separate distribution
> mechanism (which I have also described), but that was limited to
> "just" distribution.
>=20
> I really don't understand what "replacing communications" means.

I think the point is that AN *adds* a new mode of communication between
self-managing nodes, that doesn't exist in a strictly north/south
scenario. That is meant to reduce the amount of north/south communication=

and reduce the amount of centralized work. So there's a trade-off -
more autonomic messaging means less north-south messaging.

Best regards
   Brian

>=20
> regards,
> John
>=20
> On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <
> brian.e.carpenter@gmail.com> wrote:
>=20
>>> However, in ANIMA network, perhaps we do not have a centralized parsi=
ng
>> node (controller), which can replace almost all the communications bet=
ween
>> nodes.
>>
>> Correct, and there is also no presumption that individual nodes have
>> perfect
>> knowledge of the topology. Of course the ACP routing protocol will hav=
e
>> perfect
>> knowledge of the ACP topology, but individual ASAs only know what they=
 have
>> gleaned from discovery and what they have been told by Intent.
>>
>> It is, I think, a very different world view than the SDN or SUPA or NV=
O4
>> or NFV view.
>>
>> Regards
>>    Brian
>>
>> On 26/03/2016 16:46, Duzongpeng wrote:
>>> Hi John:
>>>
>>> Thanks for your reply.
>>>
>>> 
>>> How is this different than running a script, or invoking an API? Ther=
e
>> are plenty of southbound APIs (from a management function to a device)=
=2E
>> There are very few northbound APIs. That's what the operators that I h=
ave
>> talked to want.
>>> 
>>>
>>> I agree that intent belongs to northbound interface in the SDN networ=
k.
>> And ANIMA network perhaps can share the same northbound interface.
>>>
>>> But, my understanding is that ANIMA network is a little different fro=
m
>> SDN network.
>>>
>>> SDN controller will parse the intent, and transform to detailed
>> configurations of every device. It is a good model, and little
>> communications between nodes are needed.
>>>
>>> Some network level parameters can be handled in SDN controller.
>>>
>>> However, in ANIMA network, perhaps we do not have a centralized parsi=
ng
>> node (controller), which can replace almost all the communications bet=
ween
>> nodes.
>>>
>>> So when a network operator is operating a ANIMA network, the network
>> level parameters are still needed to be flooded.
>>>
>>> I mean the ANIMA interface is perhaps something like =E2=80=9Cnorthbo=
und + part
>> of southbound=E2=80=9D.
>>>
>>> Best Regards
>>> Zongpeng Du
>>>
>>>
>>> From: John Strassner [mailto:strazpdj@gmail.com]
>>> Sent: Saturday, March 26, 2016 9:41 AM
>>> To: Duzongpeng; John Strassner
>>> Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang;
>> anima@ietf.org; J=C3=A9ferson Campos Nobre
>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>>
>>> Hi Zongpeng,
>>>
>>> I think we have a misunderstanding here.
>>>
>>>
>>> But I do not support that the intent writer should look every detail =
in
>> the catalog.
>>> 
>>> Most catalogs that I have seen have very few details. They instead te=
ll
>> the user what services or products are available.
>>> 
>>>
>>> The intent writer may mainly care about what they want, but how to do=
 it
>> can be left to the autonomic network to deal with.
>>> 
>>> I agree completely.
>>> 
>>>
>>> Perhaps several solutions exist, and autonomic management system choo=
se
>> one according to the abilities collected.
>>> 
>>> This is one of the reasons why I would like to see capabilities (and
>> needs) advertised.
>>> 
>>>
>>> ...
>>> I mean in current status, some of the Autonomic Functions should be
>> exposed to the network operator, and some of the Autonomic Functions c=
an be
>> hidden to the network operator
>>> ...
>>> 
>>> How is this different than running a script, or invoking an API? Ther=
e
>> are plenty of southbound APIs (from a management function to a device)=
=2E
>> There are very few northbound APIs. That's what the operators that I h=
ave
>> talked to want.
>>>
>>> regards,
>>> John
>>>
>>> On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng > > wrote:
>>> Hi John
>>>
>>> Some personal understanding here. If any misunderstanding, please
>> correct me. Thanks.
>>>
>>> 
>>> I would suggest a slight modification, to enable users of
>>> intent (and even intent writers) to NOT have to have
>>> detailed implementation information.
>>>
>>> What if all of the Autonomic Functions, after they joined
>>> the network, advertised both their capabilities as well as
>>> their needs? We can put aside the latter (needs) as this
>>> is more complex, but if the capabilities are advertised,
>>> then the autonomic management system can assemble
>>> those needs into a catalog. Then, the intent writer can
>>> look at the catalog and write a declarative request to
>>> use the functionality of the catalog.
>>>
>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>> behind...).
>>> < /jcs2>
>>>
>>> I agree that autonomic node should have that self-advertisement abili=
ty,
>> and autonomic management system should have this catalog.
>>>
>>> But I do not support that the intent writer should look every detail =
in
>> the catalog.
>>>
>>> The intent writer may mainly care about what they want, but how to do=
 it
>> can be left to the autonomic network to deal with.
>>> Perhaps several solutions exist, and autonomic management system choo=
se
>> one according to the abilities collected.
>>>
>>> In the ideal, the network operator may communicate with the autonomic=

>> network using network-level level intent and network-level report.
>>> I mean in current status, some of the Autonomic Functions should be
>> exposed to the network operator, and some of the Autonomic Functions c=
an be
>> hidden to the network operator
>>> Perhaps in future, when the network nodes become more intelligent, th=
e
>> hidden part will increase.
>>>
>>>
>>> Best regards
>>> Zongpeng Du
>>>
>>> From: Anima [mailto:anima-bounces@ietf.org]
>> On Behalf Of John Strassner
>>> Sent: Thursday, March 24, 2016 10:41 AM
>>> To: Michael Behringer (mbehring); John Strassner
>>> Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org> anima@ietf.org>; J=C3=A9ferson Campos Nobre
>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>>
>>> Hi Michael,
>>>
>>> thanks for your reply. I do think that we largely agree, and would en=
joy
>> chatting during or (hopefully before) BA. Please see ... =
Also
>> trying to unclutter, so I am eliding points that we agree on.
>>>
>>>
>>> While *CIEs don=E2=80=99t like magic, they do understand the concept =
of a
>> default behaviour which they can influence if needed. Any reasonable
>> network management approach tries to define =E2=80=9Cdefault=E2=80=9D =
behaviour and
>> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=E2=
=80=9D part very nicely, and *CIEs
>> get that.
>>> 
>>> Agreed!
>>> 
>>>
>>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discu=
ssion: My view is that we shouldn=E2=80=99t
>> try to define exactly *what* intent allows and doesn=E2=80=99t allow, =
since we=E2=80=99ve
>> heard from many sides that the boundary is not clear, and it may be
>> impossible to have a really clear, usable definition. So my point was:=

>> Let=E2=80=99s focus on the distribution model: Intent is flooded to al=
l nodes.
>> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific no=
des, etc,
>> because that would blur the boundary even worse.
>>> 
>>> I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v=
1
>> was 2006, almost 10 years ago), we used a message bus and pub-sub. In =
v2,
>> we flooded; in v3, we went back to pub-sub because we wanted to do som=
e
>> optimization and auditing. But I would be happy to try some type of
>> controlled flooding again.
>>> 
>>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>> owners define the content of what they want to push in their bit of th=
e
>> Intent. If we insist of full flooding, we encourage the right behaviou=
r.
>> And if someone really puts into his Intent =E2=80=9Cnode A: do this; n=
ode B: do the
>> other; node C: do the third=E2=80=9D it is awkward enough for people t=
o take a step
>> back and think whether they=E2=80=99re doing the right thing =E2=98=BA=

>>> 
>>> I agree in principle with what you said. Plus, see above.
>>> < /jcs2>
>>> 
>>> Do you mean that the intent writer will **know the difference**
>>> between the fabric and each autonomic function? I hope not,
>>> but perhaps I'm misunderstanding you...
>>> < /jcs>
>>> Not sure I understand your concern. Yes, in my model there is a secti=
on
>> for the fabric, and sections for the various functions. Seems a natura=
l
>> structure?!? 
>>> The intent **writer** would have to know implementation
>>> details in order to place everything in a single file. I would
>>> rather not require this, and instead, have the intent engine
>>> (e.g., compiler and associated logic) do this.
>>> < /jcs2>
>>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to stru=
cture Intent around
>> Autonomic Functions. This is not optimal from a scientific point of vi=
ew,
>> but aligns very well with the way how things are organised. Probably w=
e
>> need to discuss that in BA. But refer to my YANG model sent a few days=
 ago.
>>> 
>>> I would suggest a slight modification, to enable users of
>>> intent (and even intent writers) to NOT have to have
>>> detailed implementation information.
>>>
>>> What if all of the Autonomic Functions, after they joined
>>> the network, advertised both their capabilities as well as
>>> their needs? We can put aside the latter (needs) as this
>>> is more complex, but if the capabilities are advertised,
>>> then the autonomic management system can assemble
>>> those needs into a catalog. Then, the intent writer can
>>> look at the catalog and write a declarative request to
>>> use the functionality of the catalog.
>>>
>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>> behind...).
>>> < /jcs2>
>>>
>>>
>>> regards,
>>> John
>>>
>>> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <
>> mbehring@cisco.com> wrote:
>>> John,
>>>
>>> I think we=E2=80=99re on the same page. Couple of points, top posting=
 here,
>> otherwise it gets too messy.
>>>
>>> While *CIEs don=E2=80=99t like magic, they do understand the concept =
of a
>> default behaviour which they can influence if needed. Any reasonable
>> network management approach tries to define =E2=80=9Cdefault=E2=80=9D =
behaviour and
>> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=E2=
=80=9D part very nicely, and *CIEs
>> get that.
>>>
>>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discu=
ssion: My view is that we shouldn=E2=80=99t
>> try to define exactly *what* intent allows and doesn=E2=80=99t allow, =
since we=E2=80=99ve
>> heard from many sides that the boundary is not clear, and it may be
>> impossible to have a really clear, usable definition. So my point was:=

>> Let=E2=80=99s focus on the distribution model: Intent is flooded to al=
l nodes.
>> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific no=
des, etc,
>> because that would blur the boundary even worse.
>>>
>>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>> owners define the content of what they want to push in their bit of th=
e
>> Intent. If we insist of full flooding, we encourage the right behaviou=
r.
>> And if someone really puts into his Intent =E2=80=9Cnode A: do this; n=
ode B: do the
>> other; node C: do the third=E2=80=9D it is awkward enough for people t=
o take a step
>> back and think whether they=E2=80=99re doing the right thing =E2=98=BA=

>>>
>>> 
>>> Do you mean that the intent writer will **know the difference**
>>> between the fabric and each autonomic function? I hope not,
>>> but perhaps I'm misunderstanding you...
>>> 
>>>
>>> Not sure I understand your concern. Yes, in my model there is a secti=
on
>> for the fabric, and sections for the various functions. Seems a natura=
l
>> structure?!?
>>>
>>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to stru=
cture Intent around
>> Autonomic Functions. This is not optimal from a scientific point of vi=
ew,
>> but aligns very well with the way how things are organised. Probably w=
e
>> need to discuss that in BA. But refer to my YANG model sent a few days=
 ago.
>>>
>>> We should schedule some discussion time outside the ANIMA meeting
>> window...
>>>
>>> Michael
>>>
>>>
>>> From: John Strassner [mailto:strazpdj@gmail.com> strazpdj@gmail.com>]
>>> Sent: 22 March 2016 21:32
>>> To: Michael Behringer (mbehring) > mbehring@cisco.com>>; John Strassner > strazpdj@gmail.com>>
>>> Cc: Duzongpeng >;=

>> anima@ietf.org; Laurent Ciavaglia <
>> laurent.ciavaglia@nokia.com>;
>> J=C3=A9ferson Campos Nobre >;
>> Sheng Jiang >
>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>>
>>> In general, I agree with Michael. I'd like to emphasize and expand on=
 a
>> couple of key points:
>>>
>>>
>>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D=
 or =E2=80=9Csteering=E2=80=9D a network,
>> at least not initially. There is still traditional config, plus all th=
e
>> existing and emerging SDN models. My view has always been that ANIMA s=
ets a
>> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are=
 more specific and
>> can override intent. This is described in RFC7575. Main point here: AN=
IMA
>> Intent isn=E2=80=99t the only tool in a network.
>>>
>>> 
>>> Absolutely agree. Note that even in the Congress case I mentioned
>>> earlier, it is NOT pushing config, and it is NOT viewed as the
>>> "only" policy
>>> 
>>>
>>>
>>> Intent could perfectly well cover a high level policy such as =E2=80=9C=
all nodes
>> of type x do this; type y does that=E2=80=9D. (Which I think is what y=
ou mention
>> below.) When we started with Intent, the vision was to keep it very hi=
gh
>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=98=
=BA
>>>
>>> 
>>> Agreed. Although to me, this means that we need to consider the types=
 of
>> actors that are using intent. For example, most of my CCIE and JCIE fr=
iends
>> have no interest in using an intent language for pushing configuration=
=2E On
>> the other hand, when I talk to operators and application developers, t=
hey
>> would love to be able to use a high-level intent and have the intent e=
ngine
>> take care of the details for them.
>>> 
>>>
>>> It has become clear that in today=E2=80=99s networking, people will w=
ant to push
>> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I=
 think we=E2=80=99ll just have to acknowledge
>> that.
>>>
>>> 
>>> Sorry Michael, I don't follow here. What do you mean
>>> by "parameters"?
>>> 
>>>
>>>
>>> Personally I think it=E2=80=99ll pan out like this: Intent will have =
sections
>> per autonomic function. An Intent parser will validate overall correct=
ness
>> (signature, etc), then segment the Intent into its bits: One part will=
 be
>> the autonomic fabric itself, then there will be bits for each autonomi=
c
>> function.
>>>
>>> 
>>> Do you mean that the intent writer will **know the difference**
>>> between the fabric and each autonomic function? I hope not,
>>> but perhaps I'm misunderstanding you...
>>> 
>>>
>>>
>>> So the Intent parser on a node will grab intent, see there is some
>> Intent for autonomic function A, will see whether this is running loca=
lly,
>> and if so, it=E2=80=99ll just pass that piece of Intent to that functi=
on.
>>>
>>> 
>>> Agreed.
>>> 
>>>
>>>
>>> In this way, a function owner defines both the intent itself, and how=
 it
>> is consumed by his/her function.
>>>
>>> 
>>> I'm not sure what you mean by "function owner". (Sorry to be
>>> picky, I'm likely under-caffeinated.)
>>>
>>> The above seemed like an autonomous decision, made by the parser.
>>> I don't think that the intent writer should be cognizant of how many =
or
>>> what types of autonomic functions are present.
>>> 
>>>
>>>
>>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people=
 can do what they
>> want, in the worst case even push config, which is clearly NOT what In=
tent
>> is meant to do.
>>>
>>> 
>>> And hence, the "HAL 2000" scenario. :-)
>>> 
>>>
>>>
>>> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =
=E2=80=9Cnode 17: here is
>> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CL=
I=E2=80=9D. But what
>> I describe above COULD be abused in this way. RFC7575 says this is NOT=

>> Intent.
>>>
>>> 
>>> Absolutely agree!
>>> 
>>>
>>>  regards,
>>> John
>>>
>>>
>>>
>>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
>> mbehring@cisco.com> wrote:
>>> Hi Zongpeng,
>>>
>>> Yes, this is a recurring question and topic. Let me give my view on t=
wo
>> levels:
>>>
>>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D=
 or =E2=80=9Csteering=E2=80=9D a network,
>> at least not initially. There is still traditional config, plus all th=
e
>> existing and emerging SDN models. My view has always been that ANIMA s=
ets a
>> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are=
 more specific and
>> can override intent. This is described in RFC7575. Main point here: AN=
IMA
>> Intent isn=E2=80=99t the only tool in a network.
>>>
>>> Intent could perfectly well cover a high level policy such as =E2=80=9C=
all nodes
>> of type x do this; type y does that=E2=80=9D. (Which I think is what y=
ou mention
>> below.) When we started with Intent, the vision was to keep it very hi=
gh
>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=98=
=BA  It has become clear that
>> in today=E2=80=99s networking, people will want to push more =E2=80=9C=
parameters=E2=80=9D than
>> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to ackno=
wledge that.
>>>
>>> Personally I think it=E2=80=99ll pan out like this: Intent will have =
sections
>> per autonomic function. An Intent parser will validate overall correct=
ness
>> (signature, etc), then segment the Intent into its bits: One part will=
 be
>> the autonomic fabric itself, then there will be bits for each autonomi=
c
>> function.
>>>
>>> So the Intent parser on a node will grab intent, see there is some
>> Intent for autonomic function A, will see whether this is running loca=
lly,
>> and if so, it=E2=80=99ll just pass that piece of Intent to that functi=
on.
>>>
>>> In this way, a function owner defines both the intent itself, and how=
 it
>> is consumed by his/her function.
>>>
>>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people=
 can do what they
>> want, in the worst case even push config, which is clearly NOT what In=
tent
>> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writi=
ng Intent like =E2=80=9Cnode
>> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: her=
e is your
>> CLI=E2=80=9D. But what I describe above COULD be abused in this way. R=
FC7575 says
>> this is NOT Intent.
>>>
>>> In other words: Let=E2=80=99s not be too prescriptive, and hope that =
folks will
>> do more or less reasonable things.
>>>
>>> I=E2=80=99ve been meaning to comment on the Intent drafts and possibl=
y
>> contribute, I=E2=80=99m just struggling with priorities at the moment =
(as you can
>> see by my relative silence over the past weeks).
>>>
>>> Michael
>>>
>>>
>>>
>>> From: Anima [mailto:anima-bounces@ietf.org]
>> On Behalf Of Duzongpeng
>>> Sent: 16 March 2016 02:38
>>> To: anima@ietf.org
>>> Cc: Laurent Ciavaglia > laurent.ciavaglia@nokia.com>>; J=C3=A9ferson Campos Nobre > >; Sheng Jiang > jiangsheng@huawei.com>>
>>> Subject: [Anima] Does Autonomic network need some intervention beyond=

>> intent
>>>
>>> Hello, everyone
>>>
>>>          I understand that Autonomic network should be able to work w=
ith
>> as less configuration as possible. In the ideal case, an Autonomic net=
work
>> should be able to work well while only intent is needed from human
>> operators.
>>>
>>>          However, I wonder whether Autonomic network may need some
>> intervention from human operators through some low-level policies beyo=
nd
>> intents.
>>>
>>>          For example, in the
>> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it =
is
>> suggested that the prefix lengths for the CSG, ASG, RSG (different rol=
es in
>> IP RAN) can be assigned as an "intent". These configuration parameters=
 need
>> to be distributed in the autonomic domain to influence the detail
>> configurations on each autonomic node.
>>>
>>>          As intent is described as an abstract, declarative, high-lev=
el
>> policy used to operate an autonomic domain, such as an enterprise netw=
ork.
>> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cpre=
fix-management ID=E2=80=9D to some
>> other words.
>>>
>>> Is =E2=80=9CASA parameters" ok here?
>>>
>>> I think some characteristics for this kind of low-level policies are
>>>
>>> 1.       Network level parameters, need to be distributed by GRASP an=
d
>> ACP
>>>
>>> 2.       Relatively static compared to intent, perhaps only needed to=
 be
>> configured once
>>>
>>> 3.       Unavoidable in current network environment, mostly for
>> establishing the network infrastructure
>>>
>>> 4.       Rarely need coordinations with others parameters, usually is=

>> closed related to one specific autonomic function
>>>
>>> Welcome for comment. Your suggestions will be appreciated.
>>>
>>>
>>> Best Regards
>>> Zongpeng Du
>>>
>>> _______________________________________________
>>> Anima mailing list
>>> Anima@ietf.org
>>> https://www.ietf.org/mailman/listinfo/anima
>>>
>>>
>>>
>>> --
>>> regards,
>>> John
>>>
>>>
>>>
>>> --
>>> regards,
>>> John
>>>
>>>
>>>
>>> --
>>> regards,
>>> John
>>>
>>>
>>>
>>> _______________________________________________
>>> Anima mailing list
>>> Anima@ietf.org
>>> https://www.ietf.org/mailman/listinfo/anima
>>>
>>
>>
>=20
>=20


From nobody Sat Mar 26 13:21:24 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4637712D10C for ; Sat, 26 Mar 2016 13:21:23 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.722
X-Spam-Level: 
X-Spam-Status: No, score=-2.722 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id vmme75HmWDkA for ; Sat, 26 Mar 2016 13:21:20 -0700 (PDT)
Received: from mailb2.tigertech.net (mailb2.tigertech.net [208.80.4.154]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 543FE12D0A1 for ; Sat, 26 Mar 2016 13:21:20 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by mailb2.tigertech.net (Postfix) with ESMTP id F07661C08CE; Sat, 26 Mar 2016 13:21:19 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1459023679; bh=ANE7xk2pO4sJZv5HmGMqjECSq3m6yNpMh/15KB5D4Y8=; h=Subject:To:References:Cc:From:Date:In-Reply-To:From; b=alF3Tya1NxVPwh8ww+Y0S1kKvNreA5F2pK9TuBwEONzNGzXI0al9FFfmwmUJ7ELH0 LDouN7QtbhhvlReJEyvv7Xig7e9XEvgO1NiMoc+qyrI7v86+SR5N5OZgQEpQXGW6aj 3f9shc0zVfHS5RtxyT648Y1X6sVDWE51/KqKYad0=
X-Virus-Scanned: Debian amavisd-new at b2.tigertech.net
Received: from Joels-MacBook-Pro.local (209-255-163-147.ip.mcleodusa.net [209.255.163.147]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mailb2.tigertech.net (Postfix) with ESMTPSA id AF0BA1C04EB; Sat, 26 Mar 2016 13:21:18 -0700 (PDT)
To: "Alexander Clemm (alex)" , Duzongpeng , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>   <1dd14a9abe20419592e877235979585e@XCH-RTP-001.cisco.com>
From: "Joel M. Halpern" 
Message-ID: <56F6EF3C.7040403@joelhalpern.com>
Date: Sat, 26 Mar 2016 16:21:16 -0400
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <1dd14a9abe20419592e877235979585e@XCH-RTP-001.cisco.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 8bit
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "Michael Behringer \(mbehring\)" , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 20:21:23 -0000

If we want to flood the data, I would appreciate knowing whether RFC 
2334 is potentially useful.

Yours,
Joel

On 3/25/16 2:18 PM, Alexander Clemm (alex) wrote:
> Hi Zongpeng,
>
> looking at your comments, I completely agree.  I think we’re on the same
> page.
>
> Regards
>
> --- Alex
>
> *From:*Duzongpeng [mailto:duzongpeng@huawei.com]
> *Sent:* Friday, March 25, 2016 4:43 AM
> *To:* John Strassner ; Alexander Clemm (alex)
> 
> *Cc:* Michael Behringer (mbehring) ; Laurent
> Ciavaglia ; Sheng Jiang
> ; anima@ietf.org; Jéferson Campos Nobre
> 
> *Subject:* RE: [Anima] Does Autonomic network need some intervention
> beyond intent
>
> Hi, John, Alex:
>
>           Some personal understanding below. Please see inline.
>
> Best regards
>
> Zongpeng Du
>
> *From:*John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* Thursday, March 24, 2016 11:41 AM
> *To:* Alexander Clemm (alex); John Strassner
> *Cc:* Michael Behringer (mbehring); Duzongpeng; Laurent Ciavaglia; Sheng
> Jiang; anima@ietf.org ; Jéferson Campos Nobre
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
> Hi Alex,
>
> I'm not sure I follow a couple of your points, please see
> inline...though in retrospect, perhaps this is merely a case of who the
> actor is (in my case, I am worried about the writer and user of intent,
> not how we build it).
>
> I really like your suggestion of flooding intent to all nodes (even if
> there may be optimizations later).  It provides a level of conceptual
> clarity.  Every node should be aware of overall intent (and need to
> decide for itself if there are actions for it to take).
>
> 
>
> Why should **every** node be aware of all intent? Many times, this won't
> be applicable.
>
> The only reason I can see for doing this is to enable collaboration
> and/or emergent behavior (in the AI sense). However, if you want
> emergent behavior from autonomic nodes, then each node that is part of
> that collaboration must interact. I don't think we know enough about
> this to support or prohibit such behavior, other than this places an
> extra requirement on the computational and reasoning abilities of the node.
>
> If you have a different purpose, please do explain.
>
> 
>
> [duzongpeng]
>
> I agree that we can have every node in the autonomic domain have the
> same intent. Within my understanding, it can be a candidate solution.
>
> The advantage of the solution is simple. By using some synchronization
> mechanism, it makes all the node in the autonomic domain can work
> cooperatively.
>
> Another consideration is that when a new node join in the autonomic
> domain, perhaps it can receive the intent from its neighbor.
>
> I means Administrative Intent here, not a service intent. (Section 8 of
> https://tools.ietf.org/html/draft-du-anima-an-intent-03)
>
> Perhaps the later only needs to be distributed to the related node, not
> all the node.
>
> The service intent may even be related to multiple autonomic domains.
>
> [/duzongpeng]
>
> If there is some intent that applies only to some parts of the network
> and not others, that can be addressed by using different Autonomic
> Domains.  And, of course, Autonomic Domains may be associated with
> certain Autonomic Functions.  Intervention beyond intent should NOT be
> allowed for the Autonomic Function(s) being served an Autonomic Domain,
> but of course other interventions (for different functions, hence
> outside the scope) are still possible.
>
> 
>
> I think that Autonomic Domains should (must?) be hidden from the
> user/intent writer. They are clearly of huge internal benefit, but why
> should the user of intent know about them? In addition, I think that we
> should use roles as an indirection mechanism for scalability when
> working with Autonomic Domains.
>
> 
>
> [duzongpeng]
>
> Within my understanding, an autonomic domain will have an intent, and it
> can only take effect within this domain.
>
> When a user issues an intent, she/he should know the domain name. I mean
> an intent can contain the domain name as one of its element.
>
> I agree that using roles is good for scalability.
>
> [/duzongpeng]
>
> regards,
>
> John
>
> On Wed, Mar 23, 2016 at 10:56 AM, Alexander Clemm (alex)  > wrote:
>
> I really like your suggestion of flooding intent to all nodes (even if
> there may be optimizations later).  It provides a level of conceptual
> clarity.  Every node should be aware of overall intent (and need to
> decide for itself if there are actions for it to take).
>
> If there is some intent that applies only to some parts of the network
> and not others, that can be addressed by using different Autonomic
> Domains.  And, of course, Autonomic Domains may be associated with
> certain Autonomic Functions.  Intervention beyond intent should NOT be
> allowed for the Autonomic Function(s) being served an Autonomic Domain,
> but of course other interventions (for different functions, hence
> outside the scope) are still possible.
>
> --- Alex
>
> *From:*Anima [mailto:anima-bounces@ietf.org
> ] *On Behalf Of *Michael Behringer (mbehring)
> *Sent:* Wednesday, March 23, 2016 3:57 AM
> *To:* John Strassner >
> *Cc:* Duzongpeng >;
> Laurent Ciavaglia  >; Sheng Jiang
> >; anima@ietf.org
> ; Jéferson Campos Nobre  >
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
> John,
>
> I think we’re on the same page. Couple of points, top posting here,
> otherwise it gets too messy.
>
> While *CIEs don’t like magic, they do understand the concept of a
> default behaviour which they can influence if needed. Any reasonable
> network management approach tries to define “default” behaviour and
> “exceptions”. The AN story fills the “default” part very nicely, and
> *CIEs get that.
>
> The “parameter” versus “policy” discussion: My view is that we shouldn’t
> try to define exactly **what** intent allows and doesn’t allow, since
> we’ve heard from many sides that the boundary is not clear, and it may
> be impossible to have a really clear, usable definition. So my point
> was: Let’s focus on the distribution model: Intent is flooded to all
> nodes. Let’s NOT do selective flooding, unicast Intent to specific
> nodes, etc, because that would blur the boundary even worse.
>
> My suggestion: Intent is flooded to all nodes in a domain. The ASA
> owners define the content of what they want to push in their bit of the
> Intent. If we insist of full flooding, we encourage the right behaviour.
> And if someone really puts into his Intent “node A: do this; node B: do
> the other; node C: do the third” it is awkward enough for people to take
> a step back and think whether they’re doing the right thing J
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
> Not sure I understand your concern. Yes, in my model there is a section
> for the fabric, and sections for the various functions. Seems a natural
> structure?!?
>
> “Function owner”: Yes, I’m suggesting to structure Intent around
> Autonomic Functions. This is not optimal from a scientific point of
> view, but aligns very well with the way how things are organised.
> Probably we need to discuss that in BA. But refer to my YANG model sent
> a few days ago.
>
> We should schedule some discussion time outside the ANIMA meeting window...
>
> Michael
>
> *From:*John Strassner [mailto:strazpdj@gmail.com]
> *Sent:* 22 March 2016 21:32
> *To:* Michael Behringer (mbehring)  >; John Strassner  >
> *Cc:* Duzongpeng >;
> anima@ietf.org ; Laurent Ciavaglia
> >;
> Jéferson Campos Nobre  >; Sheng Jiang  >
> *Subject:* Re: [Anima] Does Autonomic network need some intervention
> beyond intent
>
> In general, I agree with Michael. I'd like to emphasize and expand on a
> couple of key points:
>
> First, ANIMA isn’t the only thing “influencing” or “steering” a network,
> at least not initially. There is still traditional config, plus all the
> existing and emerging SDN models. My view has always been that ANIMA
> sets a “default” policy in a network, and all other means are more
> specific and can override intent. This is described in RFC7575. Main
> point here: ANIMA Intent isn’t the only tool in a network.
>
> 
>
> Absolutely agree. Note that even in the Congress case I mentioned
>
> earlier, it is NOT pushing config, and it is NOT viewed as the
>
> "only" policy
>
> 
>
> Intent could perfectly well cover a high level policy such as “all nodes
> of type x do this; type y does that”. (Which I think is what you mention
> below.) When we started with Intent, the vision was to keep it very high
> level, more along the line “do the right thing” J
>
> 
>
> Agreed. Although to me, this means that we need to consider the types of
> actors that are using intent. For example, most of my CCIE and JCIE
> friends have no interest in using an intent language for pushing
> configuration. On the other hand, when I talk to operators and
> application developers, they would love to be able to use a high-level
> intent and have the intent engine take care of the details for them.
>
> 
>
> It has become clear that in today’s networking, people will want to push
> more “parameters”than “policy”, and I think we’ll just have to
> acknowledge that.
>
> 
>
> Sorry Michael, I don't follow here. What do you mean
>
> by "parameters"?
>
> 
>
> Personally I think it’ll pan out like this: Intent will have sections
> per autonomic function. An Intent parser will validate overall
> correctness (signature, etc), then segment the Intent into its bits: One
> part will be the autonomic fabric itself, then there will be bits for
> each autonomic function.
>
> 
>
> Do you mean that the intent writer will **know the difference**
>
> between the fabric and each autonomic function? I hope not,
>
> but perhaps I'm misunderstanding you...
>
> 
>
> So the Intent parser on a node will grab intent, see there is some
> Intent for autonomic function A, will see whether this is running
> locally, and if so, it’ll just pass that piece of Intent to that function.
>
> 
>
> Agreed.
>
> 
>
> In this way, a function owner defines both the intent itself, and how it
> is consumed by his/her function.
>
> 
>
> I'm not sure what you mean by "function owner". (Sorry to be
>
> picky, I'm likely under-caffeinated.)
>
> The above seemed like an autonomous decision, made by the parser.
>
> I don't think that the intent writer should be cognizant of how many or
>
> what types of autonomic functions are present.
>
> 
>
> The downside is that there is no “control” and people can do what they
> want, in the worst case even push config, which is clearly NOT what
> Intent is meant to do.
>
> 
>
> And hence, the "HAL 2000" scenario. :-)
>
> 
>
> I’d hope that people don’t start writing Intent like “node 17: here is
> your CLI; node 23: here is your CLI”; node 37: here is your CLI”. But
> what I describe above COULD be abused in this way. RFC7575 says this is
> NOT Intent.
>
> 
>
> Absolutely agree!
>
> 
>
>   regards,
>
> John
>
> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring)
> > wrote:
>
>     Hi Zongpeng,
>
>     Yes, this is a recurring question and topic. Let me give my view on
>     two levels:
>
>     First, ANIMA isn’t the only thing “influencing” or “steering” a
>     network, at least not initially. There is still traditional config,
>     plus all the existing and emerging SDN models. My view has always
>     been that ANIMA sets a “default” policy in a network, and all other
>     means are more specific and can override intent. This is described
>     in RFC7575. Main point here: ANIMA Intent isn’t the only tool in a
>     network.
>
>     Intent could perfectly well cover a high level policy such as “all
>     nodes of type x do this; type y does that”. (Which I think is what
>     you mention below.) When we started with Intent, the vision was to
>     keep it very high level, more along the line “do the right thing” J
>     It has become clear that in today’s networking, people will want to
>     push more “parameters”than “policy”, and I think we’ll just have to
>     acknowledge that.
>
>     Personally I think it’ll pan out like this: Intent will have
>     sections per autonomic function. An Intent parser will validate
>     overall correctness (signature, etc), then segment the Intent into
>     its bits: One part will be the autonomic fabric itself, then there
>     will be bits for each autonomic function.
>
>     So the Intent parser on a node will grab intent, see there is some
>     Intent for autonomic function A, will see whether this is running
>     locally, and if so, it’ll just pass that piece of Intent to that
>     function.
>
>     In this way, a function owner defines both the intent itself, and
>     how it is consumed by his/her function.
>
>     The downside is that there is no “control” and people can do what
>     they want, in the worst case even push config, which is clearly NOT
>     what Intent is meant to do. I’d hope that people don’t start writing
>     Intent like “node 17: here is your CLI; node 23: here is your CLI”;
>     node 37: here is your CLI”. But what I describe above COULD be
>     abused in this way. RFC7575 says this is NOT Intent.
>
>     In other words: Let’s not be too prescriptive, and hope that folks
>     will do more or less reasonable things.
>
>     I’ve been meaning to comment on the Intent drafts and possibly
>     contribute, I’m just struggling with priorities at the moment (as
>     you can see by my relative silence over the past weeks).
>
>     Michael
>
>     *From:*Anima [mailto:anima-bounces@ietf.org
>     ] *On Behalf Of *Duzongpeng
>     *Sent:* 16 March 2016 02:38
>     *To:* anima@ietf.org 
>     *Cc:* Laurent Ciavaglia      >; Jéferson Campos Nobre
>     >; Sheng Jiang
>     >
>     *Subject:* [Anima] Does Autonomic network need some intervention
>     beyond intent
>
>     Hello, everyone
>
>               I understand that Autonomic network should be able to work
>     with as less configuration as possible. In the ideal case, an
>     Autonomic network should be able to work well while only intent is
>     needed from human operators.
>
>               However, I wonder whether Autonomic network may need some
>     intervention from human operators through some low-level policies
>     beyond intents.
>
>               For example, in the
>     https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,
>     it is suggested that the prefix lengths for the CSG, ASG, RSG
>     (different roles in IP RAN) can be assigned as an "intent". These
>     configuration parameters need to be distributed in the autonomic
>     domain to influence the detail configurations on each autonomic node.
>
>               As intent is described as an abstract, declarative,
>     high-level policy used to operate an autonomic domain, such as an
>     enterprise network. Perhaps we can rename the “intent” in the
>     “prefix-management ID” to some other words.
>
>     Is “ASA parameters" ok here?
>
>     I think some characteristics for this kind of low-level policies are
>
>     1.Network level parameters, need to be distributed by GRASP and ACP
>
>     2.Relatively static compared to intent, perhaps only needed to be
>     configured once
>
>     3.Unavoidable in current network environment, mostly for
>     establishing the network infrastructure
>
>     4.Rarely need coordinations with others parameters, usually is
>     closed related to one specific autonomic function
>
>     Welcome for comment. Your suggestions will be appreciated.
>
>     Best Regards
>
>     Zongpeng Du
>
>
>     _______________________________________________
>     Anima mailing list
>     Anima@ietf.org 
>     https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
> --
>
> regards,
>
> John
>
>
>
>
> --
>
> regards,
>
> John
>
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>


From nobody Sat Mar 26 13:42:23 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 06A9A12D5C2 for ; Sat, 26 Mar 2016 13:42:22 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9XR5SkWS57GD for ; Sat, 26 Mar 2016 13:42:18 -0700 (PDT)
Received: from mail-pa0-x22c.google.com (mail-pa0-x22c.google.com [IPv6:2607:f8b0:400e:c03::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 926F212D194 for ; Sat, 26 Mar 2016 13:42:18 -0700 (PDT)
Received: by mail-pa0-x22c.google.com with SMTP id fe3so69101321pab.1 for ; Sat, 26 Mar 2016 13:42:18 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=i1KxF+Hc7MYnKjLvOcQ1OXgo8l5+muAtxgYL8KhycM8=; b=lcApI4HeeiUWeyHRHYVQ6IfX6Sqzfosw23/KUHmFLKVu5ShTz7qPj3STf8PFnuycMi PYkpFGeXO8QdX86zlhX4W+XDnsRsBolncpFLtiiHrOCdOMovk4dym2r9dxKUi5wtrPw4 1xU/0gnMxRk7o/qwwEQH4b1b5dJn/2i97YXNH0Dv+bHkMwkYUESPYJY3mGoEbFVVCalI E/9+fF5o9o4HajcFqtpQEOgvpcYzHLmobr9Pa0oHlYlZDWtxTShFiZIA//I0Ud10k4+C zA1s6Pla3uaf0Atx81idHG/Xbj0v437bOp6fTnxYRbmfYxs4/345XkKNtX5wQUmhJe3U nZmw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=i1KxF+Hc7MYnKjLvOcQ1OXgo8l5+muAtxgYL8KhycM8=; b=XVIZ9LOy2/eFFHcCqMG/zw7uafpk7EE4EaT5q+GGCbRfnT4+tzPnRdFUTJkU8wZt9D ebuxoyQy3q/gszKIhieZ6hYB1mMZxyLssdH5LjJSo2Ze3c5ZByR63uLPh+oOAx7IHl51 r43VVeqqikhX5tuL2N5Z+ZDrbQi+SbB81kCnd7ezjQ4o/N44qZFbmz3nUBEzDK9WSLI8 N0FrT4ooPtr0LNCNftzMYdN0FkirJ0ui14X21phkoruR++VnVZCN5vQoWTKSe7pQeSv1 p+b9cB2K4Lpp8OHJpZQg58C12kIXI3E5tPkNTcjHL2v7UobNtobqADJLsIVs5IBsFZyK VxKQ==
X-Gm-Message-State: AD7BkJIao8oq32EZLo2rGxeha7jCiLEAlV0f+swWspdqLy98DLPTp/uhtKxko0R82tdMFg==
X-Received: by 10.66.119.202 with SMTP id kw10mr30760394pab.110.1459024938112;  Sat, 26 Mar 2016 13:42:18 -0700 (PDT)
Received: from ?IPv6:2406:e007:738b:1:28cc:dc4c:9703:6781? ([2406:e007:738b:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id wb7sm25033411pab.3.2016.03.26.13.42.12 (version=TLSv1/SSLv3 cipher=OTHER); Sat, 26 Mar 2016 13:42:16 -0700 (PDT)
To: "Joel M. Halpern" , "Alexander Clemm (alex)" , Duzongpeng , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com> <6aca826a56994d5898481b70e0cb5196@XCH-RTP-001.cisco.com>   <1dd14a9abe20419592e877235979585e@XCH-RTP-001.cisco.com> <56F6EF3C.7040403@joelhalpern.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F6F431.2080304@gmail.com>
Date: Sun, 27 Mar 2016 09:42:25 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56F6EF3C.7040403@joelhalpern.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , "Michael Behringer \(mbehring\)" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sat, 26 Mar 2016 20:42:22 -0000

Oh, I had completely forgotten that existed. At least, it should be menti=
oned
in the Appendix to GRASP. Were there implementations?

I have the impression that it's best suited to scenarios in which any nod=
e
is equally likely to originate an update; that isn't quite what we expect=
 for
an AN scenario.

The algorithm looks quite similar to what DNCP does (after all, the DNCP =
authors
originally proposed a set of OSPF extensions, iirc.) Also quite similar t=
o what
Michael B described. I'm sure we can learn from it.

Boy, that's a sad litany of dead companies in the author list: Bay Networ=
ks, Bellcore, Newbridge.

Regards
   Brian

On 27/03/2016 09:21, Joel M. Halpern wrote:
> If we want to flood the data, I would appreciate knowing whether RFC 23=
34 is potentially useful.
>=20
> Yours,
> Joel
>=20
> On 3/25/16 2:18 PM, Alexander Clemm (alex) wrote:
>> Hi Zongpeng,
>>
>> looking at your comments, I completely agree.  I think we=E2=80=99re o=
n the same
>> page.
>>
>> Regards
>>
>> --- Alex
>>
>> *From:*Duzongpeng [mailto:duzongpeng@huawei.com]
>> *Sent:* Friday, March 25, 2016 4:43 AM
>> *To:* John Strassner ; Alexander Clemm (alex)
>> 
>> *Cc:* Michael Behringer (mbehring) ; Laurent
>> Ciavaglia ; Sheng Jiang
>> ; anima@ietf.org; J=C3=A9ferson Campos Nobre
>> 
>> *Subject:* RE: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>
>> Hi, John, Alex:
>>
>>           Some personal understanding below. Please see inline.
>>
>> Best regards
>>
>> Zongpeng Du
>>
>> *From:*John Strassner [mailto:strazpdj@gmail.com]
>> *Sent:* Thursday, March 24, 2016 11:41 AM
>> *To:* Alexander Clemm (alex); John Strassner
>> *Cc:* Michael Behringer (mbehring); Duzongpeng; Laurent Ciavaglia; She=
ng
>> Jiang; anima@ietf.org ; J=C3=A9ferson Campos No=
bre
>> *Subject:* Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>
>> Hi Alex,
>>
>> I'm not sure I follow a couple of your points, please see
>> inline...though in retrospect, perhaps this is merely a case of who th=
e
>> actor is (in my case, I am worried about the writer and user of intent=
,
>> not how we build it).
>>
>> I really like your suggestion of flooding intent to all nodes (even if=

>> there may be optimizations later).  It provides a level of conceptual
>> clarity.  Every node should be aware of overall intent (and need to
>> decide for itself if there are actions for it to take).
>>
>> 
>>
>> Why should **every** node be aware of all intent? Many times, this won=
't
>> be applicable.
>>
>> The only reason I can see for doing this is to enable collaboration
>> and/or emergent behavior (in the AI sense). However, if you want
>> emergent behavior from autonomic nodes, then each node that is part of=

>> that collaboration must interact. I don't think we know enough about
>> this to support or prohibit such behavior, other than this places an
>> extra requirement on the computational and reasoning abilities of the =
node.
>>
>> If you have a different purpose, please do explain.
>>
>> 
>>
>> [duzongpeng]
>>
>> I agree that we can have every node in the autonomic domain have the
>> same intent. Within my understanding, it can be a candidate solution.
>>
>> The advantage of the solution is simple. By using some synchronization=

>> mechanism, it makes all the node in the autonomic domain can work
>> cooperatively.
>>
>> Another consideration is that when a new node join in the autonomic
>> domain, perhaps it can receive the intent from its neighbor.
>>
>> I means Administrative Intent here, not a service intent. (Section 8 o=
f
>> https://tools.ietf.org/html/draft-du-anima-an-intent-03)
>>
>> Perhaps the later only needs to be distributed to the related node, no=
t
>> all the node.
>>
>> The service intent may even be related to multiple autonomic domains.
>>
>> [/duzongpeng]
>>
>> If there is some intent that applies only to some parts of the network=

>> and not others, that can be addressed by using different Autonomic
>> Domains.  And, of course, Autonomic Domains may be associated with
>> certain Autonomic Functions.  Intervention beyond intent should NOT be=

>> allowed for the Autonomic Function(s) being served an Autonomic Domain=
,
>> but of course other interventions (for different functions, hence
>> outside the scope) are still possible.
>>
>> 
>>
>> I think that Autonomic Domains should (must?) be hidden from the
>> user/intent writer. They are clearly of huge internal benefit, but why=

>> should the user of intent know about them? In addition, I think that w=
e
>> should use roles as an indirection mechanism for scalability when
>> working with Autonomic Domains.
>>
>> 
>>
>> [duzongpeng]
>>
>> Within my understanding, an autonomic domain will have an intent, and =
it
>> can only take effect within this domain.
>>
>> When a user issues an intent, she/he should know the domain name. I me=
an
>> an intent can contain the domain name as one of its element.
>>
>> I agree that using roles is good for scalability.
>>
>> [/duzongpeng]
>>
>> regards,
>>
>> John
>>
>> On Wed, Mar 23, 2016 at 10:56 AM, Alexander Clemm (alex) > > wrote:
>>
>> I really like your suggestion of flooding intent to all nodes (even if=

>> there may be optimizations later).  It provides a level of conceptual
>> clarity.  Every node should be aware of overall intent (and need to
>> decide for itself if there are actions for it to take).
>>
>> If there is some intent that applies only to some parts of the network=

>> and not others, that can be addressed by using different Autonomic
>> Domains.  And, of course, Autonomic Domains may be associated with
>> certain Autonomic Functions.  Intervention beyond intent should NOT be=

>> allowed for the Autonomic Function(s) being served an Autonomic Domain=
,
>> but of course other interventions (for different functions, hence
>> outside the scope) are still possible.
>>
>> --- Alex
>>
>> *From:*Anima [mailto:anima-bounces@ietf.org
>> ] *On Behalf Of *Michael Behringer (mbe=
hring)
>> *Sent:* Wednesday, March 23, 2016 3:57 AM
>> *To:* John Strassner >
>> *Cc:* Duzongpeng =
>;
>> Laurent Ciavaglia > >; Sheng Jiang
>> >; anima@ietf.org=

>> ; J=C3=A9ferson Campos Nobre > >
>> *Subject:* Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>
>> John,
>>
>> I think we=E2=80=99re on the same page. Couple of points, top posting =
here,
>> otherwise it gets too messy.
>>
>> While *CIEs don=E2=80=99t like magic, they do understand the concept o=
f a
>> default behaviour which they can influence if needed. Any reasonable
>> network management approach tries to define =E2=80=9Cdefault=E2=80=9D =
behaviour and
>> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=E2=
=80=9D part very nicely, and
>> *CIEs get that.
>>
>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discus=
sion: My view is that we shouldn=E2=80=99t
>> try to define exactly **what** intent allows and doesn=E2=80=99t allow=
, since
>> we=E2=80=99ve heard from many sides that the boundary is not clear, an=
d it may
>> be impossible to have a really clear, usable definition. So my point
>> was: Let=E2=80=99s focus on the distribution model: Intent is flooded =
to all
>> nodes. Let=E2=80=99s NOT do selective flooding, unicast Intent to spec=
ific
>> nodes, etc, because that would blur the boundary even worse.
>>
>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>> owners define the content of what they want to push in their bit of th=
e
>> Intent. If we insist of full flooding, we encourage the right behaviou=
r.
>> And if someone really puts into his Intent =E2=80=9Cnode A: do this; n=
ode B: do
>> the other; node C: do the third=E2=80=9D it is awkward enough for peop=
le to take
>> a step back and think whether they=E2=80=99re doing the right thing J
>>
>> 
>>
>> Do you mean that the intent writer will **know the difference**
>>
>> between the fabric and each autonomic function? I hope not,
>>
>> but perhaps I'm misunderstanding you...
>>
>> 
>>
>> Not sure I understand your concern. Yes, in my model there is a sectio=
n
>> for the fabric, and sections for the various functions. Seems a natura=
l
>> structure?!?
>>
>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to struc=
ture Intent around
>> Autonomic Functions. This is not optimal from a scientific point of
>> view, but aligns very well with the way how things are organised.
>> Probably we need to discuss that in BA. But refer to my YANG model sen=
t
>> a few days ago.
>>
>> We should schedule some discussion time outside the ANIMA meeting wind=
ow...
>>
>> Michael
>>
>> *From:*John Strassner [mailto:strazpdj@gmail.com]
>> *Sent:* 22 March 2016 21:32
>> *To:* Michael Behringer (mbehring) > >; John Strassner > >
>> *Cc:* Duzongpeng =
>;
>> anima@ietf.org ; Laurent Ciavaglia
>> >;
>> J=C3=A9ferson Campos Nobre > >; Sheng Jiang > >
>> *Subject:* Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>
>> In general, I agree with Michael. I'd like to emphasize and expand on =
a
>> couple of key points:
>>
>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=9D=
 or =E2=80=9Csteering=E2=80=9D a network,
>> at least not initially. There is still traditional config, plus all th=
e
>> existing and emerging SDN models. My view has always been that ANIMA
>> sets a =E2=80=9Cdefault=E2=80=9D policy in a network, and all other me=
ans are more
>> specific and can override intent. This is described in RFC7575. Main
>> point here: ANIMA Intent isn=E2=80=99t the only tool in a network.
>>
>> 
>>
>> Absolutely agree. Note that even in the Congress case I mentioned
>>
>> earlier, it is NOT pushing config, and it is NOT viewed as the
>>
>> "only" policy
>>
>> 
>>
>> Intent could perfectly well cover a high level policy such as =E2=80=9C=
all nodes
>> of type x do this; type y does that=E2=80=9D. (Which I think is what y=
ou mention
>> below.) When we started with Intent, the vision was to keep it very hi=
gh
>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D J
>>
>> 
>>
>> Agreed. Although to me, this means that we need to consider the types =
of
>> actors that are using intent. For example, most of my CCIE and JCIE
>> friends have no interest in using an intent language for pushing
>> configuration. On the other hand, when I talk to operators and
>> application developers, they would love to be able to use a high-level=

>> intent and have the intent engine take care of the details for them.
>>
>> 
>>
>> It has become clear that in today=E2=80=99s networking, people will wa=
nt to push
>> more =E2=80=9Cparameters=E2=80=9Dthan =E2=80=9Cpolicy=E2=80=9D, and I =
think we=E2=80=99ll just have to
>> acknowledge that.
>>
>> 
>>
>> Sorry Michael, I don't follow here. What do you mean
>>
>> by "parameters"?
>>
>> 
>>
>> Personally I think it=E2=80=99ll pan out like this: Intent will have s=
ections
>> per autonomic function. An Intent parser will validate overall
>> correctness (signature, etc), then segment the Intent into its bits: O=
ne
>> part will be the autonomic fabric itself, then there will be bits for
>> each autonomic function.
>>
>> 
>>
>> Do you mean that the intent writer will **know the difference**
>>
>> between the fabric and each autonomic function? I hope not,
>>
>> but perhaps I'm misunderstanding you...
>>
>> 
>>
>> So the Intent parser on a node will grab intent, see there is some
>> Intent for autonomic function A, will see whether this is running
>> locally, and if so, it=E2=80=99ll just pass that piece of Intent to th=
at function.
>>
>> 
>>
>> Agreed.
>>
>> 
>>
>> In this way, a function owner defines both the intent itself, and how =
it
>> is consumed by his/her function.
>>
>> 
>>
>> I'm not sure what you mean by "function owner". (Sorry to be
>>
>> picky, I'm likely under-caffeinated.)
>>
>> The above seemed like an autonomous decision, made by the parser.
>>
>> I don't think that the intent writer should be cognizant of how many o=
r
>>
>> what types of autonomic functions are present.
>>
>> 
>>
>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people =
can do what they
>> want, in the worst case even push config, which is clearly NOT what
>> Intent is meant to do.
>>
>> 
>>
>> And hence, the "HAL 2000" scenario. :-)
>>
>> 
>>
>> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =E2=
=80=9Cnode 17: here is
>> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CL=
I=E2=80=9D. But
>> what I describe above COULD be abused in this way. RFC7575 says this i=
s
>> NOT Intent.
>>
>> 
>>
>> Absolutely agree!
>>
>> 
>>
>>   regards,
>>
>> John
>>
>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring)
>> > wrote:
>>
>>     Hi Zongpeng,
>>
>>     Yes, this is a recurring question and topic. Let me give my view o=
n
>>     two levels:
>>
>>     First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=
=9D or =E2=80=9Csteering=E2=80=9D a
>>     network, at least not initially. There is still traditional config=
,
>>     plus all the existing and emerging SDN models. My view has always
>>     been that ANIMA sets a =E2=80=9Cdefault=E2=80=9D policy in a netwo=
rk, and all other
>>     means are more specific and can override intent. This is described=

>>     in RFC7575. Main point here: ANIMA Intent isn=E2=80=99t the only t=
ool in a
>>     network.
>>
>>     Intent could perfectly well cover a high level policy such as =E2=80=
=9Call
>>     nodes of type x do this; type y does that=E2=80=9D. (Which I think=
 is what
>>     you mention below.) When we started with Intent, the vision was to=

>>     keep it very high level, more along the line =E2=80=9Cdo the right=
 thing=E2=80=9D J
>>     It has become clear that in today=E2=80=99s networking, people wil=
l want to
>>     push more =E2=80=9Cparameters=E2=80=9Dthan =E2=80=9Cpolicy=E2=80=9D=
, and I think we=E2=80=99ll just have to
>>     acknowledge that.
>>
>>     Personally I think it=E2=80=99ll pan out like this: Intent will ha=
ve
>>     sections per autonomic function. An Intent parser will validate
>>     overall correctness (signature, etc), then segment the Intent into=

>>     its bits: One part will be the autonomic fabric itself, then there=

>>     will be bits for each autonomic function.
>>
>>     So the Intent parser on a node will grab intent, see there is some=

>>     Intent for autonomic function A, will see whether this is running
>>     locally, and if so, it=E2=80=99ll just pass that piece of Intent t=
o that
>>     function.
>>
>>     In this way, a function owner defines both the intent itself, and
>>     how it is consumed by his/her function.
>>
>>     The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and peo=
ple can do what
>>     they want, in the worst case even push config, which is clearly NO=
T
>>     what Intent is meant to do. I=E2=80=99d hope that people don=E2=80=
=99t start writing
>>     Intent like =E2=80=9Cnode 17: here is your CLI; node 23: here is y=
our CLI=E2=80=9D;
>>     node 37: here is your CLI=E2=80=9D. But what I describe above COUL=
D be
>>     abused in this way. RFC7575 says this is NOT Intent.
>>
>>     In other words: Let=E2=80=99s not be too prescriptive, and hope th=
at folks
>>     will do more or less reasonable things.
>>
>>     I=E2=80=99ve been meaning to comment on the Intent drafts and poss=
ibly
>>     contribute, I=E2=80=99m just struggling with priorities at the mom=
ent (as
>>     you can see by my relative silence over the past weeks).
>>
>>     Michael
>>
>>     *From:*Anima [mailto:anima-bounces@ietf.org
>>     ] *On Behalf Of *Duzongpeng
>>     *Sent:* 16 March 2016 02:38
>>     *To:* anima@ietf.org 
>>     *Cc:* Laurent Ciavaglia >     >; J=C3=A9ferson Campos Nobre
>>     >; Sheng Jiang
>>     >
>>     *Subject:* [Anima] Does Autonomic network need some intervention
>>     beyond intent
>>
>>     Hello, everyone
>>
>>               I understand that Autonomic network should be able to wo=
rk
>>     with as less configuration as possible. In the ideal case, an
>>     Autonomic network should be able to work well while only intent is=

>>     needed from human operators.
>>
>>               However, I wonder whether Autonomic network may need som=
e
>>     intervention from human operators through some low-level policies
>>     beyond intents.
>>
>>               For example, in the
>>     https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,=

>>     it is suggested that the prefix lengths for the CSG, ASG, RSG
>>     (different roles in IP RAN) can be assigned as an "intent". These
>>     configuration parameters need to be distributed in the autonomic
>>     domain to influence the detail configurations on each autonomic no=
de.
>>
>>               As intent is described as an abstract, declarative,
>>     high-level policy used to operate an autonomic domain, such as an
>>     enterprise network. Perhaps we can rename the =E2=80=9Cintent=E2=80=
=9D in the
>>     =E2=80=9Cprefix-management ID=E2=80=9D to some other words.
>>
>>     Is =E2=80=9CASA parameters" ok here?
>>
>>     I think some characteristics for this kind of low-level policies a=
re
>>
>>     1.Network level parameters, need to be distributed by GRASP and AC=
P
>>
>>     2.Relatively static compared to intent, perhaps only needed to be
>>     configured once
>>
>>     3.Unavoidable in current network environment, mostly for
>>     establishing the network infrastructure
>>
>>     4.Rarely need coordinations with others parameters, usually is
>>     closed related to one specific autonomic function
>>
>>     Welcome for comment. Your suggestions will be appreciated.
>>
>>     Best Regards
>>
>>     Zongpeng Du
>>
>>
>>     _______________________________________________
>>     Anima mailing list
>>     Anima@ietf.org 
>>     https://www.ietf.org/mailman/listinfo/anima
>>
>>
>>
>>
>> --=20
>>
>> regards,
>>
>> John
>>
>>
>>
>>
>> --=20
>>
>> regards,
>>
>> John
>>
>>
>>
>> _______________________________________________
>> Anima mailing list
>> Anima@ietf.org
>> https://www.ietf.org/mailman/listinfo/anima
>>
>=20
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>=20


From nobody Sat Mar 26 22:13:08 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6B72B12D0C9 for ; Sat, 26 Mar 2016 22:13:06 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IMrCXW7g8EQv for ; Sat, 26 Mar 2016 22:13:02 -0700 (PDT)
Received: from mail-lf0-x22a.google.com (mail-lf0-x22a.google.com [IPv6:2a00:1450:4010:c07::22a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6136A12D0B6 for ; Sat, 26 Mar 2016 22:13:01 -0700 (PDT)
Received: by mail-lf0-x22a.google.com with SMTP id f67so20897276lfb.0 for ; Sat, 26 Mar 2016 22:13:01 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=bAxdClQhh5icCYn3JJyHy7dy3GsJH26Q4aYhVE04/hY=; b=d3hlsvPE3Kng2XXoFNr99LlzW+mvxCJ3CLkQLP9SvZA6T4cw/XqoIgCBBMew5NpLyw niCa0HqzZB7LTDWfIgDvQXlXg8ey57MgixAYoXjH078Jh207dZh20lh361YK0sccLRfE 1J88WhcOjLM5rmHBWzx2pEl1nRVs6+79ylJJGsuoIQwilNovT3bX8Ixu70Rcx79d5gdF JgfPmOfWb0ZwidOZUXE2grticrx0xDo3kkPIWSzgQgus2qsOa9iVGPpsb0oCz+Ge6k3j Q6X+IJeh4b6EqFZ9IaVFhY4swqD1YmGeyyySmtfNTO0vRkxrkoiNxyNtrqDYDykISj8M 1NYw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=bAxdClQhh5icCYn3JJyHy7dy3GsJH26Q4aYhVE04/hY=; b=DQiLSn7fq0zTAIQq5rpHRdEK0MjKxnSz3zKUCk9xpdKNJP633FZm7P8ohG6nXs99nl A8ND+OcB67eYS+K3K752rQzEoAy13uFZQ2VkQQ+DT1RsbM4zpwvdR4cRzf214xAYmETA HHGqru8RgudDaB6gZTa87uVgE68C7VIdqgDzNQPMAJE3YZnLSz8WP08gO1VxHH+2KgTq lAk4yl5bRG9tfw4j5eRoMQKviJkjk0XSNH1A/Hz9CFTD5MXTDDgRU9CguOEMVyj73gp2 aT01yRN3B8c92k5SRh+Rqyk/2f7ha+XVsf7o1LZfnPgqGZY6saD215W06V1xRIM46zHl aUPw==
X-Gm-Message-State: AD7BkJJA4Y+HWXzJLRBIliUJf+76avinnp/ThJTSo7kN6pTw0U7Pql/f18LGQXr9IVRb08qglicIStgDqUajhw==
MIME-Version: 1.0
X-Received: by 10.25.205.7 with SMTP id d7mr8171811lfg.70.1459055579210; Sat, 26 Mar 2016 22:12:59 -0700 (PDT)
Received: by 10.25.22.19 with HTTP; Sat, 26 Mar 2016 22:12:58 -0700 (PDT)
In-Reply-To: <56F6DAA1.8000206@gmail.com>
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>     <56F60CE4.9040901@gmail.com>  <56F6DAA1.8000206@gmail.com>
Date: Sat, 26 Mar 2016 22:12:58 -0700
Message-ID: 
From: John Strassner 
To: Brian E Carpenter , John Strassner 
Content-Type: multipart/alternative; boundary=001a1142091a81a7f4052f00d9d8
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sun, 27 Mar 2016 05:13:06 -0000

--001a1142091a81a7f4052f00d9d8
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Brian

Sorry, I'm trying to picture in my head the user model that you
(and presumably Zongpeng) have.

> I can't speak for Zongpeng, but I was triggered by the mention of
> northbound and southbound APIs. There is no presumed direction in
> Anima relationships, since in the limit a network is self-forming;
> all relationships are peer-to-peer.

Sure, but **intent** is **not** self-forming.

> Of course, when we put Intent into the picture, there's a presumption
> that Intent comes from some authority such as a NOC, but that's an
> external semantic, not an assumption of the autonomic model.

Well, no - if the autonomic system is going to use intent, or at least
respond to it, then it must understand those external semantics.
This is even more important if the network is supposed to form
according to one or more intent rules.

If intent is formulated at an abstract level, how is an autonomic
network going to consume intent that may have drastically different
syntactical structures, and likely, different semantics associated with
those syntactical structures?

> I think the point is that AN *adds* a new mode of communication
> between self-managing nodes, that doesn't exist in a strictly
> north/south scenario. That is meant to reduce the amount of
> north/south communication and reduce the amount of
> centralized work. So there's a trade-off - more autonomic
> messaging means less north-south messaging.

There may indeed be a new mode of communication required.
However, if intent is not standardized, how does this work?

I understand how unstructured and structured P2P networks
work. Let's assume this is a structured P2P network. In that case,
a DHT is used to assign ownership of each file to a peer. I fail to
understand how this works in an intent case. Files are passive,
intent is active. Please explain.

Regards,
John

On Sat, Mar 26, 2016 at 11:53 AM, Brian E Carpenter <
brian.e.carpenter@gmail.com> wrote:

> Hi John,
> On 27/03/2016 04:54, John Strassner wrote:
> >> Correct, and there is also no presumption that individual nodes
> >> have perfect knowledge of the topology. Of course the ACP
> >> routing protocol will have perfect knowledge of the ACP topology,
> >> but individual ASAs only know what they have gleaned from
> >> discovery and what they have been told by Intent.
> >>
> >> It is, I think, a very different world view than the SDN or SUPA or
> >> NVO4 or NFV view.
> >
> > Please note that I did NOT say anything about SDN, NFV, or SUPA
> > here. In fact, I'm confused as to why you both think I did.
>
> I can't speak for Zongpeng, but I was triggered by the mention of
> northbound and southbound APIs. There is no presumed direction in
> Anima relationships, since in the limit a network is self-forming;
> all relationships are peer-to-peer.
>
> Of course, when we put Intent into the picture, there's a presumption
> that Intent comes from some authority such as a NOC, but that's an
> external semantic, not an assumption of the autonomic model.
>
> > I specifically said "There are plenty of southbound APIs (from a
> > management function to a device). There are very few northbound APIs."
> >
> > Note the word "management function".
> >
> > For the record, I don't think that either Phase 2 of NFV or the SDN 1.1
> > architecture have treated policy in any detail. Both architectures lack=
 a
> > policy management component. I hope that ANIMA doesn't make the
> > same mistake.
>
> Certainly, Intent needs a number of properties such as self-consistency
> and stability that imply it is managed in some way.
>
> >>> However, in ANIMA network, perhaps we do not have a centralized
> >>> parsing node (controller), which can replace almost all the
> >>> communications between nodes.
> >
> > Earlier I wrote about intent requiring a translation function. I never
> said
> > that it was centralized or distributed. In the autonomic architectures =
I
> > have built, it was in fact a decentralized (agent-driven) function.
> > Furthermore, it did not "replace almost all the communications" - it
> > was used for compilation. There was, in fact, a separate distribution
> > mechanism (which I have also described), but that was limited to
> > "just" distribution.
> >
> > I really don't understand what "replacing communications" means.
>
> I think the point is that AN *adds* a new mode of communication between
> self-managing nodes, that doesn't exist in a strictly north/south
> scenario. That is meant to reduce the amount of north/south communication
> and reduce the amount of centralized work. So there's a trade-off -
> more autonomic messaging means less north-south messaging.
>
> Best regards
>    Brian
>
> >
> > regards,
> > John
> >
> > On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <
> > brian.e.carpenter@gmail.com> wrote:
> >
> >>> However, in ANIMA network, perhaps we do not have a centralized parsi=
ng
> >> node (controller), which can replace almost all the communications
> between
> >> nodes.
> >>
> >> Correct, and there is also no presumption that individual nodes have
> >> perfect
> >> knowledge of the topology. Of course the ACP routing protocol will hav=
e
> >> perfect
> >> knowledge of the ACP topology, but individual ASAs only know what they
> have
> >> gleaned from discovery and what they have been told by Intent.
> >>
> >> It is, I think, a very different world view than the SDN or SUPA or NV=
O4
> >> or NFV view.
> >>
> >> Regards
> >>    Brian
> >>
> >> On 26/03/2016 16:46, Duzongpeng wrote:
> >>> Hi John:
> >>>
> >>> Thanks for your reply.
> >>>
> >>> 
> >>> How is this different than running a script, or invoking an API? Ther=
e
> >> are plenty of southbound APIs (from a management function to a device)=
.
> >> There are very few northbound APIs. That's what the operators that I
> have
> >> talked to want.
> >>> 
> >>>
> >>> I agree that intent belongs to northbound interface in the SDN networ=
k.
> >> And ANIMA network perhaps can share the same northbound interface.
> >>>
> >>> But, my understanding is that ANIMA network is a little different fro=
m
> >> SDN network.
> >>>
> >>> SDN controller will parse the intent, and transform to detailed
> >> configurations of every device. It is a good model, and little
> >> communications between nodes are needed.
> >>>
> >>> Some network level parameters can be handled in SDN controller.
> >>>
> >>> However, in ANIMA network, perhaps we do not have a centralized parsi=
ng
> >> node (controller), which can replace almost all the communications
> between
> >> nodes.
> >>>
> >>> So when a network operator is operating a ANIMA network, the network
> >> level parameters are still needed to be flooded.
> >>>
> >>> I mean the ANIMA interface is perhaps something like =E2=80=9Cnorthbo=
und + part
> >> of southbound=E2=80=9D.
> >>>
> >>> Best Regards
> >>> Zongpeng Du
> >>>
> >>>
> >>> From: John Strassner [mailto:strazpdj@gmail.com]
> >>> Sent: Saturday, March 26, 2016 9:41 AM
> >>> To: Duzongpeng; John Strassner
> >>> Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang;
> >> anima@ietf.org; J=C3=A9ferson Campos Nobre
> >>> Subject: Re: [Anima] Does Autonomic network need some intervention
> >> beyond intent
> >>>
> >>> Hi Zongpeng,
> >>>
> >>> I think we have a misunderstanding here.
> >>>
> >>>
> >>> But I do not support that the intent writer should look every detail =
in
> >> the catalog.
> >>> 
> >>> Most catalogs that I have seen have very few details. They instead te=
ll
> >> the user what services or products are available.
> >>> 
> >>>
> >>> The intent writer may mainly care about what they want, but how to do
> it
> >> can be left to the autonomic network to deal with.
> >>> 
> >>> I agree completely.
> >>> 
> >>>
> >>> Perhaps several solutions exist, and autonomic management system choo=
se
> >> one according to the abilities collected.
> >>> 
> >>> This is one of the reasons why I would like to see capabilities (and
> >> needs) advertised.
> >>> 
> >>>
> >>> ...
> >>> I mean in current status, some of the Autonomic Functions should be
> >> exposed to the network operator, and some of the Autonomic Functions
> can be
> >> hidden to the network operator
> >>> ...
> >>> 
> >>> How is this different than running a script, or invoking an API? Ther=
e
> >> are plenty of southbound APIs (from a management function to a device)=
.
> >> There are very few northbound APIs. That's what the operators that I
> have
> >> talked to want.
> >>>
> >>> regards,
> >>> John
> >>>
> >>> On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng  >> > wrote:
> >>> Hi John
> >>>
> >>> Some personal understanding here. If any misunderstanding, please
> >> correct me. Thanks.
> >>>
> >>> 
> >>> I would suggest a slight modification, to enable users of
> >>> intent (and even intent writers) to NOT have to have
> >>> detailed implementation information.
> >>>
> >>> What if all of the Autonomic Functions, after they joined
> >>> the network, advertised both their capabilities as well as
> >>> their needs? We can put aside the latter (needs) as this
> >>> is more complex, but if the capabilities are advertised,
> >>> then the autonomic management system can assemble
> >>> those needs into a catalog. Then, the intent writer can
> >>> look at the catalog and write a declarative request to
> >>> use the functionality of the catalog.
> >>>
> >>> I will look at your YANG model in a bit (sorry, I'm woefully
> >>> behind...).
> >>> < /jcs2>
> >>>
> >>> I agree that autonomic node should have that self-advertisement
> ability,
> >> and autonomic management system should have this catalog.
> >>>
> >>> But I do not support that the intent writer should look every detail =
in
> >> the catalog.
> >>>
> >>> The intent writer may mainly care about what they want, but how to do
> it
> >> can be left to the autonomic network to deal with.
> >>> Perhaps several solutions exist, and autonomic management system choo=
se
> >> one according to the abilities collected.
> >>>
> >>> In the ideal, the network operator may communicate with the autonomic
> >> network using network-level level intent and network-level report.
> >>> I mean in current status, some of the Autonomic Functions should be
> >> exposed to the network operator, and some of the Autonomic Functions
> can be
> >> hidden to the network operator
> >>> Perhaps in future, when the network nodes become more intelligent, th=
e
> >> hidden part will increase.
> >>>
> >>>
> >>> Best regards
> >>> Zongpeng Du
> >>>
> >>> From: Anima [mailto:anima-bounces@ietf.org anima-bounces@ietf.org>]
> >> On Behalf Of John Strassner
> >>> Sent: Thursday, March 24, 2016 10:41 AM
> >>> To: Michael Behringer (mbehring); John Strassner
> >>> Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org >> anima@ietf.org>; J=C3=A9ferson Campos Nobre
> >>> Subject: Re: [Anima] Does Autonomic network need some intervention
> >> beyond intent
> >>>
> >>> Hi Michael,
> >>>
> >>> thanks for your reply. I do think that we largely agree, and would
> enjoy
> >> chatting during or (hopefully before) BA. Please see ...
> Also
> >> trying to unclutter, so I am eliding points that we agree on.
> >>>
> >>>
> >>> While *CIEs don=E2=80=99t like magic, they do understand the concept =
of a
> >> default behaviour which they can influence if needed. Any reasonable
> >> network management approach tries to define =E2=80=9Cdefault=E2=80=9D =
behaviour and
> >> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=
=E2=80=9D part very nicely, and
> *CIEs
> >> get that.
> >>> 
> >>> Agreed!
> >>> 
> >>>
> >>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discu=
ssion: My view is that we
> shouldn=E2=80=99t
> >> try to define exactly *what* intent allows and doesn=E2=80=99t allow, =
since
> we=E2=80=99ve
> >> heard from many sides that the boundary is not clear, and it may be
> >> impossible to have a really clear, usable definition. So my point was:
> >> Let=E2=80=99s focus on the distribution model: Intent is flooded to al=
l nodes.
> >> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific no=
des, etc,
> >> because that would blur the boundary even worse.
> >>> 
> >>> I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v=
1
> >> was 2006, almost 10 years ago), we used a message bus and pub-sub. In
> v2,
> >> we flooded; in v3, we went back to pub-sub because we wanted to do som=
e
> >> optimization and auditing. But I would be happy to try some type of
> >> controlled flooding again.
> >>> 
> >>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
> >> owners define the content of what they want to push in their bit of th=
e
> >> Intent. If we insist of full flooding, we encourage the right behaviou=
r.
> >> And if someone really puts into his Intent =E2=80=9Cnode A: do this; n=
ode B: do
> the
> >> other; node C: do the third=E2=80=9D it is awkward enough for people t=
o take a
> step
> >> back and think whether they=E2=80=99re doing the right thing =E2=98=BA
> >>> 
> >>> I agree in principle with what you said. Plus, see above.
> >>> < /jcs2>
> >>> 
> >>> Do you mean that the intent writer will **know the difference**
> >>> between the fabric and each autonomic function? I hope not,
> >>> but perhaps I'm misunderstanding you...
> >>> < /jcs>
> >>> Not sure I understand your concern. Yes, in my model there is a secti=
on
> >> for the fabric, and sections for the various functions. Seems a natura=
l
> >> structure?!? 
> >>> The intent **writer** would have to know implementation
> >>> details in order to place everything in a single file. I would
> >>> rather not require this, and instead, have the intent engine
> >>> (e.g., compiler and associated logic) do this.
> >>> < /jcs2>
> >>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to stru=
cture Intent around
> >> Autonomic Functions. This is not optimal from a scientific point of
> view,
> >> but aligns very well with the way how things are organised. Probably w=
e
> >> need to discuss that in BA. But refer to my YANG model sent a few days
> ago.
> >>> 
> >>> I would suggest a slight modification, to enable users of
> >>> intent (and even intent writers) to NOT have to have
> >>> detailed implementation information.
> >>>
> >>> What if all of the Autonomic Functions, after they joined
> >>> the network, advertised both their capabilities as well as
> >>> their needs? We can put aside the latter (needs) as this
> >>> is more complex, but if the capabilities are advertised,
> >>> then the autonomic management system can assemble
> >>> those needs into a catalog. Then, the intent writer can
> >>> look at the catalog and write a declarative request to
> >>> use the functionality of the catalog.
> >>>
> >>> I will look at your YANG model in a bit (sorry, I'm woefully
> >>> behind...).
> >>> < /jcs2>
> >>>
> >>>
> >>> regards,
> >>> John
> >>>
> >>> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <
> >> mbehring@cisco.com> wrote:
> >>> John,
> >>>
> >>> I think we=E2=80=99re on the same page. Couple of points, top posting=
 here,
> >> otherwise it gets too messy.
> >>>
> >>> While *CIEs don=E2=80=99t like magic, they do understand the concept =
of a
> >> default behaviour which they can influence if needed. Any reasonable
> >> network management approach tries to define =E2=80=9Cdefault=E2=80=9D =
behaviour and
> >> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefault=
=E2=80=9D part very nicely, and
> *CIEs
> >> get that.
> >>>
> >>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D discu=
ssion: My view is that we
> shouldn=E2=80=99t
> >> try to define exactly *what* intent allows and doesn=E2=80=99t allow, =
since
> we=E2=80=99ve
> >> heard from many sides that the boundary is not clear, and it may be
> >> impossible to have a really clear, usable definition. So my point was:
> >> Let=E2=80=99s focus on the distribution model: Intent is flooded to al=
l nodes.
> >> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific no=
des, etc,
> >> because that would blur the boundary even worse.
> >>>
> >>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
> >> owners define the content of what they want to push in their bit of th=
e
> >> Intent. If we insist of full flooding, we encourage the right behaviou=
r.
> >> And if someone really puts into his Intent =E2=80=9Cnode A: do this; n=
ode B: do
> the
> >> other; node C: do the third=E2=80=9D it is awkward enough for people t=
o take a
> step
> >> back and think whether they=E2=80=99re doing the right thing =E2=98=BA
> >>>
> >>> 
> >>> Do you mean that the intent writer will **know the difference**
> >>> between the fabric and each autonomic function? I hope not,
> >>> but perhaps I'm misunderstanding you...
> >>> 
> >>>
> >>> Not sure I understand your concern. Yes, in my model there is a secti=
on
> >> for the fabric, and sections for the various functions. Seems a natura=
l
> >> structure?!?
> >>>
> >>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to stru=
cture Intent around
> >> Autonomic Functions. This is not optimal from a scientific point of
> view,
> >> but aligns very well with the way how things are organised. Probably w=
e
> >> need to discuss that in BA. But refer to my YANG model sent a few days
> ago.
> >>>
> >>> We should schedule some discussion time outside the ANIMA meeting
> >> window...
> >>>
> >>> Michael
> >>>
> >>>
> >>> From: John Strassner [mailto:strazpdj@gmail.com >> strazpdj@gmail.com>]
> >>> Sent: 22 March 2016 21:32
> >>> To: Michael Behringer (mbehring)  >> mbehring@cisco.com>>; John Strassner  >> strazpdj@gmail.com>>
> >>> Cc: Duzongpeng >;
> >> anima@ietf.org; Laurent Ciavaglia <
> >> laurent.ciavaglia@nokia.com>;
> >> J=C3=A9ferson Campos Nobre  >>;
> >> Sheng Jiang >
> >>> Subject: Re: [Anima] Does Autonomic network need some intervention
> >> beyond intent
> >>>
> >>> In general, I agree with Michael. I'd like to emphasize and expand on=
 a
> >> couple of key points:
> >>>
> >>>
> >>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=
=9D or =E2=80=9Csteering=E2=80=9D a
> network,
> >> at least not initially. There is still traditional config, plus all th=
e
> >> existing and emerging SDN models. My view has always been that ANIMA
> sets a
> >> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are=
 more specific and
> >> can override intent. This is described in RFC7575. Main point here:
> ANIMA
> >> Intent isn=E2=80=99t the only tool in a network.
> >>>
> >>> 
> >>> Absolutely agree. Note that even in the Congress case I mentioned
> >>> earlier, it is NOT pushing config, and it is NOT viewed as the
> >>> "only" policy
> >>> 
> >>>
> >>>
> >>> Intent could perfectly well cover a high level policy such as =E2=80=
=9Call
> nodes
> >> of type x do this; type y does that=E2=80=9D. (Which I think is what y=
ou mention
> >> below.) When we started with Intent, the vision was to keep it very hi=
gh
> >> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=98=
=BA
> >>>
> >>> 
> >>> Agreed. Although to me, this means that we need to consider the types
> of
> >> actors that are using intent. For example, most of my CCIE and JCIE
> friends
> >> have no interest in using an intent language for pushing configuration=
.
> On
> >> the other hand, when I talk to operators and application developers,
> they
> >> would love to be able to use a high-level intent and have the intent
> engine
> >> take care of the details for them.
> >>> 
> >>>
> >>> It has become clear that in today=E2=80=99s networking, people will w=
ant to
> push
> >> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and I=
 think we=E2=80=99ll just have to
> acknowledge
> >> that.
> >>>
> >>> 
> >>> Sorry Michael, I don't follow here. What do you mean
> >>> by "parameters"?
> >>> 
> >>>
> >>>
> >>> Personally I think it=E2=80=99ll pan out like this: Intent will have =
sections
> >> per autonomic function. An Intent parser will validate overall
> correctness
> >> (signature, etc), then segment the Intent into its bits: One part will
> be
> >> the autonomic fabric itself, then there will be bits for each autonomi=
c
> >> function.
> >>>
> >>> 
> >>> Do you mean that the intent writer will **know the difference**
> >>> between the fabric and each autonomic function? I hope not,
> >>> but perhaps I'm misunderstanding you...
> >>> 
> >>>
> >>>
> >>> So the Intent parser on a node will grab intent, see there is some
> >> Intent for autonomic function A, will see whether this is running
> locally,
> >> and if so, it=E2=80=99ll just pass that piece of Intent to that functi=
on.
> >>>
> >>> 
> >>> Agreed.
> >>> 
> >>>
> >>>
> >>> In this way, a function owner defines both the intent itself, and how
> it
> >> is consumed by his/her function.
> >>>
> >>> 
> >>> I'm not sure what you mean by "function owner". (Sorry to be
> >>> picky, I'm likely under-caffeinated.)
> >>>
> >>> The above seemed like an autonomous decision, made by the parser.
> >>> I don't think that the intent writer should be cognizant of how many =
or
> >>> what types of autonomic functions are present.
> >>> 
> >>>
> >>>
> >>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people=
 can do what they
> >> want, in the worst case even push config, which is clearly NOT what
> Intent
> >> is meant to do.
> >>>
> >>> 
> >>> And hence, the "HAL 2000" scenario. :-)
> >>> 
> >>>
> >>>
> >>> I=E2=80=99d hope that people don=E2=80=99t start writing Intent like =
=E2=80=9Cnode 17: here is
> >> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your CL=
I=E2=80=9D. But
> what
> >> I describe above COULD be abused in this way. RFC7575 says this is NOT
> >> Intent.
> >>>
> >>> 
> >>> Absolutely agree!
> >>> 
> >>>
> >>>  regards,
> >>> John
> >>>
> >>>
> >>>
> >>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
> >> mbehring@cisco.com> wrote:
> >>> Hi Zongpeng,
> >>>
> >>> Yes, this is a recurring question and topic. Let me give my view on t=
wo
> >> levels:
> >>>
> >>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=
=9D or =E2=80=9Csteering=E2=80=9D a
> network,
> >> at least not initially. There is still traditional config, plus all th=
e
> >> existing and emerging SDN models. My view has always been that ANIMA
> sets a
> >> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means are=
 more specific and
> >> can override intent. This is described in RFC7575. Main point here:
> ANIMA
> >> Intent isn=E2=80=99t the only tool in a network.
> >>>
> >>> Intent could perfectly well cover a high level policy such as =E2=80=
=9Call
> nodes
> >> of type x do this; type y does that=E2=80=9D. (Which I think is what y=
ou mention
> >> below.) When we started with Intent, the vision was to keep it very hi=
gh
> >> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=98=
=BA  It has become clear
> that
> >> in today=E2=80=99s networking, people will want to push more =E2=80=9C=
parameters=E2=80=9D than
> >> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to ackno=
wledge that.
> >>>
> >>> Personally I think it=E2=80=99ll pan out like this: Intent will have =
sections
> >> per autonomic function. An Intent parser will validate overall
> correctness
> >> (signature, etc), then segment the Intent into its bits: One part will
> be
> >> the autonomic fabric itself, then there will be bits for each autonomi=
c
> >> function.
> >>>
> >>> So the Intent parser on a node will grab intent, see there is some
> >> Intent for autonomic function A, will see whether this is running
> locally,
> >> and if so, it=E2=80=99ll just pass that piece of Intent to that functi=
on.
> >>>
> >>> In this way, a function owner defines both the intent itself, and how
> it
> >> is consumed by his/her function.
> >>>
> >>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and people=
 can do what they
> >> want, in the worst case even push config, which is clearly NOT what
> Intent
> >> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start writi=
ng Intent like
> =E2=80=9Cnode
> >> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: her=
e is your
> >> CLI=E2=80=9D. But what I describe above COULD be abused in this way. R=
FC7575
> says
> >> this is NOT Intent.
> >>>
> >>> In other words: Let=E2=80=99s not be too prescriptive, and hope that =
folks will
> >> do more or less reasonable things.
> >>>
> >>> I=E2=80=99ve been meaning to comment on the Intent drafts and possibl=
y
> >> contribute, I=E2=80=99m just struggling with priorities at the moment =
(as you
> can
> >> see by my relative silence over the past weeks).
> >>>
> >>> Michael
> >>>
> >>>
> >>>
> >>> From: Anima [mailto:anima-bounces@ietf.org anima-bounces@ietf.org>]
> >> On Behalf Of Duzongpeng
> >>> Sent: 16 March 2016 02:38
> >>> To: anima@ietf.org
> >>> Cc: Laurent Ciavaglia  >> laurent.ciavaglia@nokia.com>>; J=C3=A9ferson Campos Nobre <
> jcnobre@inf.ufrgs.br
> >> >; Sheng Jiang   >> jiangsheng@huawei.com>>
> >>> Subject: [Anima] Does Autonomic network need some intervention beyond
> >> intent
> >>>
> >>> Hello, everyone
> >>>
> >>>          I understand that Autonomic network should be able to work
> with
> >> as less configuration as possible. In the ideal case, an Autonomic
> network
> >> should be able to work well while only intent is needed from human
> >> operators.
> >>>
> >>>          However, I wonder whether Autonomic network may need some
> >> intervention from human operators through some low-level policies beyo=
nd
> >> intents.
> >>>
> >>>          For example, in the
> >> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it
> is
> >> suggested that the prefix lengths for the CSG, ASG, RSG (different
> roles in
> >> IP RAN) can be assigned as an "intent". These configuration parameters
> need
> >> to be distributed in the autonomic domain to influence the detail
> >> configurations on each autonomic node.
> >>>
> >>>          As intent is described as an abstract, declarative, high-lev=
el
> >> policy used to operate an autonomic domain, such as an enterprise
> network.
> >> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cpre=
fix-management ID=E2=80=9D to some
> >> other words.
> >>>
> >>> Is =E2=80=9CASA parameters" ok here?
> >>>
> >>> I think some characteristics for this kind of low-level policies are
> >>>
> >>> 1.       Network level parameters, need to be distributed by GRASP an=
d
> >> ACP
> >>>
> >>> 2.       Relatively static compared to intent, perhaps only needed to
> be
> >> configured once
> >>>
> >>> 3.       Unavoidable in current network environment, mostly for
> >> establishing the network infrastructure
> >>>
> >>> 4.       Rarely need coordinations with others parameters, usually is
> >> closed related to one specific autonomic function
> >>>
> >>> Welcome for comment. Your suggestions will be appreciated.
> >>>
> >>>
> >>> Best Regards
> >>> Zongpeng Du
> >>>
> >>> _______________________________________________
> >>> Anima mailing list
> >>> Anima@ietf.org
> >>> https://www.ietf.org/mailman/listinfo/anima
> >>>
> >>>
> >>>
> >>> --
> >>> regards,
> >>> John
> >>>
> >>>
> >>>
> >>> --
> >>> regards,
> >>> John
> >>>
> >>>
> >>>
> >>> --
> >>> regards,
> >>> John
> >>>
> >>>
> >>>
> >>> _______________________________________________
> >>> Anima mailing list
> >>> Anima@ietf.org
> >>> https://www.ietf.org/mailman/listinfo/anima
> >>>
> >>
> >>
> >
> >
>
>


--=20
regards,
John

--001a1142091a81a7f4052f00d9d8
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


Hi Brian

Sorry, I'm tryi=
ng to picture in my head the user model that you
(and presumably =
Zongpeng) have.

> I can't speak for Zongpen=
g, but I was triggered by the mention of
 > northbound and southbound=
 APIs. There is no presumed direction in
 > Anima relationships, sinc=
e in the limit a network is self-forming;
 > all relationships are pe=
er-to-peer.

Sure, but **intent** is **not** se=
lf-forming.

>=C2=A0Of course, when we put Intent into the =
picture, there's a presumption
 > that Intent comes from some aut=
hority such as a NOC, but that's an
 > external semantic, not an =
assumption of the autonomic model.

Well, no -=C2=
=A0if the autonomic system is going to use intent, or at least
re=
spond to it, then it must understand those external semantics.
Th=
is is even more important if the network is supposed to form
acco=
rding to one or more intent rules.

If intent is fo=
rmulated at an abstract level, how is an autonomic
network going =
to consume intent that may have drastically different
syntactical=
 structures, and likely, different semantics associated with
thos=
e syntactical structures?

> I think the point i=
s that AN *adds* a new mode of communication
>=C2=A0between se=
lf-managing nodes, that doesn't exist in a strictly
> nort=
h/south scenario. That is meant to reduce the amount of
> nort=
h/south communication and reduce the amount of
> centralized w=
ork. So there's a trade-off - more autonomic
> messaging m=
eans less north-south messaging.

There may ind=
eed be a new mode of communication required.
However, if intent i=
s not standardized, how does this work?

I understa=
nd how unstructured and structured P2P networks
work. Let's a=
ssume this is a structured P2P network. In that case,
a DHT is us=
ed to assign ownership of each file to a peer. I fail to
understa=
nd how this works in an intent case. Files are passive,
intent is=
 active. Please explain.

Regards,
John

On =
Sat, Mar 26, 2016 at 11:53 AM, Brian E Carpenter <brian.e.carpen=
ter@gmail.com> wrote:
Hi Jo=
hn,

On 27/03/2016 04:54, John Strassner wrote:

>> Correct, and there is also no presumption that individual nodes
>> have perfect knowledge of the topology. Of course the ACP

>> routing protocol will have perfect knowledge of the ACP topology,<=
br>
>> but individual ASAs only know what they have gleaned from

>> discovery and what they have been told by Intent.

>>

>> It is, I think, a very different world view than the SDN or SUPA o=
r

>> NVO4 or NFV view.

>

> Please note that I did NOT say anything about SDN, NFV, or SUPA

> here. In fact, I'm confused as to why you both think I did.



I can't speak for Zongpeng, but I was triggered by the mention of

northbound and southbound APIs. There is no presumed direction in

Anima relationships, since in the limit a network is self-forming;

all relationships are peer-to-peer.



Of course, when we put Intent into the picture, there's a presumption
that Intent comes from some authority such as a NOC, but that's an

external semantic, not an assumption of the autonomic model.



> I specifically said "There are plenty of southbound APIs (from a<=
br>
> management function to a device). There are very few northbound APIs.&=
quot;

>

> Note the word "management function".

>

> For the record, I don't think that either Phase 2 of NFV or the SD=
N 1.1

> architecture have treated policy in any detail. Both architectures lac=
k a

> policy management component. I hope that ANIMA doesn't make the
> same mistake.



Certainly, Intent needs a number of properties such as self-consistency

and stability that imply it is managed in some way.



>>> However, in ANIMA network, perhaps we do not have a centralize=
d

>>> parsing node (controller), which can replace almost all the
>>> communications between nodes.

>

> Earlier I wrote about intent requiring a translation function. I never=
 said

> that it was centralized or distributed. In the autonomic architectures=
 I

> have built, it was in fact a decentralized (agent-driven) function.
> Furthermore, it did not "replace almost all the communications&qu=
ot; - it

> was used for compilation. There was, in fact, a separate distribution<=
br>
> mechanism (which I have also described), but that was limited to

> "just" distribution.

>

> I really don't understand what "replacing communications"=
; means.



I think the point is that AN *adds* a new mode of communication between

self-managing nodes, that doesn't exist in a strictly north/south

scenario. That is meant to reduce the amount of north/south communication
and reduce the amount of centralized work. So there's a trade-off -

more autonomic messaging means less north-south messaging.



Best regards

=C2=A0 =C2=A0Brian



>

> regards,

> John

>

> On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <

> brian.e.carpenter@gmail=
.com> wrote:

>

>>> However, in ANIMA network, perhaps we do not have a centralize=
d parsing

>> node (controller), which can replace almost all the communications=
 between

>> nodes.

>>

>> Correct, and there is also no presumption that individual nodes ha=
ve

>> perfect

>> knowledge of the topology. Of course the ACP routing protocol will=
 have

>> perfect

>> knowledge of the ACP topology, but individual ASAs only know what =
they have

>> gleaned from discovery and what they have been told by Intent.

>>

>> It is, I think, a very different world view than the SDN or SUPA o=
r NVO4

>> or NFV view.

>>

>> Regards

>>=C2=A0 =C2=A0 Brian

>>

>> On 26/03/2016 16:46, Duzongpeng wrote:

>>> Hi John:

>>>

>>> Thanks for your reply.

>>>

>>> <jcs>

>>> How is this different than running a script, or invoking an AP=
I? There

>> are plenty of southbound APIs (from a management function to a dev=
ice).

>> There are very few northbound APIs. That's what the operators =
that I have

>> talked to want.

>>> </jcs>

>>>

>>> I agree that intent belongs to northbound interface in the SDN=
 network.

>> And ANIMA network perhaps can share the same northbound interface.=


>>>

>>> But, my understanding is that ANIMA network is a little differ=
ent from

>> SDN network.

>>>

>>> SDN controller will parse the intent, and transform to detaile=
d

>> configurations of every device. It is a good model, and little

>> communications between nodes are needed.

>>>

>>> Some network level parameters can be handled in SDN controller=
.

>>>

>>> However, in ANIMA network, perhaps we do not have a centralize=
d parsing

>> node (controller), which can replace almost all the communications=
 between

>> nodes.

>>>

>>> So when a network operator is operating a ANIMA network, the n=
etwork

>> level parameters are still needed to be flooded.

>>>

>>> I mean the ANIMA interface is perhaps something like =E2=80=9C=
northbound + part

>> of southbound=E2=80=9D.

>>>

>>> Best Regards

>>> Zongpeng Du

>>>

>>>

>>> From: John Strassner [mailto:strazpdj@gmail.com]

>>> Sent: Saturday, March 26, 2016 9:41 AM

>>> To: Duzongpeng; John Strassner

>>> Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jia=
ng;

>> anima@ietf.org; J=C3=A9ferso=
n Campos Nobre

>>> Subject: Re: [Anima] Does Autonomic network need some interven=
tion

>> beyond intent

>>>

>>> Hi Zongpeng,

>>>

>>> I think we have a misunderstanding here.

>>>

>>>

>>> But I do not support that the intent writer should look every =
detail in

>> the catalog.

>>> <jcs>

>>> Most catalogs that I have seen have very few details. They ins=
tead tell

>> the user what services or products are available.

>>> </jcs>

>>>

>>> The intent writer may mainly care about what they want, but ho=
w to do it

>> can be left to the autonomic network to deal with.

>>> <jcs>

>>> I agree completely.

>>> </jcs>

>>>

>>> Perhaps several solutions exist, and autonomic management syst=
em choose

>> one according to the abilities collected.

>>> <jcs>

>>> This is one of the reasons why I would like to see capabilitie=
s (and

>> needs) advertised.

>>> </jcs>

>>>

>>> ...

>>> I mean in current status, some of the Autonomic Functions shou=
ld be

>> exposed to the network operator, and some of the Autonomic Functio=
ns can be

>> hidden to the network operator

>>> ...

>>> <jcs>

>>> How is this different than running a script, or invoking an AP=
I? There

>> are plenty of southbound APIs (from a management function to a dev=
ice).

>> There are very few northbound APIs. That's what the operators =
that I have

>> talked to want.

>>>

>>> regards,

>>> John

>>>

>>> On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng <duzongpeng@huawei.com

>> <mailto:duzongpeng@hua=
wei.com>> wrote:

>>> Hi John

>>>

>>> Some personal understanding here. If any misunderstanding, ple=
ase

>> correct me. Thanks.

>>>

>>> <jcs2>

>>> I would suggest a slight modification, to enable users of

>>> intent (and even intent writers) to NOT have to have

>>> detailed implementation information.

>>>

>>> What if all of the Autonomic Functions, after they joined

>>> the network, advertised both their capabilities as well as

>>> their needs? We can put aside the latter (needs) as this

>>> is more complex, but if the capabilities are advertised,

>>> then the autonomic management system can assemble

>>> those needs into a catalog. Then, the intent writer can

>>> look at the catalog and write a declarative request to

>>> use the functionality of the catalog.

>>>

>>> I will look at your YANG model in a bit (sorry, I'm woeful=
ly

>>> behind...).

>>> < /jcs2>

>>>

>>> I agree that autonomic node should have that self-advertisemen=
t ability,

>> and autonomic management system should have this catalog.

>>>

>>> But I do not support that the intent writer should look every =
detail in

>> the catalog.

>>>

>>> The intent writer may mainly care about what they want, but ho=
w to do it

>> can be left to the autonomic network to deal with.

>>> Perhaps several solutions exist, and autonomic management syst=
em choose

>> one according to the abilities collected.

>>>

>>> In the ideal, the network operator may communicate with the au=
tonomic

>> network using network-level level intent and network-level report.=


>>> I mean in current status, some of the Autonomic Functions shou=
ld be

>> exposed to the network operator, and some of the Autonomic Functio=
ns can be

>> hidden to the network operator

>>> Perhaps in future, when the network nodes become more intellig=
ent, the

>> hidden part will increase.

>>>

>>>

>>> Best regards

>>> Zongpeng Du

>>>

>>> From: Anima [mailto:=
anima-bounces@ietf.org<mailto:anima-bounces@ietf.org>]

>> On Behalf Of John Strassner

>>> Sent: Thursday, March 24, 2016 10:41 AM

>>> To: Michael Behringer (mbehring); John Strassner

>>> Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org<mailto:

>> anima@ietf.org>; J=C3=A9f=
erson Campos Nobre

>>> Subject: Re: [Anima] Does Autonomic network need some interven=
tion

>> beyond intent

>>>

>>> Hi Michael,

>>>

>>> thanks for your reply. I do think that we largely agree, and w=
ould enjoy

>> chatting during or (hopefully before) BA. Please see <jcs2>.=
.</jcs2>. Also

>> trying to unclutter, so I am eliding points that we agree on.

>>>

>>>

>>> While *CIEs don=E2=80=99t like magic, they do understand the c=
oncept of a

>> default behaviour which they can influence if needed. Any reasonab=
le

>> network management approach tries to define =E2=80=9Cdefault=E2=80=
=9D behaviour and

>> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefa=
ult=E2=80=9D part very nicely, and *CIEs

>> get that.

>>> <jcs2>

>>> Agreed!

>>> </jcs2>

>>>

>>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=
=9D discussion: My view is that we shouldn=E2=80=99t

>> try to define exactly *what* intent allows and doesn=E2=80=99t all=
ow, since we=E2=80=99ve

>> heard from many sides that the boundary is not clear, and it may b=
e

>> impossible to have a really clear, usable definition. So my point =
was:

>> Let=E2=80=99s focus on the distribution model: Intent is flooded t=
o all nodes.

>> Let=E2=80=99s NOT do selective flooding, unicast Intent to specifi=
c nodes, etc,

>> because that would blur the boundary even worse.

>>> <jcs2>

>>> I like flooding intent to all nodes. In FOCALE (v1-3, and reme=
mber, v1

>> was 2006, almost 10 years ago), we used a message bus and pub-sub.=
 In v2,

>> we flooded; in v3, we went back to pub-sub because we wanted to do=
 some

>> optimization and auditing. But I would be happy to try some type o=
f

>> controlled flooding again.

>>> </jcs2>

>>> My suggestion: Intent is flooded to all nodes in a domain. The=
 ASA

>> owners define the content of what they want to push in their bit o=
f the

>> Intent. If we insist of full flooding, we encourage the right beha=
viour.

>> And if someone really puts into his Intent =E2=80=9Cnode A: do thi=
s; node B: do the

>> other; node C: do the third=E2=80=9D it is awkward enough for peop=
le to take a step

>> back and think whether they=E2=80=99re doing the right thing =E2=
=98=BA

>>> <jcs2>

>>> I agree in principle with what you said. Plus, see above.

>>> < /jcs2>

>>> <jcs>

>>> Do you mean that the intent writer will **know the difference*=
*

>>> between the fabric and each autonomic function? I hope not,
>>> but perhaps I'm misunderstanding you...

>>> < /jcs>

>>> Not sure I understand your concern. Yes, in my model there is =
a section

>> for the fabric, and sections for the various functions. Seems a na=
tural

>> structure?!? <jcs2>

>>> The intent **writer** would have to know implementation

>>> details in order to place everything in a single file. I would=


>>> rather not require this, and instead, have the intent engine
>>> (e.g., compiler and associated logic) do this.

>>> < /jcs2>

>>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting =
to structure Intent around

>> Autonomic Functions. This is not optimal from a scientific point o=
f view,

>> but aligns very well with the way how things are organised. Probab=
ly we

>> need to discuss that in BA. But refer to my YANG model sent a few =
days ago.

>>> <jcs2>

>>> I would suggest a slight modification, to enable users of

>>> intent (and even intent writers) to NOT have to have

>>> detailed implementation information.

>>>

>>> What if all of the Autonomic Functions, after they joined

>>> the network, advertised both their capabilities as well as

>>> their needs? We can put aside the latter (needs) as this

>>> is more complex, but if the capabilities are advertised,

>>> then the autonomic management system can assemble

>>> those needs into a catalog. Then, the intent writer can

>>> look at the catalog and write a declarative request to

>>> use the functionality of the catalog.

>>>

>>> I will look at your YANG model in a bit (sorry, I'm woeful=
ly

>>> behind...).

>>> < /jcs2>

>>>

>>>

>>> regards,

>>> John

>>>

>>> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) =
<

>> mbehring@cisco.com<ma=
ilto:mbehring@cisco.com>> w=
rote:

>>> John,

>>>

>>> I think we=E2=80=99re on the same page. Couple of points, top =
posting here,

>> otherwise it gets too messy.

>>>

>>> While *CIEs don=E2=80=99t like magic, they do understand the c=
oncept of a

>> default behaviour which they can influence if needed. Any reasonab=
le

>> network management approach tries to define =E2=80=9Cdefault=E2=80=
=9D behaviour and

>> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefa=
ult=E2=80=9D part very nicely, and *CIEs

>> get that.

>>>

>>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=
=9D discussion: My view is that we shouldn=E2=80=99t

>> try to define exactly *what* intent allows and doesn=E2=80=99t all=
ow, since we=E2=80=99ve

>> heard from many sides that the boundary is not clear, and it may b=
e

>> impossible to have a really clear, usable definition. So my point =
was:

>> Let=E2=80=99s focus on the distribution model: Intent is flooded t=
o all nodes.

>> Let=E2=80=99s NOT do selective flooding, unicast Intent to specifi=
c nodes, etc,

>> because that would blur the boundary even worse.

>>>

>>> My suggestion: Intent is flooded to all nodes in a domain. The=
 ASA

>> owners define the content of what they want to push in their bit o=
f the

>> Intent. If we insist of full flooding, we encourage the right beha=
viour.

>> And if someone really puts into his Intent =E2=80=9Cnode A: do thi=
s; node B: do the

>> other; node C: do the third=E2=80=9D it is awkward enough for peop=
le to take a step

>> back and think whether they=E2=80=99re doing the right thing =E2=
=98=BA

>>>

>>> <jcs>

>>> Do you mean that the intent writer will **know the difference*=
*

>>> between the fabric and each autonomic function? I hope not,
>>> but perhaps I'm misunderstanding you...

>>> </jcs>

>>>

>>> Not sure I understand your concern. Yes, in my model there is =
a section

>> for the fabric, and sections for the various functions. Seems a na=
tural

>> structure?!?

>>>

>>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting =
to structure Intent around

>> Autonomic Functions. This is not optimal from a scientific point o=
f view,

>> but aligns very well with the way how things are organised. Probab=
ly we

>> need to discuss that in BA. But refer to my YANG model sent a few =
days ago.

>>>

>>> We should schedule some discussion time outside the ANIMA meet=
ing

>> window...

>>>

>>> Michael

>>>

>>>

>>> From: John Strassner [mailto:strazpdj@gmail.com<mailto:

>> strazpdj@gmail.com>]<=
br>
>>> Sent: 22 March 2016 21:32

>>> To: Michael Behringer (mbehring) <mbehring@cisco.com<mailto:

>> mbehring@cisco.com>&g=
t;; John Strassner <strazpdj@gmail=
.com<mailto:

>> strazpdj@gmail.com>&g=
t;

>>> Cc: Duzongpeng <du=
zongpeng@huawei.com<mailto:=
duzongpeng@huawei.com>>;

>> anima@ietf.org<mailto:anima@ietf.org>; Laurent Ciavaglia &l=
t;

>> laurent.ciavaglia@n=
okia.com<mailto:laure=
nt.ciavaglia@nokia.com>>;

>> J=C3=A9ferson Campos Nobre <jcnobre@inf.ufrgs.br<mailto:jcnobre@inf.ufrgs.br>>;

>> Sheng Jiang <jiangshen=
g@huawei.com<mailto:jiangsh=
eng@huawei.com>>

>>> Subject: Re: [Anima] Does Autonomic network need some interven=
tion

>> beyond intent

>>>

>>> In general, I agree with Michael. I'd like to emphasize an=
d expand on a

>> couple of key points:

>>>

>>>

>>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=
=E2=80=9D or =E2=80=9Csteering=E2=80=9D a network,

>> at least not initially. There is still traditional config, plus al=
l the

>> existing and emerging SDN models. My view has always been that ANI=
MA sets a

>> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means=
 are more specific and

>> can override intent. This is described in RFC7575. Main point here=
: ANIMA

>> Intent isn=E2=80=99t the only tool in a network.

>>>

>>> <jcs>

>>> Absolutely agree. Note that even in the Congress case I mentio=
ned

>>> earlier, it is NOT pushing config, and it is NOT viewed as the=


>>> "only" policy

>>> </jcs>

>>>

>>>

>>> Intent could perfectly well cover a high level policy such as =
=E2=80=9Call nodes

>> of type x do this; type y does that=E2=80=9D. (Which I think is wh=
at you mention

>> below.) When we started with Intent, the vision was to keep it ver=
y high

>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =
=E2=98=BA

>>>

>>> <jcs>

>>> Agreed. Although to me, this means that we need to consider th=
e types of

>> actors that are using intent. For example, most of my CCIE and JCI=
E friends

>> have no interest in using an intent language for pushing configura=
tion. On

>> the other hand, when I talk to operators and application developer=
s, they

>> would love to be able to use a high-level intent and have the inte=
nt engine

>> take care of the details for them.

>>> </jcs>

>>>

>>> It has become clear that in today=E2=80=99s networking, people=
 will want to push

>> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, a=
nd I think we=E2=80=99ll just have to acknowledge

>> that.

>>>

>>> <jcs>

>>> Sorry Michael, I don't follow here. What do you mean

>>> by "parameters"?

>>> </jcs>

>>>

>>>

>>> Personally I think it=E2=80=99ll pan out like this: Intent wil=
l have sections

>> per autonomic function. An Intent parser will validate overall cor=
rectness

>> (signature, etc), then segment the Intent into its bits: One part =
will be

>> the autonomic fabric itself, then there will be bits for each auto=
nomic

>> function.

>>>

>>> <jcs>

>>> Do you mean that the intent writer will **know the difference*=
*

>>> between the fabric and each autonomic function? I hope not,
>>> but perhaps I'm misunderstanding you...

>>> </jcs>

>>>

>>>

>>> So the Intent parser on a node will grab intent, see there is =
some

>> Intent for autonomic function A, will see whether this is running =
locally,

>> and if so, it=E2=80=99ll just pass that piece of Intent to that fu=
nction.

>>>

>>> <jcs>

>>> Agreed.

>>> </jcs>

>>>

>>>

>>> In this way, a function owner defines both the intent itself, =
and how it

>> is consumed by his/her function.

>>>

>>> <jcs>

>>> I'm not sure what you mean by "function owner". =
(Sorry to be

>>> picky, I'm likely under-caffeinated.)

>>>

>>> The above seemed like an autonomous decision, made by the pars=
er.

>>> I don't think that the intent writer should be cognizant o=
f how many or

>>> what types of autonomic functions are present.

>>> </jcs>

>>>

>>>

>>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and=
 people can do what they

>> want, in the worst case even push config, which is clearly NOT wha=
t Intent

>> is meant to do.

>>>

>>> <jcs>

>>> And hence, the "HAL 2000" scenario. :-)

>>> </jcs>

>>>

>>>

>>> I=E2=80=99d hope that people don=E2=80=99t start writing Inten=
t like =E2=80=9Cnode 17: here is

>> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is you=
r CLI=E2=80=9D. But what

>> I describe above COULD be abused in this way. RFC7575 says this is=
 NOT

>> Intent.

>>>

>>> <jcs>

>>> Absolutely agree!

>>> </jcs>

>>>

>>>=C2=A0 regards,

>>> John

>>>

>>>

>>>

>>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) =
<

>> mbehring@cisco.com<ma=
ilto:mbehring@cisco.com>> w=
rote:

>>> Hi Zongpeng,

>>>

>>> Yes, this is a recurring question and topic. Let me give my vi=
ew on two

>> levels:

>>>

>>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=
=E2=80=9D or =E2=80=9Csteering=E2=80=9D a network,

>> at least not initially. There is still traditional config, plus al=
l the

>> existing and emerging SDN models. My view has always been that ANI=
MA sets a

>> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means=
 are more specific and

>> can override intent. This is described in RFC7575. Main point here=
: ANIMA

>> Intent isn=E2=80=99t the only tool in a network.

>>>

>>> Intent could perfectly well cover a high level policy such as =
=E2=80=9Call nodes

>> of type x do this; type y does that=E2=80=9D. (Which I think is wh=
at you mention

>> below.) When we started with Intent, the vision was to keep it ver=
y high

>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =
=E2=98=BA=C2=A0 It has become clear that

>> in today=E2=80=99s networking, people will want to push more =E2=
=80=9Cparameters=E2=80=9D than

>> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to a=
cknowledge that.

>>>

>>> Personally I think it=E2=80=99ll pan out like this: Intent wil=
l have sections

>> per autonomic function. An Intent parser will validate overall cor=
rectness

>> (signature, etc), then segment the Intent into its bits: One part =
will be

>> the autonomic fabric itself, then there will be bits for each auto=
nomic

>> function.

>>>

>>> So the Intent parser on a node will grab intent, see there is =
some

>> Intent for autonomic function A, will see whether this is running =
locally,

>> and if so, it=E2=80=99ll just pass that piece of Intent to that fu=
nction.

>>>

>>> In this way, a function owner defines both the intent itself, =
and how it

>> is consumed by his/her function.

>>>

>>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and=
 people can do what they

>> want, in the worst case even push config, which is clearly NOT wha=
t Intent

>> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start w=
riting Intent like =E2=80=9Cnode

>> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37:=
 here is your

>> CLI=E2=80=9D. But what I describe above COULD be abused in this wa=
y. RFC7575 says

>> this is NOT Intent.

>>>

>>> In other words: Let=E2=80=99s not be too prescriptive, and hop=
e that folks will

>> do more or less reasonable things.

>>>

>>> I=E2=80=99ve been meaning to comment on the Intent drafts and =
possibly

>> contribute, I=E2=80=99m just struggling with priorities at the mom=
ent (as you can

>> see by my relative silence over the past weeks).

>>>

>>> Michael

>>>

>>>

>>>

>>> From: Anima [mailto:=
anima-bounces@ietf.org<mailto:anima-bounces@ietf.org>]

>> On Behalf Of Duzongpeng

>>> Sent: 16 March 2016 02:38

>>> To: anima@ietf.org<ma=
ilto:anima@ietf.org>

>>> Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com<mailto:

>> laurent.ciavaglia@n=
okia.com>>; J=C3=A9ferson Campos Nobre <jcnobre@inf.ufrgs.br

>> <mailto:jcnobre@inf.ufr=
gs.br>>; Sheng Jiang <jiangsheng@huawei.com<mailto:

>> jiangsheng@huawei.com=
>>

>>> Subject: [Anima] Does Autonomic network need some intervention=
 beyond

>> intent

>>>

>>> Hello, everyone

>>>

>>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 I understand that Autonomic =
network should be able to work with

>> as less configuration as possible. In the ideal case, an Autonomic=
 network

>> should be able to work well while only intent is needed from human=


>> operators.

>>>

>>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 However, I wonder whether Au=
tonomic network may need some

>> intervention from human operators through some low-level policies =
beyond

>> intents.

>>>

>>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 For example, in the

>> https://tools.ietf.org/htm=
l/draft-ietf-anima-prefix-management-00, it is

>> suggested that the prefix lengths for the CSG, ASG, RSG (different=
 roles in

>> IP RAN) can be assigned as an "intent". These configurat=
ion parameters need

>> to be distributed in the autonomic domain to influence the detail<=
br>
>> configurations on each autonomic node.

>>>

>>>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 As intent is described as an=
 abstract, declarative, high-level

>> policy used to operate an autonomic domain, such as an enterprise =
network.

>> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=
=9Cprefix-management ID=E2=80=9D to some

>> other words.

>>>

>>> Is =E2=80=9CASA parameters" ok here?

>>>

>>> I think some characteristics for this kind of low-level polici=
es are

>>>

>>> 1.=C2=A0 =C2=A0 =C2=A0 =C2=A0Network level parameters, need to=
 be distributed by GRASP and

>> ACP

>>>

>>> 2.=C2=A0 =C2=A0 =C2=A0 =C2=A0Relatively static compared to int=
ent, perhaps only needed to be

>> configured once

>>>

>>> 3.=C2=A0 =C2=A0 =C2=A0 =C2=A0Unavoidable in current network en=
vironment, mostly for

>> establishing the network infrastructure

>>>

>>> 4.=C2=A0 =C2=A0 =C2=A0 =C2=A0Rarely need coordinations with ot=
hers parameters, usually is

>> closed related to one specific autonomic function

>>>

>>> Welcome for comment. Your suggestions will be appreciated.

>>>

>>>

>>> Best Regards

>>> Zongpeng Du

>>>

>>> _______________________________________________

>>> Anima mailing list

>>> Anima@ietf.org<mailto=
:Anima@ietf.org>

>>> https://www.ietf.org/mailman/listinfo/anima<=
/a>

>>>

>>>

>>>

>>> --

>>> regards,

>>> John

>>>

>>>

>>>

>>> --

>>> regards,

>>> John

>>>

>>>

>>>

>>> --

>>> regards,

>>> John

>>>

>>>

>>>

>>> _______________________________________________

>>> Anima mailing list

>>> Anima@ietf.org

>>> https://www.ietf.org/mailman/listinfo/anima<=
/a>

>>>

>>

>>

>

>







-- 
regards,
John




--001a1142091a81a7f4052f00d9d8--


From nobody Sun Mar 27 10:32:31 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 44A8012D1EC for ; Sun, 27 Mar 2016 10:32:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.699
X-Spam-Level: 
X-Spam-Status: No, score=-2.699 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qEXMJKAG2wpX for ; Sun, 27 Mar 2016 10:32:27 -0700 (PDT)
Received: from mail-lf0-x236.google.com (mail-lf0-x236.google.com [IPv6:2a00:1450:4010:c07::236]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 666FC127058 for ; Sun, 27 Mar 2016 10:32:26 -0700 (PDT)
Received: by mail-lf0-x236.google.com with SMTP id c62so77561720lfc.1 for ; Sun, 27 Mar 2016 10:32:26 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc; bh=AmvvNfg80XcdB7I5vzR0ZiIVm8to9jClaCIssC64Zo4=; b=lrf9pQ696ItPIKN7V7NfvCn7w6XbHti6Nr8jgdDmp5MjmljBbekILvxJxD3WIph8BV SnexsAC8nRgpHXWP0NnmtlXHLVmgAgVV8jLfEtwVQPPMixFSyjVyERM0KjOW65dlYM7l fKV6RAB0dmDelTdBa5rWTlWY4CWG7ct4KE0IscaTLnga2HGTK6l5sgC3eKJP1VHTcGtr pKTHX3u7f304dbHh2jKbWoWkdt8Kky3vu1VUzRj/pUeyKMpq7cItLXbZuZi3PQTnDNCY BwdXmSuZVRySFzgi5kTe5Fgf2rmMnnV2JODNzeNypu+9+SAyB0/4oAWfkIkBqzES6h2h 8V+g==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc; bh=AmvvNfg80XcdB7I5vzR0ZiIVm8to9jClaCIssC64Zo4=; b=XtIDj6/bFzo9GttQ4WtmcCF2ZJAn9UfyO4EWBh96aFcMAciPcWtkP0g2QI6r2gSvlE 5kxqxKPzwCO63oss2c5j7P7W8KYGiPdrMHPlqTKwugOqWBSN9RtpfKf/6XO1creWV428 jy7GtRDowHcFOyiarbh+/HFp1eoFK1VYAEgNl2MGJtZeTKZ8oDzFAh3NUrY3YU6bTgrt duhm6p2CAyUsFnNbnz0E7zSF3KXaJrcmlEWBSKUqYbzYsI2ZDHId56nFxauEkVK9yUG3 kAuBXRbCGXigjzGrugkIE+TelxO+grzW3YjsrTJU2iNwfDP7oB9S+y5bULoDSvyb6SRc PUww==
X-Gm-Message-State: AD7BkJKXK7sPhn6X+p1Vx7+AOGDcZuu8IcOofx0cM7ECYs3ZMhsUh6uCMkw3YK5yT8ge/TTKavHblu0eh2k7uw==
MIME-Version: 1.0
X-Received: by 10.25.218.196 with SMTP id r187mr8921853lfg.6.1459099944471; Sun, 27 Mar 2016 10:32:24 -0700 (PDT)
Received: by 10.25.22.19 with HTTP; Sun, 27 Mar 2016 10:32:24 -0700 (PDT)
In-Reply-To: 
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com> <56F4718D.8010104@alcatel-lucent.com>  
Date: Sun, 27 Mar 2016 10:32:24 -0700
Message-ID: 
From: John Strassner 
To: "Michael Behringer (mbehring)" 
Content-Type: multipart/alternative; boundary=001a11401f3ce1d3f3052f0b2d9b
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=C3=A9ferson_Campos_Nobre?= , "anima@ietf.org" 
Subject: Re: [Anima] Reference Model Issues
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sun, 27 Mar 2016 17:32:30 -0000

--001a11401f3ce1d3f3052f0b2d9b
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

HI Michael,

I agree. I'll work on detailed responses after Easter.

regards,
John

On Fri, Mar 25, 2016 at 7:23 AM, Michael Behringer (mbehring) <
mbehring@cisco.com> wrote:

> Yes, I=E2=80=99m thinking that we move EVERYTHING that isn=E2=80=99t in t=
he current
> charter into a =E2=80=9Cfutures=E2=80=9D section. So that would include:
>
> -          Intent
>
> -          feedback loops
>
> -          aggregated reporting
>
> -          APIs
>
> -          current section 9
>
> -          constrained devices
>
> -          cross-domain behaviour
>
> -          etc
>
>
>
> This way the document would be very clearly structured, and we could stil=
l
> keep the current content, which, frankly, I don=E2=80=99t want to lose.
>
>
>
> Michael
>
>
>
>
>
> *From:* J=C3=A9ferson Campos Nobre [mailto:jcnobre@inf.ufrgs.br]
> *Sent:* 25 March 2016 15:17
> *To:* Laurent Ciavaglia ; Michael Behringer
> (mbehring) ; anima@ietf.org
> *Subject:* Re: [Anima] Reference Model Issues
>
>
>
> Hi.
>
> + 1 for the grouping.
>
> Michael, are you thinking in maintaining subsections in the future work
> section?
>
> Best.
>
> J=C3=A9ferson
>
>
>
> Em qui, 24 de mar de 2016 =C3=A0s 20:00, Laurent Ciavaglia <
> laurent.ciavaglia@nokia.com> escreveu:
>
> Dear Michael, all,
>
> Thanks for your continuous care of the document and taking some hindsight=
!
>
> I am also more in favor of grouping non-chartered aspects into a future
> work section, and allow the text there to be meaningful for 1) the
> relationship with (chartered) ANIMA components, 2) the overall picture, a=
nd
> 3) describing the implications of these future items with enough depth.
>
> I think the reference document is a pivot one in ANIMA to prepare what
> will come next (in ANIMA or elsewhere) and improve our collective
> understanding on future areas to investigate.
> I think it would be a mistake to neglect this opportunity by restricting
> the document on chartered items.
>
> Thanks, Laurent.
> ps: of course my above support has nothing to do with the fact that I'm
> too lazy busy to condense the section I am editing ;)
>
>
>
> On 24/03/2016 18:52, EXT Michael Behringer (mbehring) wrote:
>
> I just read through the entire document again, noting down things that ne=
ed fixing, or discussion. The below is my personal view, for discussion. I =
cleaned up the github repository, and created a new file, corresponding to =
the latest version plus some small edits I made today while reading through=
 again.
>
>
>
> One important point: Our chairs want us to take out or reduce drastically=
 all the stuff that doesn't belong into the current charter. I agree that i=
t is not always 100% clear if something is out of charter. But I'm having a=
 hard time reducing some sections. What if we make a new section "Future wo=
rk" (which clearly indicates out of scope for this charter), where we colle=
ct those points, but then are a bit more liberal about the volume of each s=
ection? I think that would be a good compromise. Sheng?
>
>
>
> I'll summarise the below points in the WG meeting in BA. Maybe we can alr=
eady clarify some stuff before.
>
>
>
> Michael
>
>
>
>
>
>
>
>
>
> Reference model issues
>
> 24 Mar 2016
>
>
>
> 4.1 Naming:
>
> - "It is recommended that the names are generated by the autonomic nodes
>
>    themselves." - how? Should names / addresses not come from the registr=
ar?
>
>    Suggestion: Name *and* ACP address should be assigned by the registrar
>
>    at enrolment time. zone-id is dynamic, rest is fixed.
>
>
>
> - "a specific naming convention is out of scope":
>
>    Suggestion: We define a default naming and addressing scheme.
>
>    1st we settle on an addressing scheme from the ACP draft.
>
>    That defines a "device-ID" (last n bits of ACP address).
>
>    convert that to a string, and use it as a name. This would result in
>
>    something like: 0123-4567-89ab-0001.example.com.
>
>    The name would go into the domain certificate --> need to put this int=
o
>
>    the bootstrap draft as well.
>
>
>
> - to write in the doc: "registrar picks a naming scheme; all registrars
>
>   in a domain must use the same scheme. Example is: ...
>
>
>
> 4.4 ACP
>
> - should move to the end of section 4, it breaks the flow where it is rig=
ht now.
>
>
>
> 4.7 Intent Distribution
>
> - This section should only explain the basics, real work is out of scope =
for this phase.
>
> - should point to the intent distribution draft
>
> - should explain that intent is flooded to all nodes in a domain (if we a=
gree),
>
>   or other methods (if not).
>
> - should explain that we expect Intent to have a long life time (months),=
 thus
>
>   Intent distribution is expected to be very infrequent.
>
> - should explain that the entire Intent file is flooded in one go (if we =
agree).
>
>
>
> 5. Functional Overview
>
> - title is not good. This section really describes how an autonomic node =
behaves. maybe call it "Behaviour of an autonomic node"?
>
> - we need to describe the bring-up better, specifically insecure discover=
y,
>
>   ACP negotiation, ACP bringup, and subsequent operations, and which prot=
ocols to use
>
>   where. Or should this go into the ACP draft??
>
>
>
> 6. Security and Trust Infrastructure
>
> - needs a review from someone in the security space.
>
> - we need to describe the certificate format; this should likely go into =
bootstrap draft. This draft should explain where the certificate format is =
described.
>
> - the domain certificate should also contain the ACP IP address (where th=
e zone bits are set to zero).
>
>
>
> 7.1
>
> - broken paragraph; needs fixing.
>
> - there is a mention of constrained devices - take out.
>
>
>
> 7.2 Specific ASAs
>
> - should explain MUST implement
>
> - generally, need to work through text in this section; relate back to se=
ction 5.
>
>
>
> 8.1
>
> - remove last reference to RFC7575.
>
>
>
> 8.2 Intent
>
> - explain not part of current charter first.
>
> - not sure whether we should keep this level of detail, or shorten furthe=
r?
>
>
>
> 8.3 Aggregated Reporting
>
> - explain not part of current charter first.
>
> - my feeling: Should be reduced to 1-2 paragraphs, only high level idea.
>
>
>
> General:
>
> - should we combine all items that are not part of the current charter in=
to a separate section "Future Topics" or something like that? Then all (*) =
topic would be nicely contained. Yes, I think I'd prefer that to taking the=
m out completely, or shortening them to next-to-nothing.
>
>
>
> 10. Security Considerations
>
> - needs to be worked over.
>
>
>
> General:
>
> - uppercase Intent everywhere in the doc.
>
>
>
> _______________________________________________
>
> Anima mailing list
>
> Anima@ietf.org
>
> https://www.ietf.org/mailman/listinfo/anima
>
>
>
>
>
> --
>
> Laurent Ciavaglia
>
> Senior Research Manager
>
> Bell Labs, Nokia
>
>
>
> +33 160 402 636
>
> Route de Villejust | 91620 Nozay | France
>
> LinkedIn: laurentciavaglia 
>
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>


--=20
regards,
John

--001a11401f3ce1d3f3052f0b2d9b
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


HI Michael,

I agree. I'l=
l work on detailed responses after Easter.

regards=
,
John

On Fri, Mar 25, 2016 at 7:23 AM, Michael Behringer (mbehring) <mbehring@cisco.com> wrote:











=
Yes, I=E2=80=99m thinking that we move EVERYTHING that =
isn=E2=80=99t in the current charter into a =E2=80=9Cfutures=E2=80=9D
 section. So that would include: 


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
Intent

-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
feedback loops<=
/font>


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
aggregated reporting


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
APIs


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
current section 9


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
constrained devices


-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
cross-domain behaviour=



-=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0
etc


=
=C2=A0


=
This way the document would be very clearly structured,=
 and we could still keep the current
 content, which, frankly, I don=E2=80=99t want to lose. 


=
=C2=A0


=
Michael


=
=C2=A0


=
=C2=A0








From:
 J=C3=A9ferson Campos Nobre [mailto:jcnobre@inf.ufrgs.br] 

Sent: 25 March 2016 15:17
To: Laurent Ciavaglia <laurent.ciav=
aglia@nokia.com>; Michael Behringer (mbehring) <mbehring@cisco.com>; anima@ietf.org

Subject: Re: [Anima] Referen=
ce Model Issues






=C2=A0








Hi.






+ 1 for the grouping.






Michael, are you thinking in maintaining subsections i=
n the future work section?






Best.






J=C3=A9ferson






=C2=A0






Em qui, 24 de mar de 2016 =C3=A0s 20:00, Laurent Ciava=
glia <l=
aurent.ciavaglia@nokia.com> escreveu:







Dear Michael, all,



Thanks for your continuous care of the document and taking some hindsight!<=
br>


I am also more in favor of grouping non-chartered aspects into a future wor=
k section, and allow the text there to be meaningful for 1) the relationshi=
p with (chartered) ANIMA components, 2) the overall picture, and 3) describ=
ing the implications of these future
 items with enough depth.



I think the reference document is a pivot one in ANIMA to prepare what will=
 come next (in ANIMA or elsewhere) and improve our collective understanding=
 on future areas to investigate.

I think it would be a mistake to neglect this opportunity by restricting th=
e document on chartered items.










Thanks, Laurent.

ps: of course my above support has nothing to do with the fact that I'm=
 too lazy busy to condense the section I am editing ;)=







=C2=A0




On 24/03/2016 18:52, EXT Michael Behringer (mbehring) =
wrote:






I=
 just read through the entire document again, noting down things that need =
fixing, or discussion. The below is my personal view, for discussion. I cle=
aned up the github repository, and created a new file, corresponding to the=
 latest version plus some small edits I made today while reading through ag=
ain. 


<=
u>=C2=A0


O=
ne important point: Our chairs want us to take out or reduce drastically al=
l the stuff that doesn't belong into the current charter. I agree that =
it is not always 100% clear if something is out of charter. But I'm hav=
ing a hard time reducing some sections. What if we make a new section "=
;Future work" (which clearly indicates out of scope for this charter),=
 where we collect those points, but then are a bit more liberal about the v=
olume of each section? I think that would be a good compromise. Sheng? <=
/u>


<=
u>=C2=A0


I=
'll summarise the below points in the WG meeting in BA. Maybe we can al=
ready clarify some stuff before. 


<=
u>=C2=A0


M=
ichael


<=
u>=C2=A0


<=
u>=C2=A0


<=
u>=C2=A0


<=
u>=C2=A0


R=
eference model issues


2=
4 Mar 2016


<=
u>=C2=A0


4=
.1 Naming: 


-=
 "It is recommended that the names are generated by the autonomic node=
s


=
=C2=A0=C2=A0 themselves." - how? Should names / addresses not come fro=
m the registrar? 


=
=C2=A0=C2=A0=C2=A0Suggestion: Name *and* ACP address should be assigned by =
the registrar


=
=C2=A0=C2=A0 at enrolment time. zone-id is dynamic, rest is fixed. <=
u>


<=
u>=C2=A0


-=
 "a specific naming convention is out of scope": 


=
=C2=A0=C2=A0=C2=A0Suggestion: We define a default naming and addressing sch=
eme. 


=
=C2=A0=C2=A0=C2=A01st we settle on an addressing scheme from the ACP draft.=
 


=
=C2=A0=C2=A0=C2=A0That defines a "device-ID" (last n bits of ACP =
address). 


=
=C2=A0=C2=A0=C2=A0convert that to a string, and use it as a name. This woul=
d result in


=
=C2=A0=C2=A0 something like: 0123-4567-89ab-0001.example.com. <=
/span>


=
=C2=A0=C2=A0=C2=A0The name would go into the domain certificate --> need=
 to put this into 


=
=C2=A0=C2=A0=C2=A0the bootstrap draft as well. =



=
=C2=A0=C2=A0=C2=A0


-=
 to write in the doc: "registrar picks a naming scheme; all registrars=
 


=
=C2=A0=C2=A0in a domain must use the same scheme. Example is: ... 


=
=C2=A0=C2=A0=C2=A0


4=
.4 ACP


-=
 should move to the end of section 4, it breaks the flow where it is right =
now. 


<=
u>=C2=A0


4=
.7 Intent Distribution


-=
 This section should only explain the basics, real work is out of scope for=
 this phase.


-=
 should point to the intent distribution draft<=
/pre>

-=
 should explain that intent is flooded to all nodes in a domain (if we agre=
e), 


=
=C2=A0=C2=A0or other methods (if not). 


-=
 should explain that we expect Intent to have a long life time (months), th=
us


=
=C2=A0 Intent distribution is expected to be very infrequent. 


-=
 should explain that the entire Intent file is flooded in one go (if we agr=
ee). 


<=
u>=C2=A0


5=
. Functional Overview


-=
 title is not good. This section really describes how an autonomic node beh=
aves. maybe call it "Behaviour of an autonomic node"? <=
/u>


-=
 we need to describe the bring-up better, specifically insecure discovery, =



=
=C2=A0=C2=A0ACP negotiation, ACP bringup, and subsequent operations, and wh=
ich protocols to use


=
=C2=A0 where. Or should this go into the ACP draft?? <=
/font>


<=
u>=C2=A0


6=
. Security and Trust Infrastructure


-=
 needs a review from someone in the security space.


-=
 we need to describe the certificate format; this should likely go into boo=
tstrap draft. This draft should explain where the certificate format is des=
cribed. 


-=
 the domain certificate should also contain the ACP IP address (where the z=
one bits are set to zero). 


<=
u>=C2=A0


7=
.1 


-=
 broken paragraph; needs fixing. 


-=
 there is a mention of constrained devices - take out. 


<=
u>=C2=A0


7=
.2 Specific ASAs


-=
 should explain MUST implement


-=
 generally, need to work through text in this section; relate back to secti=
on 5. 


<=
u>=C2=A0


8=
.1 


-=
 remove last reference to RFC7575. 


<=
u>=C2=A0


8=
.2 Intent


-=
 explain not part of current charter first. 

-=
 not sure whether we should keep this level of detail, or shorten further? =



<=
u>=C2=A0


8=
.3 Aggregated Reporting


-=
 explain not part of current charter first. 

-=
 my feeling: Should be reduced to 1-2 paragraphs, only high level idea.<=
/u>


<=
u>=C2=A0


G=
eneral: 


-=
 should we combine all items that are not part of the current charter into =
a separate section "Future Topics" or something like that? Then a=
ll (*) topic would be nicely contained. Yes, I think I'd prefer that to=
 taking them out completely, or shortening them to next-to-nothing.<=
u>


<=
u>=C2=A0


1=
0. Security Considerations


-=
 needs to be worked over. 


<=
u>=C2=A0


G=
eneral: 


-=
 uppercase Intent everywhere in the doc. 


<=
u>=C2=A0


_=
______________________________________________<=
/pre>

A=
nima mailing list


<=
a href=3D"mailto:Anima@ietf.org" target=3D"_blank">Anima@ietf.org


<=
a href=3D"https://www.ietf.org/mailman/listinfo/anima" target=3D"_blank">ht=
tps://www.ietf.org/mailman/listinfo/anima

<=
u>=C2=A0




=C2=A0








--





Laurent Ciavaglia


Senior Research Manager

Bell Labs, Nokia


=C2=A0


+33=C2=A0160 402=C2=A0636



Route de Villejust | 91620 Nozay | France


LinkedIn:
<=
span lang=3D"FR">laurentciavaglia

=C2=A0








_______________________________________________

Anima mailing list

Anima@ietf.org

h=
ttps://www.ietf.org/mailman/listinfo/anima<=
/p>














_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.org/mailman/listinfo/anima





-- 
regards,
John




--001a11401f3ce1d3f3052f0b2d9b--


From nobody Sun Mar 27 12:09:31 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 65F3E12D093 for ; Sun, 27 Mar 2016 12:09:30 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level: 
X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9,  DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0M1SK5A8pUnH for ; Sun, 27 Mar 2016 12:09:27 -0700 (PDT)
Received: from mail-pa0-x229.google.com (mail-pa0-x229.google.com [IPv6:2607:f8b0:400e:c03::229]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 25D8312D133 for ; Sun, 27 Mar 2016 12:09:27 -0700 (PDT)
Received: by mail-pa0-x229.google.com with SMTP id tt10so82480225pab.3 for ; Sun, 27 Mar 2016 12:09:27 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=G5O+4+gn/5eRvCaFovGgjSnkCSaHh+Zfgu/72hwdHf4=; b=XtmuFBZ6SatLgL3AtRUte8jjeUNYNI/59K0dfig9OBJqirZ83P+AcD9jWdJ6QpmIR5 +W0Cr5QHAw9rv91CUkGmbg8jw3cvOAojR/S7+83vHCjr+vfn2AZqp5uEIkgd4dKFjZjV iQmqWG07sYqVqp14Qf8HyVcLJZ7tkT7KPgIs1ECu2DhlOkW0mr3vk0UUI3OAqq8K1K8m NxfWmjIK9BZtIaZ/A0pwKx3fRXqK9pepShMM8pt0752bWARy2YbYWkqiMhuL9CHnUfgd P2piz347+aPnL2iTtqHSaMQiJQV+ppbuTYkNsCyh71Kku2gdMXQ2X8L6xxXN4KHNxqk0 9KEw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=G5O+4+gn/5eRvCaFovGgjSnkCSaHh+Zfgu/72hwdHf4=; b=FfMSIDclB9c77SXseDfeMd9tdzIsCjyJJpX7UMRInD5pWwjPIW8MEuh23LvYQK6mzm NTz5/AKk1QmKTYH4TrK5blPZh5H/EkzGkHPvnRmXrtvBDX5mob0LR4W5LRE0clNEIxm0 WXlfm+CHspRpQMl4E9cqKfEXTLriaJDbAsCZmR1zan1wRZ5WHa0iuQNO9D4AOy7tWCab kkFQ7UfOL5AfnovnG0YHn39DYDirm25Z8aIOBoKunXKOUrLHkjGOaD9K8YC81YIYWqYT FkT9ptrHEV5kuO89AF4uZHa7/GD+VJBsvKjpJRSdZWwnCLP/L95gd1/KDwc+sjc1RZ50 4/2g==
X-Gm-Message-State: AD7BkJLxf/GHTGYb0HT4K5M08WMoMj8QVtW5OmDz4fsYog8fBgSbQR3deUF3yAoo6PEeHg==
X-Received: by 10.66.55.102 with SMTP id r6mr36825849pap.67.1459105766624; Sun, 27 Mar 2016 12:09:26 -0700 (PDT)
Received: from ?IPv6:2406:e007:7e16:1:28cc:dc4c:9703:6781? ([2406:e007:7e16:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id f8sm30088075pfj.49.2016.03.27.12.09.20 (version=TLSv1/SSLv3 cipher=OTHER); Sun, 27 Mar 2016 12:09:24 -0700 (PDT)
To: John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>     <56F60CE4.9040901@gmail.com>  <56F6DAA1.8000206@gmail.com> 
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56F82FDD.7090405@gmail.com>
Date: Mon, 28 Mar 2016 08:09:17 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sun, 27 Mar 2016 19:09:30 -0000

Hi John,

On 27/03/2016 18:12, John Strassner wrote:
> Hi Brian
>=20
> Sorry, I'm trying to picture in my head the user model that you
> (and presumably Zongpeng) have.

I believe we are closer to the same view than you might think...

>> I can't speak for Zongpeng, but I was triggered by the mention of
>> northbound and southbound APIs. There is no presumed direction in
>> Anima relationships, since in the limit a network is self-forming;
>> all relationships are peer-to-peer.
>=20
> Sure, but **intent** is **not** self-forming.

Indeed not. An ASA will have some default behaviour that amounts to havin=
g
default settings for intent, but by definition intent comes from humans.
(One qualification: if the intent that reaches a given node is inconsiste=
nt,
then resolving that inconsistency might be autonomic.)

>> Of course, when we put Intent into the picture, there's a presumption
>> that Intent comes from some authority such as a NOC, but that's an
>> external semantic, not an assumption of the autonomic model.
>=20
> Well, no - if the autonomic system is going to use intent, or at least
> respond to it, then it must understand those external semantics.
> This is even more important if the network is supposed to form
> according to one or more intent rules.

Right, but what I meant is that the network will operate perfectly well
if no intent is sent, by using defaults. That will always happen to the
extent that intent distribution will take a finite time. Until the intent=

reaches a node, it will operate on defaults. Of course, the semantics of
intent must be well-defined, but it's an extra layer.

> If intent is formulated at an abstract level, how is an autonomic
> network going to consume intent that may have drastically different
> syntactical structures, and likely, different semantics associated with=

> those syntactical structures?

Agreed. We need a model, syntax and semantics.

>=20
>> I think the point is that AN *adds* a new mode of communication
>> between self-managing nodes, that doesn't exist in a strictly
>> north/south scenario. That is meant to reduce the amount of
>> north/south communication and reduce the amount of
>> centralized work. So there's a trade-off - more autonomic
>> messaging means less north-south messaging.
>=20
> There may indeed be a new mode of communication required.
> However, if intent is not standardized, how does this work?

It doesn't. My only point is that intent is layered on top of an
underlying autonomic behaviour based on default behaviours. Actually,
that's why the question of intent was deferred in the WG charter.

    Brian

> I understand how unstructured and structured P2P networks
> work. Let's assume this is a structured P2P network. In that case,
> a DHT is used to assign ownership of each file to a peer. I fail to
> understand how this works in an intent case. Files are passive,
> intent is active. Please explain.
>=20
> Regards,
> John
>=20
> On Sat, Mar 26, 2016 at 11:53 AM, Brian E Carpenter <
> brian.e.carpenter@gmail.com> wrote:
>=20
>> Hi John,
>> On 27/03/2016 04:54, John Strassner wrote:
>>>> Correct, and there is also no presumption that individual nodes
>>>> have perfect knowledge of the topology. Of course the ACP
>>>> routing protocol will have perfect knowledge of the ACP topology,
>>>> but individual ASAs only know what they have gleaned from
>>>> discovery and what they have been told by Intent.
>>>>
>>>> It is, I think, a very different world view than the SDN or SUPA or
>>>> NVO4 or NFV view.
>>>
>>> Please note that I did NOT say anything about SDN, NFV, or SUPA
>>> here. In fact, I'm confused as to why you both think I did.
>>
>> I can't speak for Zongpeng, but I was triggered by the mention of
>> northbound and southbound APIs. There is no presumed direction in
>> Anima relationships, since in the limit a network is self-forming;
>> all relationships are peer-to-peer.
>>
>> Of course, when we put Intent into the picture, there's a presumption
>> that Intent comes from some authority such as a NOC, but that's an
>> external semantic, not an assumption of the autonomic model.
>>
>>> I specifically said "There are plenty of southbound APIs (from a
>>> management function to a device). There are very few northbound APIs.=
"
>>>
>>> Note the word "management function".
>>>
>>> For the record, I don't think that either Phase 2 of NFV or the SDN 1=
=2E1
>>> architecture have treated policy in any detail. Both architectures la=
ck a
>>> policy management component. I hope that ANIMA doesn't make the
>>> same mistake.
>>
>> Certainly, Intent needs a number of properties such as self-consistenc=
y
>> and stability that imply it is managed in some way.
>>
>>>>> However, in ANIMA network, perhaps we do not have a centralized
>>>>> parsing node (controller), which can replace almost all the
>>>>> communications between nodes.
>>>
>>> Earlier I wrote about intent requiring a translation function. I neve=
r
>> said
>>> that it was centralized or distributed. In the autonomic architecture=
s I
>>> have built, it was in fact a decentralized (agent-driven) function.
>>> Furthermore, it did not "replace almost all the communications" - it
>>> was used for compilation. There was, in fact, a separate distribution=

>>> mechanism (which I have also described), but that was limited to
>>> "just" distribution.
>>>
>>> I really don't understand what "replacing communications" means.
>>
>> I think the point is that AN *adds* a new mode of communication betwee=
n
>> self-managing nodes, that doesn't exist in a strictly north/south
>> scenario. That is meant to reduce the amount of north/south communicat=
ion
>> and reduce the amount of centralized work. So there's a trade-off -
>> more autonomic messaging means less north-south messaging.
>>
>> Best regards
>>    Brian
>>
>>>
>>> regards,
>>> John
>>>
>>> On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <
>>> brian.e.carpenter@gmail.com> wrote:
>>>
>>>>> However, in ANIMA network, perhaps we do not have a centralized par=
sing
>>>> node (controller), which can replace almost all the communications
>> between
>>>> nodes.
>>>>
>>>> Correct, and there is also no presumption that individual nodes have=

>>>> perfect
>>>> knowledge of the topology. Of course the ACP routing protocol will h=
ave
>>>> perfect
>>>> knowledge of the ACP topology, but individual ASAs only know what th=
ey
>> have
>>>> gleaned from discovery and what they have been told by Intent.
>>>>
>>>> It is, I think, a very different world view than the SDN or SUPA or =
NVO4
>>>> or NFV view.
>>>>
>>>> Regards
>>>>    Brian
>>>>
>>>> On 26/03/2016 16:46, Duzongpeng wrote:
>>>>> Hi John:
>>>>>
>>>>> Thanks for your reply.
>>>>>
>>>>> 
>>>>> How is this different than running a script, or invoking an API? Th=
ere
>>>> are plenty of southbound APIs (from a management function to a devic=
e).
>>>> There are very few northbound APIs. That's what the operators that I=

>> have
>>>> talked to want.
>>>>> 
>>>>>
>>>>> I agree that intent belongs to northbound interface in the SDN netw=
ork.
>>>> And ANIMA network perhaps can share the same northbound interface.
>>>>>
>>>>> But, my understanding is that ANIMA network is a little different f=
rom
>>>> SDN network.
>>>>>
>>>>> SDN controller will parse the intent, and transform to detailed
>>>> configurations of every device. It is a good model, and little
>>>> communications between nodes are needed.
>>>>>
>>>>> Some network level parameters can be handled in SDN controller.
>>>>>
>>>>> However, in ANIMA network, perhaps we do not have a centralized par=
sing
>>>> node (controller), which can replace almost all the communications
>> between
>>>> nodes.
>>>>>
>>>>> So when a network operator is operating a ANIMA network, the networ=
k
>>>> level parameters are still needed to be flooded.
>>>>>
>>>>> I mean the ANIMA interface is perhaps something like =E2=80=9Cnorth=
bound + part
>>>> of southbound=E2=80=9D.
>>>>>
>>>>> Best Regards
>>>>> Zongpeng Du
>>>>>
>>>>>
>>>>> From: John Strassner [mailto:strazpdj@gmail.com]
>>>>> Sent: Saturday, March 26, 2016 9:41 AM
>>>>> To: Duzongpeng; John Strassner
>>>>> Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang;
>>>> anima@ietf.org; J=C3=A9ferson Campos Nobre
>>>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>>>> beyond intent
>>>>>
>>>>> Hi Zongpeng,
>>>>>
>>>>> I think we have a misunderstanding here.
>>>>>
>>>>>
>>>>> But I do not support that the intent writer should look every detai=
l in
>>>> the catalog.
>>>>> 
>>>>> Most catalogs that I have seen have very few details. They instead =
tell
>>>> the user what services or products are available.
>>>>> 
>>>>>
>>>>> The intent writer may mainly care about what they want, but how to =
do
>> it
>>>> can be left to the autonomic network to deal with.
>>>>> 
>>>>> I agree completely.
>>>>> 
>>>>>
>>>>> Perhaps several solutions exist, and autonomic management system ch=
oose
>>>> one according to the abilities collected.
>>>>> 
>>>>> This is one of the reasons why I would like to see capabilities (an=
d
>>>> needs) advertised.
>>>>> 
>>>>>
>>>>> ...
>>>>> I mean in current status, some of the Autonomic Functions should be=

>>>> exposed to the network operator, and some of the Autonomic Functions=

>> can be
>>>> hidden to the network operator
>>>>> ...
>>>>> 
>>>>> How is this different than running a script, or invoking an API? Th=
ere
>>>> are plenty of southbound APIs (from a management function to a devic=
e).
>>>> There are very few northbound APIs. That's what the operators that I=

>> have
>>>> talked to want.
>>>>>
>>>>> regards,
>>>>> John
>>>>>
>>>>> On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng >>> > wrote:
>>>>> Hi John
>>>>>
>>>>> Some personal understanding here. If any misunderstanding, please
>>>> correct me. Thanks.
>>>>>
>>>>> 
>>>>> I would suggest a slight modification, to enable users of
>>>>> intent (and even intent writers) to NOT have to have
>>>>> detailed implementation information.
>>>>>
>>>>> What if all of the Autonomic Functions, after they joined
>>>>> the network, advertised both their capabilities as well as
>>>>> their needs? We can put aside the latter (needs) as this
>>>>> is more complex, but if the capabilities are advertised,
>>>>> then the autonomic management system can assemble
>>>>> those needs into a catalog. Then, the intent writer can
>>>>> look at the catalog and write a declarative request to
>>>>> use the functionality of the catalog.
>>>>>
>>>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>>>> behind...).
>>>>> < /jcs2>
>>>>>
>>>>> I agree that autonomic node should have that self-advertisement
>> ability,
>>>> and autonomic management system should have this catalog.
>>>>>
>>>>> But I do not support that the intent writer should look every detai=
l in
>>>> the catalog.
>>>>>
>>>>> The intent writer may mainly care about what they want, but how to =
do
>> it
>>>> can be left to the autonomic network to deal with.
>>>>> Perhaps several solutions exist, and autonomic management system ch=
oose
>>>> one according to the abilities collected.
>>>>>
>>>>> In the ideal, the network operator may communicate with the autonom=
ic
>>>> network using network-level level intent and network-level report.
>>>>> I mean in current status, some of the Autonomic Functions should be=

>>>> exposed to the network operator, and some of the Autonomic Functions=

>> can be
>>>> hidden to the network operator
>>>>> Perhaps in future, when the network nodes become more intelligent, =
the
>>>> hidden part will increase.
>>>>>
>>>>>
>>>>> Best regards
>>>>> Zongpeng Du
>>>>>
>>>>> From: Anima [mailto:anima-bounces@ietf.org> anima-bounces@ietf.org>]
>>>> On Behalf Of John Strassner
>>>>> Sent: Thursday, March 24, 2016 10:41 AM
>>>>> To: Michael Behringer (mbehring); John Strassner
>>>>> Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org>>> anima@ietf.org>; J=C3=A9ferson Campos Nobre
>>>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>>>> beyond intent
>>>>>
>>>>> Hi Michael,
>>>>>
>>>>> thanks for your reply. I do think that we largely agree, and would
>> enjoy
>>>> chatting during or (hopefully before) BA. Please see ..=
=2E
>> Also
>>>> trying to unclutter, so I am eliding points that we agree on.
>>>>>
>>>>>
>>>>> While *CIEs don=E2=80=99t like magic, they do understand the concep=
t of a
>>>> default behaviour which they can influence if needed. Any reasonable=

>>>> network management approach tries to define =E2=80=9Cdefault=E2=80=9D=
 behaviour and
>>>> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefaul=
t=E2=80=9D part very nicely, and
>> *CIEs
>>>> get that.
>>>>> 
>>>>> Agreed!
>>>>> 
>>>>>
>>>>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D dis=
cussion: My view is that we
>> shouldn=E2=80=99t
>>>> try to define exactly *what* intent allows and doesn=E2=80=99t allow=
, since
>> we=E2=80=99ve
>>>> heard from many sides that the boundary is not clear, and it may be
>>>> impossible to have a really clear, usable definition. So my point wa=
s:
>>>> Let=E2=80=99s focus on the distribution model: Intent is flooded to =
all nodes.
>>>> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific =
nodes, etc,
>>>> because that would blur the boundary even worse.
>>>>> 
>>>>> I like flooding intent to all nodes. In FOCALE (v1-3, and remember,=
 v1
>>>> was 2006, almost 10 years ago), we used a message bus and pub-sub. I=
n
>> v2,
>>>> we flooded; in v3, we went back to pub-sub because we wanted to do s=
ome
>>>> optimization and auditing. But I would be happy to try some type of
>>>> controlled flooding again.
>>>>> 
>>>>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>>>> owners define the content of what they want to push in their bit of =
the
>>>> Intent. If we insist of full flooding, we encourage the right behavi=
our.
>>>> And if someone really puts into his Intent =E2=80=9Cnode A: do this;=
 node B: do
>> the
>>>> other; node C: do the third=E2=80=9D it is awkward enough for people=
 to take a
>> step
>>>> back and think whether they=E2=80=99re doing the right thing =E2=98=BA=

>>>>> 
>>>>> I agree in principle with what you said. Plus, see above.
>>>>> < /jcs2>
>>>>> 
>>>>> Do you mean that the intent writer will **know the difference**
>>>>> between the fabric and each autonomic function? I hope not,
>>>>> but perhaps I'm misunderstanding you...
>>>>> < /jcs>
>>>>> Not sure I understand your concern. Yes, in my model there is a sec=
tion
>>>> for the fabric, and sections for the various functions. Seems a natu=
ral
>>>> structure?!? 
>>>>> The intent **writer** would have to know implementation
>>>>> details in order to place everything in a single file. I would
>>>>> rather not require this, and instead, have the intent engine
>>>>> (e.g., compiler and associated logic) do this.
>>>>> < /jcs2>
>>>>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to st=
ructure Intent around
>>>> Autonomic Functions. This is not optimal from a scientific point of
>> view,
>>>> but aligns very well with the way how things are organised. Probably=
 we
>>>> need to discuss that in BA. But refer to my YANG model sent a few da=
ys
>> ago.
>>>>> 
>>>>> I would suggest a slight modification, to enable users of
>>>>> intent (and even intent writers) to NOT have to have
>>>>> detailed implementation information.
>>>>>
>>>>> What if all of the Autonomic Functions, after they joined
>>>>> the network, advertised both their capabilities as well as
>>>>> their needs? We can put aside the latter (needs) as this
>>>>> is more complex, but if the capabilities are advertised,
>>>>> then the autonomic management system can assemble
>>>>> those needs into a catalog. Then, the intent writer can
>>>>> look at the catalog and write a declarative request to
>>>>> use the functionality of the catalog.
>>>>>
>>>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>>>> behind...).
>>>>> < /jcs2>
>>>>>
>>>>>
>>>>> regards,
>>>>> John
>>>>>
>>>>> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <
>>>> mbehring@cisco.com> wrote:
>>>>> John,
>>>>>
>>>>> I think we=E2=80=99re on the same page. Couple of points, top posti=
ng here,
>>>> otherwise it gets too messy.
>>>>>
>>>>> While *CIEs don=E2=80=99t like magic, they do understand the concep=
t of a
>>>> default behaviour which they can influence if needed. Any reasonable=

>>>> network management approach tries to define =E2=80=9Cdefault=E2=80=9D=
 behaviour and
>>>> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefaul=
t=E2=80=9D part very nicely, and
>> *CIEs
>>>> get that.
>>>>>
>>>>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D dis=
cussion: My view is that we
>> shouldn=E2=80=99t
>>>> try to define exactly *what* intent allows and doesn=E2=80=99t allow=
, since
>> we=E2=80=99ve
>>>> heard from many sides that the boundary is not clear, and it may be
>>>> impossible to have a really clear, usable definition. So my point wa=
s:
>>>> Let=E2=80=99s focus on the distribution model: Intent is flooded to =
all nodes.
>>>> Let=E2=80=99s NOT do selective flooding, unicast Intent to specific =
nodes, etc,
>>>> because that would blur the boundary even worse.
>>>>>
>>>>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>>>> owners define the content of what they want to push in their bit of =
the
>>>> Intent. If we insist of full flooding, we encourage the right behavi=
our.
>>>> And if someone really puts into his Intent =E2=80=9Cnode A: do this;=
 node B: do
>> the
>>>> other; node C: do the third=E2=80=9D it is awkward enough for people=
 to take a
>> step
>>>> back and think whether they=E2=80=99re doing the right thing =E2=98=BA=

>>>>>
>>>>> 
>>>>> Do you mean that the intent writer will **know the difference**
>>>>> between the fabric and each autonomic function? I hope not,
>>>>> but perhaps I'm misunderstanding you...
>>>>> 
>>>>>
>>>>> Not sure I understand your concern. Yes, in my model there is a sec=
tion
>>>> for the fabric, and sections for the various functions. Seems a natu=
ral
>>>> structure?!?
>>>>>
>>>>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to st=
ructure Intent around
>>>> Autonomic Functions. This is not optimal from a scientific point of
>> view,
>>>> but aligns very well with the way how things are organised. Probably=
 we
>>>> need to discuss that in BA. But refer to my YANG model sent a few da=
ys
>> ago.
>>>>>
>>>>> We should schedule some discussion time outside the ANIMA meeting
>>>> window...
>>>>>
>>>>> Michael
>>>>>
>>>>>
>>>>> From: John Strassner [mailto:strazpdj@gmail.com>>> strazpdj@gmail.com>]
>>>>> Sent: 22 March 2016 21:32
>>>>> To: Michael Behringer (mbehring) >>> mbehring@cisco.com>>; John Strassner >>> strazpdj@gmail.com>>
>>>>> Cc: Duzongpeng =
>;
>>>> anima@ietf.org; Laurent Ciavaglia <
>>>> laurent.ciavaglia@nokia.com>;
>>>> J=C3=A9ferson Campos Nobre >>> ;
>>>> Sheng Jiang >
>>>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>>>> beyond intent
>>>>>
>>>>> In general, I agree with Michael. I'd like to emphasize and expand =
on a
>>>> couple of key points:
>>>>>
>>>>>
>>>>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=
=9D or =E2=80=9Csteering=E2=80=9D a
>> network,
>>>> at least not initially. There is still traditional config, plus all =
the
>>>> existing and emerging SDN models. My view has always been that ANIMA=

>> sets a
>>>> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means a=
re more specific and
>>>> can override intent. This is described in RFC7575. Main point here:
>> ANIMA
>>>> Intent isn=E2=80=99t the only tool in a network.
>>>>>
>>>>> 
>>>>> Absolutely agree. Note that even in the Congress case I mentioned
>>>>> earlier, it is NOT pushing config, and it is NOT viewed as the
>>>>> "only" policy
>>>>> 
>>>>>
>>>>>
>>>>> Intent could perfectly well cover a high level policy such as =E2=80=
=9Call
>> nodes
>>>> of type x do this; type y does that=E2=80=9D. (Which I think is what=
 you mention
>>>> below.) When we started with Intent, the vision was to keep it very =
high
>>>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=98=
=BA
>>>>>
>>>>> 
>>>>> Agreed. Although to me, this means that we need to consider the typ=
es
>> of
>>>> actors that are using intent. For example, most of my CCIE and JCIE
>> friends
>>>> have no interest in using an intent language for pushing configurati=
on.
>> On
>>>> the other hand, when I talk to operators and application developers,=

>> they
>>>> would love to be able to use a high-level intent and have the intent=

>> engine
>>>> take care of the details for them.
>>>>> 
>>>>>
>>>>> It has become clear that in today=E2=80=99s networking, people will=
 want to
>> push
>>>> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, and=
 I think we=E2=80=99ll just have to
>> acknowledge
>>>> that.
>>>>>
>>>>> 
>>>>> Sorry Michael, I don't follow here. What do you mean
>>>>> by "parameters"?
>>>>> 
>>>>>
>>>>>
>>>>> Personally I think it=E2=80=99ll pan out like this: Intent will hav=
e sections
>>>> per autonomic function. An Intent parser will validate overall
>> correctness
>>>> (signature, etc), then segment the Intent into its bits: One part wi=
ll
>> be
>>>> the autonomic fabric itself, then there will be bits for each autono=
mic
>>>> function.
>>>>>
>>>>> 
>>>>> Do you mean that the intent writer will **know the difference**
>>>>> between the fabric and each autonomic function? I hope not,
>>>>> but perhaps I'm misunderstanding you...
>>>>> 
>>>>>
>>>>>
>>>>> So the Intent parser on a node will grab intent, see there is some
>>>> Intent for autonomic function A, will see whether this is running
>> locally,
>>>> and if so, it=E2=80=99ll just pass that piece of Intent to that func=
tion.
>>>>>
>>>>> 
>>>>> Agreed.
>>>>> 
>>>>>
>>>>>
>>>>> In this way, a function owner defines both the intent itself, and h=
ow
>> it
>>>> is consumed by his/her function.
>>>>>
>>>>> 
>>>>> I'm not sure what you mean by "function owner". (Sorry to be
>>>>> picky, I'm likely under-caffeinated.)
>>>>>
>>>>> The above seemed like an autonomous decision, made by the parser.
>>>>> I don't think that the intent writer should be cognizant of how man=
y or
>>>>> what types of autonomic functions are present.
>>>>> 
>>>>>
>>>>>
>>>>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and peop=
le can do what they
>>>> want, in the worst case even push config, which is clearly NOT what
>> Intent
>>>> is meant to do.
>>>>>
>>>>> 
>>>>> And hence, the "HAL 2000" scenario. :-)
>>>>> 
>>>>>
>>>>>
>>>>> I=E2=80=99d hope that people don=E2=80=99t start writing Intent lik=
e =E2=80=9Cnode 17: here is
>>>> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is your =
CLI=E2=80=9D. But
>> what
>>>> I describe above COULD be abused in this way. RFC7575 says this is N=
OT
>>>> Intent.
>>>>>
>>>>> 
>>>>> Absolutely agree!
>>>>> 
>>>>>
>>>>>  regards,
>>>>> John
>>>>>
>>>>>
>>>>>
>>>>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
>>>> mbehring@cisco.com> wrote:
>>>>> Hi Zongpeng,
>>>>>
>>>>> Yes, this is a recurring question and topic. Let me give my view on=
 two
>>>> levels:
>>>>>
>>>>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=80=
=9D or =E2=80=9Csteering=E2=80=9D a
>> network,
>>>> at least not initially. There is still traditional config, plus all =
the
>>>> existing and emerging SDN models. My view has always been that ANIMA=

>> sets a
>>>> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means a=
re more specific and
>>>> can override intent. This is described in RFC7575. Main point here:
>> ANIMA
>>>> Intent isn=E2=80=99t the only tool in a network.
>>>>>
>>>>> Intent could perfectly well cover a high level policy such as =E2=80=
=9Call
>> nodes
>>>> of type x do this; type y does that=E2=80=9D. (Which I think is what=
 you mention
>>>> below.) When we started with Intent, the vision was to keep it very =
high
>>>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=98=
=BA  It has become clear
>> that
>>>> in today=E2=80=99s networking, people will want to push more =E2=80=9C=
parameters=E2=80=9D than
>>>> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to ack=
nowledge that.
>>>>>
>>>>> Personally I think it=E2=80=99ll pan out like this: Intent will hav=
e sections
>>>> per autonomic function. An Intent parser will validate overall
>> correctness
>>>> (signature, etc), then segment the Intent into its bits: One part wi=
ll
>> be
>>>> the autonomic fabric itself, then there will be bits for each autono=
mic
>>>> function.
>>>>>
>>>>> So the Intent parser on a node will grab intent, see there is some
>>>> Intent for autonomic function A, will see whether this is running
>> locally,
>>>> and if so, it=E2=80=99ll just pass that piece of Intent to that func=
tion.
>>>>>
>>>>> In this way, a function owner defines both the intent itself, and h=
ow
>> it
>>>> is consumed by his/her function.
>>>>>
>>>>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and peop=
le can do what they
>>>> want, in the worst case even push config, which is clearly NOT what
>> Intent
>>>> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start wri=
ting Intent like
>> =E2=80=9Cnode
>>>> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37: h=
ere is your
>>>> CLI=E2=80=9D. But what I describe above COULD be abused in this way.=
 RFC7575
>> says
>>>> this is NOT Intent.
>>>>>
>>>>> In other words: Let=E2=80=99s not be too prescriptive, and hope tha=
t folks will
>>>> do more or less reasonable things.
>>>>>
>>>>> I=E2=80=99ve been meaning to comment on the Intent drafts and possi=
bly
>>>> contribute, I=E2=80=99m just struggling with priorities at the momen=
t (as you
>> can
>>>> see by my relative silence over the past weeks).
>>>>>
>>>>> Michael
>>>>>
>>>>>
>>>>>
>>>>> From: Anima [mailto:anima-bounces@ietf.org> anima-bounces@ietf.org>]
>>>> On Behalf Of Duzongpeng
>>>>> Sent: 16 March 2016 02:38
>>>>> To: anima@ietf.org
>>>>> Cc: Laurent Ciavaglia >>> laurent.ciavaglia@nokia.com>>; J=C3=A9ferson Campos Nobre <
>> jcnobre@inf.ufrgs.br
>>>> >; Sheng Jiang > >>> jiangsheng@huawei.com>>
>>>>> Subject: [Anima] Does Autonomic network need some intervention beyo=
nd
>>>> intent
>>>>>
>>>>> Hello, everyone
>>>>>
>>>>>          I understand that Autonomic network should be able to work=

>> with
>>>> as less configuration as possible. In the ideal case, an Autonomic
>> network
>>>> should be able to work well while only intent is needed from human
>>>> operators.
>>>>>
>>>>>          However, I wonder whether Autonomic network may need some
>>>> intervention from human operators through some low-level policies be=
yond
>>>> intents.
>>>>>
>>>>>          For example, in the
>>>> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, i=
t
>> is
>>>> suggested that the prefix lengths for the CSG, ASG, RSG (different
>> roles in
>>>> IP RAN) can be assigned as an "intent". These configuration paramete=
rs
>> need
>>>> to be distributed in the autonomic domain to influence the detail
>>>> configurations on each autonomic node.
>>>>>
>>>>>          As intent is described as an abstract, declarative, high-l=
evel
>>>> policy used to operate an autonomic domain, such as an enterprise
>> network.
>>>> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9Cp=
refix-management ID=E2=80=9D to some
>>>> other words.
>>>>>
>>>>> Is =E2=80=9CASA parameters" ok here?
>>>>>
>>>>> I think some characteristics for this kind of low-level policies ar=
e
>>>>>
>>>>> 1.       Network level parameters, need to be distributed by GRASP =
and
>>>> ACP
>>>>>
>>>>> 2.       Relatively static compared to intent, perhaps only needed =
to
>> be
>>>> configured once
>>>>>
>>>>> 3.       Unavoidable in current network environment, mostly for
>>>> establishing the network infrastructure
>>>>>
>>>>> 4.       Rarely need coordinations with others parameters, usually =
is
>>>> closed related to one specific autonomic function
>>>>>
>>>>> Welcome for comment. Your suggestions will be appreciated.
>>>>>
>>>>>
>>>>> Best Regards
>>>>> Zongpeng Du
>>>>>
>>>>> _______________________________________________
>>>>> Anima mailing list
>>>>> Anima@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> regards,
>>>>> John
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> regards,
>>>>> John
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> regards,
>>>>> John
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Anima mailing list
>>>>> Anima@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>=20
>=20


From nobody Sun Mar 27 14:47:42 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0725312D5B2 for ; Sun, 27 Mar 2016 14:47:41 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.21
X-Spam-Level: 
X-Spam-Status: No, score=-4.21 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, RCVD_IN_DNSWL_MED=-2.3, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=mitre.onmicrosoft.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LEDvzK7ivHwl for ; Sun, 27 Mar 2016 14:47:37 -0700 (PDT)
Received: from smtpvmsrv1.mitre.org (smtpvmsrv1.mitre.org [192.52.194.136]) by ietfa.amsl.com (Postfix) with ESMTP id 51BF812D582 for ; Sun, 27 Mar 2016 14:47:37 -0700 (PDT)
Received: from smtpvmsrv1.mitre.org (localhost.localdomain [127.0.0.1]) by localhost (Postfix) with SMTP id 8BE7413B1AE; Sun, 27 Mar 2016 17:47:36 -0400 (EDT)
Received: from imshyb02.MITRE.ORG (imshyb02.mitre.org [129.83.29.3]) by smtpvmsrv1.mitre.org (Postfix) with ESMTP id 64A0413B0E6; Sun, 27 Mar 2016 17:47:36 -0400 (EDT)
Received: from imshyb01.MITRE.ORG (129.83.29.2) by imshyb02.MITRE.ORG (129.83.29.3) with Microsoft SMTP Server (TLS) id 15.0.1130.7; Sun, 27 Mar 2016 17:47:35 -0400
Received: from gcc01-CY1-obe.outbound.protection.outlook.com (10.140.19.249) by imshyb01.MITRE.ORG (129.83.29.2) with Microsoft SMTP Server (TLS) id 15.0.1130.7 via Frontend Transport; Sun, 27 Mar 2016 17:47:35 -0400
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=mitre.onmicrosoft.com;  s=selector1-mitre-org; h=From:To:Date:Subject:Message-ID:Content-Type:MIME-Version; bh=Z0zOS+RKcP4c9S3toMGL+rDrzgTRslrTPUqLKdJ1OBM=; b=BQKr0wpWAuk9/5Tq9gPImRB16WrxVBpxfMngsaR3SQc4y8eG674H3wZyib4YXJWq2T7asAvCqIUaEFTYqymT5nVWsVTu5kvnBHFWJ5yRZfGMpOTL6gPk1UrjpBiB66r2UaXeHVupAU+icJsr2VadvrpSVXjUsefbwv+iX/uQzqo=
Received: from CY1PR09MB0922.namprd09.prod.outlook.com (10.163.89.140) by CY1PR09MB0922.namprd09.prod.outlook.com (10.163.89.140) with Microsoft SMTP Server (TLS) id 15.1.447.15; Sun, 27 Mar 2016 21:47:34 +0000
Received: from CY1PR09MB0922.namprd09.prod.outlook.com ([10.163.89.140]) by CY1PR09MB0922.namprd09.prod.outlook.com ([10.163.89.140]) with mapi id 15.01.0447.023; Sun, 27 Mar 2016 21:47:34 +0000
From: "Natale, Bob" 
To: Brian E Carpenter , John Strassner 
Thread-Topic: [Anima] Does Autonomic network need some intervention beyond intent
Thread-Index: AQHRhwCTKQbSxNs2S22ajBGI+Y2RPp9rFhGAgAAINACAAMMngIAAMhuAgACtHwCAAOmqgIAAKd9A
Date: Sun, 27 Mar 2016 21:47:33 +0000
Message-ID: 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>     <56F60CE4.9040901@gmail.com>  <56F6DAA1.8000206@gmail.com>  <56F82FDD.7090405@gmail.com>
In-Reply-To: <56F82FDD.7090405@gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
authentication-results: gmail.com; dkim=none (message not signed) header.d=none;gmail.com; dmarc=none action=none header.from=mitre.org;
x-originating-ip: [192.80.55.86]
x-ms-office365-filtering-correlation-id: fb1d50b6-9111-42ed-cb1a-08d3568967d4
x-microsoft-exchange-diagnostics: 1; CY1PR09MB0922; 5:kaqcx75KWd7eLk9gLR8Mordkn8NNucHEiGCRpdlfztA2hx1vecqaa6BBQoAVDBOn5as4vbr1bqBo2Tc3L2OWM01+CDAd+UCDK25a8bcILRu1VATFqoHGkf27wiLSu3ycyAFZhp6yl9jtxyiyovN8qw==; 24:pZFOWQpy+XzKfQKi5gWi6tyA3lEIWh9MSEtber3DwpitxsRCO38VCLqoaRCObkJQyIpocrRaGJBaCqWYI1n7Dj3g+KdgrIHbom3LI4kEfNs=
x-microsoft-antispam: UriScan:;BCL:0;PCL:0;RULEID:;SRVR:CY1PR09MB0922;
x-microsoft-antispam-prvs: 
x-exchange-antispam-report-test: UriScan:(95692535739014);
x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(601004)(2401047)(8121501046)(5005006)(3002001)(10201501046); SRVR:CY1PR09MB0922; BCL:0; PCL:0; RULEID:; SRVR:CY1PR09MB0922; 
x-forefront-prvs: 089473E5FE
x-forefront-antispam-report: SFV:NSPM; SFS:(10009020)(6009001)(43544003)(377454003)(45074003)(13464003)(24454002)(33656002)(5004730100002)(81166005)(19580395003)(77096005)(19580405001)(15975445007)(5001770100001)(93886004)(10400500002)(5003600100002)(74316001)(5002640100001)(122556002)(66066001)(106116001)(86362001)(5008740100001)(3280700002)(3660700001)(2900100001)(92566002)(76576001)(99286002)(1096002)(3846002)(586003)(102836003)(6116002)(2950100001)(50986999)(54356999)(1220700001)(4326007)(189998001)(2906002)(76176999)(11100500001)(87936001)(579004)(559001); DIR:OUT; SFP:1101; SCL:1; SRVR:CY1PR09MB0922; H:CY1PR09MB0922.namprd09.prod.outlook.com; FPR:; SPF:None; MLV:sfv; LANG:en; 
spamdiagnosticoutput: 1:23
spamdiagnosticmetadata: NSPM
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-originalarrivaltime: 27 Mar 2016 21:47:33.2660 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: c620dc48-1d50-4952-8b39-df4d54d74d82
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CY1PR09MB0922
X-OriginatorOrg: mitre.org
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Sun, 27 Mar 2016 21:47:41 -0000
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==


From nobody Tue Mar 29 01:58:41 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3A45312D582 for ; Tue, 29 Mar 2016 01:58:40 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ox3NE3fnZ3GC for ; Tue, 29 Mar 2016 01:58:38 -0700 (PDT)
Received: from rcdn-iport-8.cisco.com (rcdn-iport-8.cisco.com [173.37.86.79]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id A60DC12D0F5 for ; Tue, 29 Mar 2016 01:58:38 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1496; q=dns/txt; s=iport; t=1459241918; x=1460451518; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=lkiyYaskumcZ6TbPNKffFmzYeRyNa63skZAlmourjJY=; b=X6MT+yWRg+bGWoiZMT5WsJ9SG1GlGfFnpSyC+ia7QO6b17r7DCjV4ct4 5NtMxLkO2kIruM8+pjM7XrII44wv/l0eKyZIj2SmwcVf6gpo0RiIGbzhj 0+2fuu5xmrY4ZFwPjlK4098+U+TfZFOVZhMSDLZmJc8O9QM/sGv2a+Gyc E=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D+AQDaQvpW/4UNJK1dgy5TgQO6cAENg?= =?us-ascii?q?XAhhWwCHIELOBQBAQEBAQEBZCeEQQEBAQMBIxFFBQsCAQgaAiYCAgIwFRACBAE?= =?us-ascii?q?NDYgXCA4DsEuQQwEBAQEBAQEBAQEBAQEBAQEBAQEBAREEfIUhhESHPYJWBZdsA?= =?us-ascii?q?YVxiA6BbY0nhhKIfAEeAQFCg2WIaH4BAQE?=
X-IronPort-AV: E=Sophos;i="5.24,410,1454976000"; d="scan'208";a="85814135"
Received: from alln-core-11.cisco.com ([173.36.13.133]) by rcdn-iport-8.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 29 Mar 2016 08:58:37 +0000
Received: from XCH-RCD-010.cisco.com (xch-rcd-010.cisco.com [173.37.102.20]) by alln-core-11.cisco.com (8.14.5/8.14.5) with ESMTP id u2T8wbpu016003 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Tue, 29 Mar 2016 08:58:37 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-010.cisco.com (173.37.102.20) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Tue, 29 Mar 2016 03:58:37 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Tue, 29 Mar 2016 03:58:37 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , John Strassner , Duzongpeng 
Thread-Topic: Another use case for Intent: "Freeze network enrolment"
Thread-Index: AQHRhgPpzIwvl+WeaEyrb+KXVUZG1J9qK3OAgAES5wCABORlsA==
Date: Tue, 29 Mar 2016 08:58:37 +0000
Message-ID: 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>
In-Reply-To: <56F5DFD6.90000@gmail.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.49.80.32]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Cc: Laurent Ciavaglia , =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 08:58:40 -0000

PiBJIGp1c3QgdGhvdWdodCB0aGF0IGlmIHRoZSBOT0Mgd2FudHMgdG8gc2VuZCBvdXQgc29tZSBz
b3J0IG9mIHVyZ2VudA0KPiBpbnN0cnVjdGlvbiB0byBhbGwgbm9kZXMsIGl0IGlzIGJ5IGl0cyBu
YXR1cmUgYSBwYXJ0aWFsIHVwZGF0ZSBvZiBJbnRlbnQuIE9mIGNvdXJzZQ0KPiBhIHBhcnRpYWwg
dXBkYXRlIGNhbiBiZSBhY2hpZXZlZCBieSBzZW5kaW5nIGV2ZXJ5dGhpbmcuIFRoYXQncyBhIHRy
YWRlLW9mZiBvZg0KPiByZXNvdXJjZSB2cyBjb21wbGV4aXR5LCBJIHN1cHBvc2UuDQoNCkFueXRo
aW5nICpjYW4qIGJlIHBhcnRpYWwuIE15IGNsYWltIGlzIHRoYXQgaWYgSW50ZW50IHJlYWxseSBp
cyBhIGhpZ2ggbGV2ZWwgcG9saWN5LCBpdCBzaG91bGQgYmUgc28gc3RhYmxlIG92ZXIgdGltZSB0
aGF0IHRoZSBvcHRpbWlzYXRpb24gdG8gc3VwcG9ydCBwYXJ0aWFsIHVwZGF0ZXMganVzdCBpc24n
dCByZXF1aXJlZC4gDQoNCkluIG15IHByb3Bvc2FsIG9uICJjb250ZW50IGRpc3RyaWJ1dGlvbiIg
KGh0dHBzOi8vbWFpbGFyY2hpdmUuaWV0Zi5vcmcvYXJjaC9tc2cvYW5pbWEvS05JZ2x4OWFiRE1F
ZWp3RVp1OWhFNG5EQTA4KSBJIHN1Z2dlc3QgYSBjb250ZW50IHR5cGU7IEkgc3VnZ2VzdCB0byBz
dGFydCB3aXRoIG9ubHkgb25lIGNvbnRlbnQgdHlwZSBzdXBwb3J0ZWQgKCJJbnRlbnQgdmVyc2lv
biAwIikgYW5kIGdldCBleHBlcmllbmNlIHdpdGggdGhhdC4gSWYgaXQgdHVybnMgb3V0IHRoYXQg
d2UgbmVlZCB0byBzbGljZSBpdCwgd2UgY291bGQgZGVmaW5lIG5ldyBjb250ZW50IHR5cGVzLCBw
b3RlbnRpYWxseSBtdWNoIG1vcmUgZ3JhbnVsYXIsIHNvIHRoYXQgd2UgY291bGQgbGF0ZXIgc3Vw
cG9ydCBlZmZlY3RpdmVseSBhIHBhcnRpYWwgdXBkYXRlLiBJZiB3ZSBkbyB0aGlzIGpvYiB3ZWxs
LCB0aGUgZGlzdHJpYnV0aW9uIHByb3RvY29sIGRvZXNuJ3QgbmVlZCB0byBjaGFuZ2UgYXQgYWxs
LiANCg0KU28gbXkgcHJvcG9zYWwgaXMgcmVhbGx5OiBMZXQncyBzdGFydCBhcyBzaW1wbGUgYXMg
cG9zc2libGUsIGJ1dCBsZWF2ZSBzcGFjZSB0byBncm93IGlmIHdlIG5lZWQgdG8uIA0KDQpNaWNo
YWVsDQoNCg==


From nobody Tue Mar 29 02:02:25 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E318412D582 for ; Tue, 29 Mar 2016 02:02:23 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EiST9DBh8nJp for ; Tue, 29 Mar 2016 02:02:22 -0700 (PDT)
Received: from rcdn-iport-6.cisco.com (rcdn-iport-6.cisco.com [173.37.86.77]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 620EA12D0F5 for ; Tue, 29 Mar 2016 02:02:22 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1466; q=dns/txt; s=iport; t=1459242142; x=1460451742; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=89NPgzTvXsNqCVxTbql4CHaNCtLWKCvrmfbM6dEnk4k=; b=K720TeGcwgQMaKvBnQY64BtUPYSu4hzIi4dPzIcIJF7t3EfVN+1/rjVT tcIVEJm2ZBIoGnAAF3SoOpLR7tYrmG98nDaEUD5/VFMcAEdphRsg32ksn OY+mDkwe1gbVg2t5H4Li7zeUuKUJMJ9whWowfPR2PETAtWcZoyAY3o7Dr s=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AAAgDQQ/pW/4UNJK1dgy6BUAavH4tRA?= =?us-ascii?q?Q2BcIYNAhyBDTgUAQEBAQEBAWQnhEEBAQEEIxFRBAIBCBEEAQEBAgIjAwICAh8?= =?us-ascii?q?RFAEICAIEARIIiAoDErBgi18NhFgBAQEBAQEBAQEBAQEBAQEBAQEBAQEVfIUhh?= =?us-ascii?q?ESCP4IUgmqCVgWXOzEBjBGBbo8UhzuHUwEeAQFCg2VsAYd7fgEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,410,1454976000"; d="scan'208";a="87597857"
Received: from alln-core-11.cisco.com ([173.36.13.133]) by rcdn-iport-6.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 29 Mar 2016 09:02:21 +0000
Received: from XCH-ALN-010.cisco.com (xch-aln-010.cisco.com [173.36.7.20]) by alln-core-11.cisco.com (8.14.5/8.14.5) with ESMTP id u2T92LCF019742 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Tue, 29 Mar 2016 09:02:21 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-010.cisco.com (173.36.7.20) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Tue, 29 Mar 2016 04:02:20 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Tue, 29 Mar 2016 04:02:20 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , "anima@ietf.org" 
Thread-Topic: 7.2 Specific ASAs [was Reference Model Issues]
Thread-Index: AQHRhvqCS/S+spVEn06mMRqFfyyXgZ9wJKfQ
Date: Tue, 29 Mar 2016 09:02:20 +0000
Message-ID: <405bab40b6694cc5893f2cd89ef8e913@XCH-RCD-006.cisco.com>
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com> <56F5DE90.6010509@gmail.com>
In-Reply-To: <56F5DE90.6010509@gmail.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.49.80.32]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Subject: Re: [Anima] 7.2 Specific ASAs [was Reference Model Issues]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 09:02:24 -0000
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From nobody Tue Mar 29 02:08:45 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C8B8112D5A2 for ; Tue, 29 Mar 2016 02:08:43 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xvS1pFms60l0 for ; Tue, 29 Mar 2016 02:08:37 -0700 (PDT)
Received: from alln-iport-1.cisco.com (alln-iport-1.cisco.com [173.37.142.88]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8D11012D5A4 for ; Tue, 29 Mar 2016 02:08:37 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1960; q=dns/txt; s=iport; t=1459242517; x=1460452117; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=xdjVkLHO42RmKXS8eqgJOqJ8Phnk/tYm0IIpAbng5Pw=; b=bM37DhI6hYhYTc3qCcBBbiOsbFpzH8PIrZgYeaM3cR4to6DvQmW0MBiM ADruC5uuvVQWHavLZRba7R0/cm9OJqzNIViA0aVLxB/dN+1vmaHMd9isO ATR0oYVGJhvkGlKPD31u79GfdZLuqCT8KAffTh+q/2xA7FVHDrpvi0F0p Q=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AAAgAMRfpW/5pdJa1dgy6BUAavH4tRA?= =?us-ascii?q?Q2BcIYNAhyBDTgUAQEBAQEBAWQnhEEBAQEEIwQNUQQCAQgRBAEBAQICIwMCAgI?= =?us-ascii?q?fERQBCAgCBAESCIgKAxKwZYtgDYRYAQEBAQEBAQEBAQEBAQEBAQEBAQEBFXyFI?= =?us-ascii?q?YREgj+CFIJqglYBBJc7MQGMEYFujxSHO4dTAR4BAUKCAxmBSWwBh3t+AQEB?=
X-IronPort-AV: E=Sophos;i="5.24,410,1454976000"; d="scan'208";a="254846426"
Received: from rcdn-core-3.cisco.com ([173.37.93.154]) by alln-iport-1.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 29 Mar 2016 09:08:36 +0000
Received: from XCH-ALN-008.cisco.com (xch-aln-008.cisco.com [173.36.7.18]) by rcdn-core-3.cisco.com (8.14.5/8.14.5) with ESMTP id u2T98a0K020954 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Tue, 29 Mar 2016 09:08:36 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-008.cisco.com (173.36.7.18) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Tue, 29 Mar 2016 04:08:35 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Tue, 29 Mar 2016 04:08:35 -0500
From: "Michael Behringer (mbehring)" 
To: Brian E Carpenter , "anima@ietf.org" 
Thread-Topic: 5. Functional Overview [was Reference Model Issues]
Thread-Index: AQHRhvmT1+gSqj1wnkekKFiNCJlig59wJepw
Date: Tue, 29 Mar 2016 09:08:35 +0000
Message-ID: 
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com> <56F5DD00.9090809@gmail.com>
In-Reply-To: <56F5DD00.9090809@gmail.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.49.80.32]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Subject: Re: [Anima] 5. Functional Overview [was Reference Model Issues]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 09:08:44 -0000
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From nobody Tue Mar 29 08:39:19 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F36AE12D8CF for ; Tue, 29 Mar 2016 08:39:17 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.702
X-Spam-Level: 
X-Spam-Status: No, score=-2.702 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6IPplZd1gWLS for ; Tue, 29 Mar 2016 08:39:16 -0700 (PDT)
Received: from mailb2.tigertech.net (mailb2.tigertech.net [208.80.4.154]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1A77C12D89A for ; Tue, 29 Mar 2016 08:38:58 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by mailb2.tigertech.net (Postfix) with ESMTP id D9D4A320064; Tue, 29 Mar 2016 08:38:57 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1459265937; bh=X0uDd02JW2UtH1C3faXPGMmt5p90TOM3nv+6XwQGt4A=; h=Subject:To:References:Cc:From:Date:In-Reply-To:From; b=GV3Jn81BDT7ETSGl0ksBOZYG8N5Cc8a6hfK2Ej99F1pjGGxXpkDAS13zlB3g4VLYW WRVXTm2+LMcgHIXA/CJauPZnllrT0r1nv8PssdKd/UFtFgo5a7yoKvMov2z2kysCXO PYhleKo0AOu/7JlJYhnNBn0Ns0smWU83kkp8nfIE=
X-Virus-Scanned: Debian amavisd-new at b2.tigertech.net
Received: from Joels-MacBook-Pro.local (209-255-163-147.ip.mcleodusa.net [209.255.163.147]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mailb2.tigertech.net (Postfix) with ESMTPSA id C5588320051; Tue, 29 Mar 2016 08:38:56 -0700 (PDT)
To: "Michael Behringer (mbehring)" , Brian E Carpenter , John Strassner , Duzongpeng 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com> 
From: "Joel M. Halpern" 
Message-ID: <56FAA188.2050901@joelhalpern.com>
Date: Tue, 29 Mar 2016 11:38:48 -0400
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.1
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 15:39:18 -0000

The notion of distributing all intents as a single blob seems like a 
really bad model.  Policy always decomposes into pieces.  Some of those 
pieces are related.  I think that assuming that the total policy 
(intent) is small is asking for trouble.

This is also why I worry a bit about flooding to everything.

Yours,
Joel

On 3/29/16 4:58 AM, Michael Behringer (mbehring) wrote:
>> I just thought that if the NOC wants to send out some sort of urgent
>> instruction to all nodes, it is by its nature a partial update of Intent. Of course
>> a partial update can be achieved by sending everything. That's a trade-off of
>> resource vs complexity, I suppose.
>
> Anything *can* be partial. My claim is that if Intent really is a high level policy, it should be so stable over time that the optimisation to support partial updates just isn't required.
>
> In my proposal on "content distribution" (https://mailarchive.ietf.org/arch/msg/anima/KNIglx9abDMEejwEZu9hE4nDA08) I suggest a content type; I suggest to start with only one content type supported ("Intent version 0") and get experience with that. If it turns out that we need to slice it, we could define new content types, potentially much more granular, so that we could later support effectively a partial update. If we do this job well, the distribution protocol doesn't need to change at all.
>
> So my proposal is really: Let's start as simple as possible, but leave space to grow if we need to.
>
> Michael
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>


From nobody Tue Mar 29 08:51:05 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 048C312D963 for ; Tue, 29 Mar 2016 08:51:04 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fs1QHF7H6CuG for ; Tue, 29 Mar 2016 08:51:02 -0700 (PDT)
Received: from rcdn-iport-4.cisco.com (rcdn-iport-4.cisco.com [173.37.86.75]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3B38412D97F for ; Tue, 29 Mar 2016 08:51:01 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=2843; q=dns/txt; s=iport; t=1459266661; x=1460476261; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=lVQFMv2LzRrTSHd1smJY3ePk/hgtA/HP4ue1CE67qko=; b=SO7sLR/ZicxrvbHewtSlEPW1XvHzq8tIijSLEvw5ovxrndeGYNuvZh6h rF5HdU4OsTjMwBZNjLOL617MJiKKTWTFe80xAzALnMhvR4VpNicm4vw8l QVotNsmQyPvmGS9z53/sDkDqKEAY0ON3+JUiLI4CB0p5sonsyHOLt2NJ/ s=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0BSBQAApPpW/49dJa1dgy5TfQavIYtfg?= =?us-ascii?q?XAXCoVsAoExOhIBAQEBAQEBZCeEQQEBAQMBAQEBawsFBwQCAQgRBAEBAScHIQY?= =?us-ascii?q?LFAkIAgQBDQUIiAoDCggOu18NhGUBAQEBAQEBAQEBAQEBAQEBAQEBAQERBIYdh?= =?us-ascii?q?ESCP4dUBY18hG2EUjEBhXGGIIFugW2NJ4YSgSmHUwEnATqDZWwBh3t+AQEB?=
X-IronPort-AV: E=Sophos;i="5.24,411,1454976000"; d="scan'208";a="87747055"
Received: from rcdn-core-7.cisco.com ([173.37.93.143]) by rcdn-iport-4.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 29 Mar 2016 15:51:00 +0000
Received: from XCH-ALN-006.cisco.com (xch-aln-006.cisco.com [173.36.7.16]) by rcdn-core-7.cisco.com (8.14.5/8.14.5) with ESMTP id u2TFp0As012680 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Tue, 29 Mar 2016 15:51:00 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-006.cisco.com (173.36.7.16) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Tue, 29 Mar 2016 10:50:59 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Tue, 29 Mar 2016 10:50:59 -0500
From: "Michael Behringer (mbehring)" 
To: "Joel M. Halpern" , Brian E Carpenter , John Strassner , Duzongpeng 
Thread-Topic: [Anima] Another use case for Intent: "Freeze network enrolment"
Thread-Index: AQHRhgPpzIwvl+WeaEyrb+KXVUZG1J9qK3OAgAES5wCABORlsIAAxzYA//+uO0A=
Date: Tue, 29 Mar 2016 15:50:59 +0000
Message-ID: <9c96c405709e4920bcb0e9c7483d1f90@XCH-RCD-006.cisco.com>
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FAA188.2050901@joelhalpern.com>
In-Reply-To: <56FAA188.2050901@joelhalpern.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Cc: Laurent Ciavaglia , =?iso-8859-1?Q?J=E9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 15:51:04 -0000

Thanks Joel, I think we agree that *if* Intent can become big, and / or *if=
* it changes frequently, we need to divide and conquer.=20

If however Intent is "high level", it should *not* change frequently. And I=
 think we probably also agree that if an update is required say once a mont=
h, we *can* flood, and deal with it in one blob. (right?)=20

So the real question is: What will Intent look like? Possibly we should go =
back to the use cases, and see what Intent for such a use case could look l=
ike.=20

Michael


> -----Original Message-----
> From: Joel M. Halpern [mailto:jmh@joelhalpern.com]
> Sent: 29 March 2016 17:39
> To: Michael Behringer (mbehring) ; Brian E Carpenter
> ; John Strassner ;
> Duzongpeng 
> Cc: Laurent Ciavaglia ; J=E9ferson Campos No=
bre
> ; anima@ietf.org; Sheng Jiang
> 
> Subject: Re: [Anima] Another use case for Intent: "Freeze network
> enrolment"
>=20
> The notion of distributing all intents as a single blob seems like a real=
ly bad
> model.  Policy always decomposes into pieces.  Some of those pieces are
> related.  I think that assuming that the total policy
> (intent) is small is asking for trouble.
>=20
> This is also why I worry a bit about flooding to everything.
>=20
> Yours,
> Joel
>=20
> On 3/29/16 4:58 AM, Michael Behringer (mbehring) wrote:
> >> I just thought that if the NOC wants to send out some sort of urgent
> >> instruction to all nodes, it is by its nature a partial update of
> >> Intent. Of course a partial update can be achieved by sending
> >> everything. That's a trade-off of resource vs complexity, I suppose.
> >
> > Anything *can* be partial. My claim is that if Intent really is a high =
level
> policy, it should be so stable over time that the optimisation to support
> partial updates just isn't required.
> >
> > In my proposal on "content distribution"
> (https://mailarchive.ietf.org/arch/msg/anima/KNIglx9abDMEejwEZu9hE4nD
> A08) I suggest a content type; I suggest to start with only one content t=
ype
> supported ("Intent version 0") and get experience with that. If it turns =
out
> that we need to slice it, we could define new content types, potentially =
much
> more granular, so that we could later support effectively a partial updat=
e. If
> we do this job well, the distribution protocol doesn't need to change at =
all.
> >
> > So my proposal is really: Let's start as simple as possible, but leave =
space to
> grow if we need to.
> >
> > Michael
> >
> > _______________________________________________
> > Anima mailing list
> > Anima@ietf.org
> > https://www.ietf.org/mailman/listinfo/anima
> >


From nobody Tue Mar 29 08:52:56 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 49F6B12D9B3 for ; Tue, 29 Mar 2016 08:52:55 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qSIuQ4PgX3ON for ; Tue, 29 Mar 2016 08:52:51 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8E0C912D99D for ; Tue, 29 Mar 2016 08:52:32 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id ADBE997C7D8CC; Tue, 29 Mar 2016 15:52:27 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2TFqStR027934 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Tue, 29 Mar 2016 15:52:28 GMT
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2TFqGbY000876 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 29 Mar 2016 17:52:27 +0200
Received: from [172.27.205.120] (135.239.27.40) by FR711WXCHHUB02.zeu.alcatel-lucent.com (135.239.2.112) with Microsoft SMTP Server (TLS) id 14.3.195.1; Tue, 29 Mar 2016 17:52:24 +0200
To: "EXT Michael Behringer (mbehring)" , Brian E Carpenter , John Strassner , Duzongpeng 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56FAA4B7.9010305@alcatel-lucent.com>
Date: Tue, 29 Mar 2016 17:52:23 +0200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com>
Content-Type: multipart/alternative; boundary="------------020904010508060403010503"
X-Originating-IP: [135.239.27.40]
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 15:52:55 -0000

--------------020904010508060403010503
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit

Michael,

It seems to me that "Freeze network enrollment" is a valid declarative 
form of policy (intent), yet it seems to be a counter-example to the 
argument of long-lasting intent, i.e. the operator will want to 
roll-back to "normal" behavior as soon as the issue that raised the 
"freeze" intent creation is cleared...
(I am not saying that the case is not a good one to investigate).

Am I over-/mis-interpreting?

Best regards, Laurent.


On 24/03/2016 14:49, EXT Michael Behringer (mbehring) wrote:
> Maybe we should combine two threads here. I think there is a perfect use case for Intent I  mentioned in another mail, which would be even applicable to what we do today:
>
> "Freeze network enrolment".
>
> Since any node can be a proxy node for bootstrap, it makes perfect sense to flood this to all nodes.
>
> Local override possible via traditional config / netconf / etc.
>
> No parameters needed. Can't get any simpler.
>
> Michael
>
>
>> -----Original Message-----
>> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
>> Sent: 24 March 2016 05:53
>> To: John Strassner ; Duzongpeng
>> 
>> Cc: Laurent Ciavaglia ; Jéferson Campos Nobre
>> ; anima@ietf.org; Michael Behringer (mbehring)
>> ; Sheng Jiang 
>> Subject: Re: [Anima] Does Autonomic network need some intervention
>> beyond intent
>>
>> On 24/03/2016 16:49, John Strassner wrote:
>>> I am confused as to why we would consider prefix-management as an
>>> autonomic intent.
>> It's the other way round. Prefix management was chosen as an autonomic
>> use case, so a follow-up question is: what intent is appropriate for that use
>> case?
>>
>> And the suggested answer is: the ISP's intent is that certain types of edge
>> network should get certain sizes of prefix. How the prefixes are actually
>> divided up and assigned is left to the autonomic nodes.
>>
>>> To me, this breaks a fundamental rule of what is intent by requiring
>>> the intent writer to understand too many details about the target of
>>> the intent. The analogy (at least in the ONF and TMF) was specifying
>>> to a taxi company that you want a taxi, but you need to take a certain
>>> route, because you (the customer) know traffic patterns better than
>>> the taxi driver.
>>>
>>> Surely we could pick a more appropriate example?
>> By all means describe other use cases and what sort of intent they need.
>>
>>      Brian
>>>
>>> regards,
>>> John
>>>
>>> On Wed, Mar 16, 2016 at 5:41 AM, Duzongpeng
>>  wrote:
>>>> Hi, Michael,
>>>>
>>>>
>>>>
>>>>           Thanks for the reply. I can understand that intent is not in
>>>> the current milestones, and you are working on several milestones drafts
>> now.
>>>>
>>>>
>>>>           You are welcome to contribute to the intent draft at any
>>>> time, and thanks in advance.
>>>>
>>>>
>>>>
>>>>           According to your statements, in my understanding, we can
>>>> keep regarding the “intent†example in “prefix-management ID†as an
>>>> Autonomic intent.
>>>>
>>>>           It’s semantics are just like “all nodes of type x do this;
>>>> type y does thatâ€, so it is a high level policy.
>>>>
>>>>
>>>>
>>>>           It really helps a lot. Thanks.
>>>>
>>>>
>>>>
>>>>           I generally agree with your statement, while some other
>>>> personal understandings exist (shown below). But perhaps they can be
>> discussed later.
>>>>
>>>>
>>>> Best
>>>> Regards
>>>>
>>>> Zongpeng Du
>>>>
>>>>
>>>>
>>>> *From:* Michael Behringer (mbehring) [mailto:mbehring@cisco.com]
>>>> *Sent:* Wednesday, March 16, 2016 4:17 PM
>>>> *To:* Duzongpeng; anima@ietf.org
>>>> *Cc:* Laurent Ciavaglia; Jéferson Campos Nobre; Sheng Jiang
>>>> *Subject:* RE: [Anima] Does Autonomic network need some intervention
>>>> beyond intent
>>>>
>>>>
>>>>
>>>> Hi Zongpeng,
>>>>
>>>>
>>>>
>>>> Yes, this is a recurring question and topic. Let me give my view on
>>>> two
>>>> levels:
>>>>
>>>>
>>>>
>>>>
>>>> First, ANIMA isn’t the only thing “influencing†or “steering†a
>>>> network, at least not initially. There is still traditional config,
>>>> plus all the existing and emerging SDN models. My view has always
>>>> been that ANIMA sets a “default†policy in a network, and all other
>>>> means are more specific and can override intent. This is described in
>>>> RFC7575. Main point here: ANIMA Intent isn’t the only tool in a network.
>>>>
>>>>
>>>>
>>>> [duzongpeng] I agree that intent can coexist with traditional config.
>>>>
>>>> I generally agree with the policy priorities, but I have some other
>>>> concerns.
>>>>
>>>> Some default policies/parameters set through “intent†perhaps are
>>>> fundamental for the uplayer services.
>>>>
>>>> It should be prevented that human operator uses some CLIs to change
>>>> them by mistake, and the uplayer services are interrupted.
>>>>
>>>> At least, some warning should be reported to the human operator.
>>>> [duzongpeng]
>>>>
>>>>
>>>>
>>>> Intent could perfectly well cover a high level policy such as “all
>>>> nodes of type x do this; type y does thatâ€. (Which I think is what
>>>> you mention
>>>> below.) When we started with Intent, the vision was to keep it very
>>>> high level, more along the line “do the right thing†J  It has become
>>>> clear that in today’s networking, people will want to push more
>>>> “parameters†than “policyâ€, and I think we’ll just have to acknowledge
>> that.
>>>>
>>>>
>>>> Personally I think it’ll pan out like this: Intent will have sections
>>>> per autonomic function. An Intent parser will validate overall
>>>> correctness (signature, etc), then segment the Intent into its bits:
>>>> One part will be the autonomic fabric itself, then there will be bits
>>>> for each autonomic function.
>>>>
>>>>
>>>>
>>>> So the Intent parser on a node will grab intent, see there is some
>>>> Intent for autonomic function A, will see whether this is running
>>>> locally, and if so, it’ll just pass that piece of Intent to that function.
>>>>
>>>>
>>>>
>>>> In this way, a function owner defines both the intent itself, and how
>>>> it is consumed by his/her function.
>>>>
>>>>
>>>>
>>>> [duzongpeng]Agree.[duzongpeng]
>>>>
>>>>
>>>>
>>>> The downside is that there is no “control†an//d people can do what
>>>> they want, in the worst case even push config, which is clearly NOT
>>>> what Intent is meant to do. I’d hope that people don’t start writing
>>>> Intent like “node
>>>> 17: here is your CLI; node 23: here is your CLIâ€; node 37: here is
>>>> your CLIâ€. But what I describe above COULD be abused in this way.
>>>> RFC7575 says this is NOT Intent.
>>>>
>>>>
>>>>
>>>> [duzongpeng]Agree. It is hard to give an exact boundary.
>>>>
>>>> However, I think we have two choices here.
>>>>
>>>> The first one is to define a policy first in every ASA, but with some
>>>> open parameters that can be configured. What we need to do is just to
>>>> distribute the network-level parameters for that function. This way
>>>> is suitable to some relatively static functions.
>>>>
>>>> The second is more advanced, in the ideal case, the human operator
>>>> can communicate with network using natural language or something
>>>> similar. This way is more flexible, but harder to realize.
>>>>
>>>> [duzongpeng]
>>>>
>>>>
>>>>
>>>> In other words: Let’s not be too prescriptive, and hope that folks
>>>> will do more or less reasonable things.
>>>>
>>>>
>>>>
>>>> [duzongpeng]Agree.[duzongpeng]
>>>>
>>>>
>>>>
>>>> I’ve been meaning to comment on the Intent drafts and possibly
>>>> contribute, I’m just struggling with priorities at the moment (as you
>>>> can see by my relative silence over the past weeks).
>>>>
>>>>
>>>>
>>>> Michael
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> *From:* Anima [mailto:anima-bounces@ietf.org
>>>> ] *On Behalf Of *Duzongpeng
>>>> *Sent:* 16 March 2016 02:38
>>>> *To:* anima@ietf.org
>>>> *Cc:* Laurent Ciavaglia ; Jéferson
>>>> Campos Nobre ; Sheng Jiang
>>>> 
>>>> *Subject:* [Anima] Does Autonomic network need some intervention
>>>> beyond intent
>>>>
>>>>
>>>>
>>>> Hello, everyone
>>>>
>>>>
>>>>
>>>>           I understand that Autonomic network should be able to work
>>>> with as less configuration as possible. In the ideal case, an
>>>> Autonomic network should be able to work well while only intent is
>>>> needed from human operators.
>>>>
>>>>
>>>>
>>>>           However, I wonder whether Autonomic network may need some
>>>> intervention from human operators through some low-level policies
>>>> beyond intents.
>>>>
>>>>
>>>>
>>>>           For example, in the
>>>> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it
>>>> is suggested that the prefix lengths for the CSG, ASG, RSG (different
>>>> roles in IP RAN) can be assigned as an "intent". These configuration
>>>> parameters need to be distributed in the autonomic domain to
>>>> influence the detail configurations on each autonomic node.
>>>>
>>>>
>>>>
>>>>           As intent is described as an abstract, declarative,
>>>> high-level policy used to operate an autonomic domain, such as an
>> enterprise network.
>>>> Perhaps we can rename the “intent†in the “prefix-management ID†to
>>>> some other words.
>>>>
>>>>
>>>>
>>>> Is “ASA parameters" ok here?
>>>>
>>>>
>>>>
>>>> I think some characteristics for this kind of low-level policies are
>>>>
>>>> 1.       Network level parameters, need to be distributed by GRASP and
>> ACP
>>>> 2.       Relatively static compared to intent, perhaps only needed to be
>>>> configured once
>>>>
>>>> 3.       Unavoidable in current network environment, mostly for
>>>> establishing the network infrastructure
>>>>
>>>> 4.       Rarely need coordinations with others parameters, usually is
>>>> closed related to one specific autonomic function
>>>>
>>>>
>>>>
>>>> Welcome for comment. Your suggestions will be appreciated.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Best Regards
>>>>
>>>> Zongpeng Du
>>>>
>>>> _______________________________________________
>>>> Anima mailing list
>>>> Anima@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>
>>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Anima mailing list
>>> Anima@ietf.org
>>> https://www.ietf.org/mailman/listinfo/anima
>>>

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------020904010508060403010503
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Michael,

      

      It seems to me that "Freeze network enrollment" is a valid
      declarative form of policy (intent), yet it seems to be a
      counter-example to the argument of long-lasting intent, i.e. the
      operator will want to roll-back to "normal" behavior as soon as
      the issue that raised the "freeze" intent creation is cleared...

      (I am not saying that the case is not a good one to investigate).

      

      Am I over-/mis-interpreting?

      

      Best regards, Laurent.

      

    

    
On 24/03/2016 14:49, EXT Michael
      Behringer (mbehring) wrote:

    

    

      
Maybe we should combine two threads here. I think there is a perfect use case for Intent I  mentioned in another mail, which would be even applicable to what we do today: 

"Freeze network enrolment". 

Since any node can be a proxy node for bootstrap, it makes perfect sense to flood this to all nodes. 

Local override possible via traditional config / netconf / etc. 

No parameters needed. Can't get any simpler. 

Michael




      

        
-----Original Message-----
From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
Sent: 24 March 2016 05:53
To: John Strassner <strazpdj@gmail.com>; Duzongpeng
<duzongpeng@huawei.com>
Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; Jéferson Campos Nobre
<jcnobre@inf.ufrgs.br>; anima@ietf.org; Michael Behringer (mbehring)
<mbehring@cisco.com>; Sheng Jiang <jiangsheng@huawei.com>
Subject: Re: [Anima] Does Autonomic network need some intervention
beyond intent

On 24/03/2016 16:49, John Strassner wrote:


        

          
I am confused as to why we would consider prefix-management as an
autonomic intent.


        

        
It's the other way round. Prefix management was chosen as an autonomic
use case, so a follow-up question is: what intent is appropriate for that use
case?

And the suggested answer is: the ISP's intent is that certain types of edge
network should get certain sizes of prefix. How the prefixes are actually
divided up and assigned is left to the autonomic nodes.



        

          
To me, this breaks a fundamental rule of what is intent by requiring
the intent writer to understand too many details about the target of
the intent. The analogy (at least in the ONF and TMF) was specifying
to a taxi company that you want a taxi, but you need to take a certain
route, because you (the customer) know traffic patterns better than
the taxi driver.

Surely we could pick a more appropriate example?


        

        
By all means describe other use cases and what sort of intent they need.

    Brian


        

          

regards,
John

On Wed, Mar 16, 2016 at 5:41 AM, Duzongpeng


        

        
<duzongpeng@huawei.com> wrote:


        

          

          

            
Hi, Michael,



         Thanks for the reply. I can understand that intent is not in
the current milestones, and you are working on several milestones drafts


          

        

        
now.


        

          

            


         You are welcome to contribute to the intent draft at any
time, and thanks in advance.



         According to your statements, in my understanding, we can
keep regarding the “intent†example in “prefix-management ID†as an
Autonomic intent.

         It’s semantics are just like “all nodes of type x do this;
type y does thatâ€, so it is a high level policy.



         It really helps a lot. Thanks.



         I generally agree with your statement, while some other
personal understandings exist (shown below). But perhaps they can be


          

        

        
discussed later.


        

          

            


Best
Regards

Zongpeng Du



*From:* Michael Behringer (mbehring) [mailto:mbehring@cisco.com]
*Sent:* Wednesday, March 16, 2016 4:17 PM
*To:* Duzongpeng; anima@ietf.org
*Cc:* Laurent Ciavaglia; Jéferson Campos Nobre; Sheng Jiang
*Subject:* RE: [Anima] Does Autonomic network need some intervention
beyond intent



Hi Zongpeng,



Yes, this is a recurring question and topic. Let me give my view on
two
levels:




First, ANIMA isn’t the only thing “influencing†or “steering†a
network, at least not initially. There is still traditional config,
plus all the existing and emerging SDN models. My view has always
been that ANIMA sets a “default†policy in a network, and all other
means are more specific and can override intent. This is described in
RFC7575. Main point here: ANIMA Intent isn’t the only tool in a network.



[duzongpeng] I agree that intent can coexist with traditional config.

I generally agree with the policy priorities, but I have some other
concerns.

Some default policies/parameters set through “intent†perhaps are
fundamental for the uplayer services.

It should be prevented that human operator uses some CLIs to change
them by mistake, and the uplayer services are interrupted.

At least, some warning should be reported to the human operator.
[duzongpeng]



Intent could perfectly well cover a high level policy such as “all
nodes of type x do this; type y does thatâ€. (Which I think is what
you mention
below.) When we started with Intent, the vision was to keep it very
high level, more along the line “do the right thing†J  It has become
clear that in today’s networking, people will want to push more
“parameters†than “policyâ€, and I think we’ll just have to acknowledge


          

        

        
that.


        

          

            


Personally I think it’ll pan out like this: Intent will have sections
per autonomic function. An Intent parser will validate overall
correctness (signature, etc), then segment the Intent into its bits:
One part will be the autonomic fabric itself, then there will be bits
for each autonomic function.



So the Intent parser on a node will grab intent, see there is some
Intent for autonomic function A, will see whether this is running
locally, and if so, it’ll just pass that piece of Intent to that function.



In this way, a function owner defines both the intent itself, and how
it is consumed by his/her function.



[duzongpeng]Agree.[duzongpeng]



The downside is that there is no “control†an//d people can do what
they want, in the worst case even push config, which is clearly NOT
what Intent is meant to do. I’d hope that people don’t start writing
Intent like “node
17: here is your CLI; node 23: here is your CLIâ€; node 37: here is
your CLIâ€. But what I describe above COULD be abused in this way.
RFC7575 says this is NOT Intent.



[duzongpeng]Agree. It is hard to give an exact boundary.

However, I think we have two choices here.

The first one is to define a policy first in every ASA, but with some
open parameters that can be configured. What we need to do is just to
distribute the network-level parameters for that function. This way
is suitable to some relatively static functions.

The second is more advanced, in the ideal case, the human operator
can communicate with network using natural language or something
similar. This way is more flexible, but harder to realize.

[duzongpeng]



In other words: Let’s not be too prescriptive, and hope that folks
will do more or less reasonable things.



[duzongpeng]Agree.[duzongpeng]



I’ve been meaning to comment on the Intent drafts and possibly
contribute, I’m just struggling with priorities at the moment (as you
can see by my relative silence over the past weeks).



Michael







*From:* Anima [mailto:anima-bounces@ietf.org
<anima-bounces@ietf.org>] *On Behalf Of *Duzongpeng
*Sent:* 16 March 2016 02:38
*To:* anima@ietf.org
*Cc:* Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; Jéferson
Campos Nobre <jcnobre@inf.ufrgs.br>; Sheng Jiang
<jiangsheng@huawei.com>
*Subject:* [Anima] Does Autonomic network need some intervention
beyond intent



Hello, everyone



         I understand that Autonomic network should be able to work
with as less configuration as possible. In the ideal case, an
Autonomic network should be able to work well while only intent is
needed from human operators.



         However, I wonder whether Autonomic network may need some
intervention from human operators through some low-level policies
beyond intents.



         For example, in the
https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it
is suggested that the prefix lengths for the CSG, ASG, RSG (different
roles in IP RAN) can be assigned as an "intent". These configuration
parameters need to be distributed in the autonomic domain to
influence the detail configurations on each autonomic node.



         As intent is described as an abstract, declarative,
high-level policy used to operate an autonomic domain, such as an


          

        

        
enterprise network.


        

          

            
Perhaps we can rename the “intent†in the “prefix-management ID†to
some other words.



Is “ASA parameters" ok here?



I think some characteristics for this kind of low-level policies are

1.       Network level parameters, need to be distributed by GRASP and


          

        

        
ACP


        

          

            
2.       Relatively static compared to intent, perhaps only needed to be
configured once

3.       Unavoidable in current network environment, mostly for
establishing the network infrastructure

4.       Rarely need coordinations with others parameters, usually is
closed related to one specific autonomic function



Welcome for comment. Your suggestions will be appreciated.





Best Regards

Zongpeng Du

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima




          

          



_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima



        

      

      

    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------020904010508060403010503--


From nobody Tue Mar 29 09:04:02 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 508AF12D9DD for ; Tue, 29 Mar 2016 09:04:01 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.702
X-Spam-Level: 
X-Spam-Status: No, score=-2.702 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aK9okFiRITzk for ; Tue, 29 Mar 2016 09:03:59 -0700 (PDT)
Received: from mailb2.tigertech.net (mailb2.tigertech.net [208.80.4.154]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CFD6412DA3B for ; Tue, 29 Mar 2016 09:01:42 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by mailb2.tigertech.net (Postfix) with ESMTP id 56DE132005D; Tue, 29 Mar 2016 09:01:42 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1459267302; bh=jaMd5mWLS7nmSzRxuoMO+chCzO2mPdFNSLG8lsaMqJQ=; h=Subject:To:References:Cc:From:Date:In-Reply-To:From; b=m3v9v9Mq3c4vKlfKZ0ezN6NlGLOysXvNisb+iKiCkheSWC85jXkHbFq9jBWFvKJQW TBeNQff3r5xg3Ldz+2Y3+rHzFECLPkhXN47/Cp2/68/CbZd13eqTVe2yICEgPtTgxa +DR0dfnXcPmopcYFYtwVWqU6XTsXICkbiUFSTF84=
X-Virus-Scanned: Debian amavisd-new at b2.tigertech.net
Received: from Joels-MacBook-Pro.local (209-255-163-147.ip.mcleodusa.net [209.255.163.147]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mailb2.tigertech.net (Postfix) with ESMTPSA id BFEAB32006A; Tue, 29 Mar 2016 09:01:41 -0700 (PDT)
To: "Michael Behringer (mbehring)" 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FAA188.2050901@joelhalpern.com> <9c96c405709e4920bcb0e9c7483d1f90@XCH-RCD-006.cisco.com>
From: "Joel M. Halpern" 
Message-ID: <56FAA6DD.30403@joelhalpern.com>
Date: Tue, 29 Mar 2016 12:01:33 -0400
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.1
MIME-Version: 1.0
In-Reply-To: <9c96c405709e4920bcb0e9c7483d1f90@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 8bit
Archived-At: 
Cc: Laurent Ciavaglia , "anima@ietf.org" 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 16:04:01 -0000

I can buy the low change rate.  In part because if we find that flooding 
is awkward, it is easy enough to add a new mechanism later.

But I find the assumption that being high level will somehow make 
intents small to be very unlikely.

Yours,
Joel

On 3/29/16 11:50 AM, Michael Behringer (mbehring) wrote:
> Thanks Joel, I think we agree that *if* Intent can become big, and / or *if* it changes frequently, we need to divide and conquer.
>
> If however Intent is "high level", it should *not* change frequently. And I think we probably also agree that if an update is required say once a month, we *can* flood, and deal with it in one blob. (right?)
>
> So the real question is: What will Intent look like? Possibly we should go back to the use cases, and see what Intent for such a use case could look like.
>
> Michael
>
>
>> -----Original Message-----
>> From: Joel M. Halpern [mailto:jmh@joelhalpern.com]
>> Sent: 29 March 2016 17:39
>> To: Michael Behringer (mbehring) ; Brian E Carpenter
>> ; John Strassner ;
>> Duzongpeng 
>> Cc: Laurent Ciavaglia ; Jéferson Campos Nobre
>> ; anima@ietf.org; Sheng Jiang
>> 
>> Subject: Re: [Anima] Another use case for Intent: "Freeze network
>> enrolment"
>>
>> The notion of distributing all intents as a single blob seems like a really bad
>> model.  Policy always decomposes into pieces.  Some of those pieces are
>> related.  I think that assuming that the total policy
>> (intent) is small is asking for trouble.
>>
>> This is also why I worry a bit about flooding to everything.
>>
>> Yours,
>> Joel
>>
>> On 3/29/16 4:58 AM, Michael Behringer (mbehring) wrote:
>>>> I just thought that if the NOC wants to send out some sort of urgent
>>>> instruction to all nodes, it is by its nature a partial update of
>>>> Intent. Of course a partial update can be achieved by sending
>>>> everything. That's a trade-off of resource vs complexity, I suppose.
>>>
>>> Anything *can* be partial. My claim is that if Intent really is a high level
>> policy, it should be so stable over time that the optimisation to support
>> partial updates just isn't required.
>>>
>>> In my proposal on "content distribution"
>> (https://mailarchive.ietf.org/arch/msg/anima/KNIglx9abDMEejwEZu9hE4nD
>> A08) I suggest a content type; I suggest to start with only one content type
>> supported ("Intent version 0") and get experience with that. If it turns out
>> that we need to slice it, we could define new content types, potentially much
>> more granular, so that we could later support effectively a partial update. If
>> we do this job well, the distribution protocol doesn't need to change at all.
>>>
>>> So my proposal is really: Let's start as simple as possible, but leave space to
>> grow if we need to.
>>>
>>> Michael
>>>
>>> _______________________________________________
>>> Anima mailing list
>>> Anima@ietf.org
>>> https://www.ietf.org/mailman/listinfo/anima
>>>
>


From nobody Tue Mar 29 09:36:34 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 571A912D9CF for ; Tue, 29 Mar 2016 09:36:33 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yVrVtRSdLnJ9 for ; Tue, 29 Mar 2016 09:36:28 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E0B0512DABA for ; Tue, 29 Mar 2016 09:29:56 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id EDACECB9B372E; Tue, 29 Mar 2016 16:29:48 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2TGTp8E017557 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Tue, 29 Mar 2016 16:29:51 GMT
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2TGTpo9021856 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 29 Mar 2016 18:29:51 +0200
Received: from [172.27.205.120] (135.239.27.39) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Tue, 29 Mar 2016 18:29:48 +0200
To: EXT Brian E Carpenter , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>     <56F60CE4.9040901@gmail.com>  <56F6DAA1.8000206@gmail.com>  <56F82FDD.7090405@gmail.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56FAAD7B.6080305@alcatel-lucent.com>
Date: Tue, 29 Mar 2016 18:29:47 +0200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <56F82FDD.7090405@gmail.com>
Content-Type: multipart/alternative; boundary="------------070605040304060404060501"
X-Originating-IP: [135.239.27.39]
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 16:36:33 -0000

--------------070605040304060404060501
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit

Hello,


> Indeed not. An ASA will have some default behaviour that amounts to having
> default settings for intent, but by definition intent comes from humans.
> (One qualification: if the intent that reaches a given node is inconsistent,
> then resolving that inconsistency might be autonomic.)

Brian: could you explain what you mean here, especially "resolving that 
inconsistency might be autonomic"?
I guess you mean that ASAs/nodes can autonomously detect inconsistency 
in intents and (try to) solve this inconsistency (on their own or via 
collaboration)...?

>
>>> Of course, when we put Intent into the picture, there's a presumption
>>> that Intent comes from some authority such as a NOC, but that's an
>>> external semantic, not an assumption of the autonomic model.
>> Well, no - if the autonomic system is going to use intent, or at least
>> respond to it, then it must understand those external semantics.
>> This is even more important if the network is supposed to form
>> according to one or more intent rules.
> Right, but what I meant is that the network will operate perfectly well
> if no intent is sent, by using defaults. That will always happen to the
> extent that intent distribution will take a finite time. Until the intent
> reaches a node, it will operate on defaults.

 From collaborations with some operators on autonomic networking 
projects: default behavior without policy must equal to "do nothing". 
But operators' views might change/evolve...
Also, and this is based on feedback from the field (e.g. 3GPP SON), 
turning on autonomic functions, even with well-thought parameter 
settings, turns out most of the times in chaotic behaviors and severe 
performance degradations.
A key element to consider here is how to "gracefully" start/ramp-up an 
autonomic network and aspects of coordination (cf 
draft-ciavaglia-anima-coordination-01).

>   Of course, the semantics of
> intent must be well-defined, but it's an extra layer.
>
>> If intent is formulated at an abstract level, how is an autonomic
>> network going to consume intent that may have drastically different
>> syntactical structures, and likely, different semantics associated with
>> those syntactical structures?
> Agreed. We need a model, syntax and semantics.
>
>>> I think the point is that AN *adds* a new mode of communication
>>> between self-managing nodes, that doesn't exist in a strictly
>>> north/south scenario. That is meant to reduce the amount of
>>> north/south communication and reduce the amount of
>>> centralized work. So there's a trade-off - more autonomic
>>> messaging means less north-south messaging.

north-south is a matter of convention / roles.
if we consider PBM / intent, for me it implies a "vertical" (authority) 
relationship.
if we consider a more "flat" system of collaborative entities, then 
let's have a look at MAS / DAI approaches and not reinvent (a squarred) 
wheel.
we may end up pushing more or less intelligence in the autonomic 
functions, but we also consider the scope of ANIMA to address "managed" 
networks (maybe not only this one), so it means some operators will 
expect certain behavior/performance from the neworks. Intent is one way 
to inform the network what is expected.

>> There may indeed be a new mode of communication required.
>> However, if intent is not standardized, how does this work?
> It doesn't. My only point is that intent is layered on top of an
> underlying autonomic behaviour based on default behaviours. Actually,
> that's why the question of intent was deferred in the WG charter.

Your explanation is "valid", but I don't remember this was the reason 
used to consider intent FFS.
Without rewriting history, we are facing now with the "heavy" 
design/conceptual discussions on the list, the effect of some decisions 
to exclude architectural analysis / investigations in the early phase of 
ANIMA...

Laurent.


>
>      Brian
>
>> I understand how unstructured and structured P2P networks
>> work. Let's assume this is a structured P2P network. In that case,
>> a DHT is used to assign ownership of each file to a peer. I fail to
>> understand how this works in an intent case. Files are passive,
>> intent is active. Please explain.
>>
>> Regards,
>> John
>>
>> On Sat, Mar 26, 2016 at 11:53 AM, Brian E Carpenter <
>> brian.e.carpenter@gmail.com> wrote:
>>
>>> Hi John,
>>> On 27/03/2016 04:54, John Strassner wrote:
>>>>> Correct, and there is also no presumption that individual nodes
>>>>> have perfect knowledge of the topology. Of course the ACP
>>>>> routing protocol will have perfect knowledge of the ACP topology,
>>>>> but individual ASAs only know what they have gleaned from
>>>>> discovery and what they have been told by Intent.
>>>>>
>>>>> It is, I think, a very different world view than the SDN or SUPA or
>>>>> NVO4 or NFV view.
>>>> Please note that I did NOT say anything about SDN, NFV, or SUPA
>>>> here. In fact, I'm confused as to why you both think I did.
>>> I can't speak for Zongpeng, but I was triggered by the mention of
>>> northbound and southbound APIs. There is no presumed direction in
>>> Anima relationships, since in the limit a network is self-forming;
>>> all relationships are peer-to-peer.
>>>
>>> Of course, when we put Intent into the picture, there's a presumption
>>> that Intent comes from some authority such as a NOC, but that's an
>>> external semantic, not an assumption of the autonomic model.
>>>
>>>> I specifically said "There are plenty of southbound APIs (from a
>>>> management function to a device). There are very few northbound APIs."
>>>>
>>>> Note the word "management function".
>>>>
>>>> For the record, I don't think that either Phase 2 of NFV or the SDN 1.1
>>>> architecture have treated policy in any detail. Both architectures lack a
>>>> policy management component. I hope that ANIMA doesn't make the
>>>> same mistake.
>>> Certainly, Intent needs a number of properties such as self-consistency
>>> and stability that imply it is managed in some way.
>>>
>>>>>> However, in ANIMA network, perhaps we do not have a centralized
>>>>>> parsing node (controller), which can replace almost all the
>>>>>> communications between nodes.
>>>> Earlier I wrote about intent requiring a translation function. I never
>>> said
>>>> that it was centralized or distributed. In the autonomic architectures I
>>>> have built, it was in fact a decentralized (agent-driven) function.
>>>> Furthermore, it did not "replace almost all the communications" - it
>>>> was used for compilation. There was, in fact, a separate distribution
>>>> mechanism (which I have also described), but that was limited to
>>>> "just" distribution.
>>>>
>>>> I really don't understand what "replacing communications" means.
>>> I think the point is that AN *adds* a new mode of communication between
>>> self-managing nodes, that doesn't exist in a strictly north/south
>>> scenario. That is meant to reduce the amount of north/south communication
>>> and reduce the amount of centralized work. So there's a trade-off -
>>> more autonomic messaging means less north-south messaging.
>>>
>>> Best regards
>>>     Brian
>>>
>>>> regards,
>>>> John
>>>>
>>>> On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <
>>>> brian.e.carpenter@gmail.com> wrote:
>>>>
>>>>>> However, in ANIMA network, perhaps we do not have a centralized parsing
>>>>> node (controller), which can replace almost all the communications
>>> between
>>>>> nodes.
>>>>>
>>>>> Correct, and there is also no presumption that individual nodes have
>>>>> perfect
>>>>> knowledge of the topology. Of course the ACP routing protocol will have
>>>>> perfect
>>>>> knowledge of the ACP topology, but individual ASAs only know what they
>>> have
>>>>> gleaned from discovery and what they have been told by Intent.
>>>>>
>>>>> It is, I think, a very different world view than the SDN or SUPA or NVO4
>>>>> or NFV view.
>>>>>
>>>>> Regards
>>>>>     Brian
>>>>>
>>>>> On 26/03/2016 16:46, Duzongpeng wrote:
>>>>>> Hi John:
>>>>>>
>>>>>> Thanks for your reply.
>>>>>>
>>>>>> 
>>>>>> How is this different than running a script, or invoking an API? There
>>>>> are plenty of southbound APIs (from a management function to a device).
>>>>> There are very few northbound APIs. That's what the operators that I
>>> have
>>>>> talked to want.
>>>>>> 
>>>>>>
>>>>>> I agree that intent belongs to northbound interface in the SDN network.
>>>>> And ANIMA network perhaps can share the same northbound interface.
>>>>>> But, my understanding is that ANIMA network is a little different from
>>>>> SDN network.
>>>>>> SDN controller will parse the intent, and transform to detailed
>>>>> configurations of every device. It is a good model, and little
>>>>> communications between nodes are needed.
>>>>>> Some network level parameters can be handled in SDN controller.
>>>>>>
>>>>>> However, in ANIMA network, perhaps we do not have a centralized parsing
>>>>> node (controller), which can replace almost all the communications
>>> between
>>>>> nodes.
>>>>>> So when a network operator is operating a ANIMA network, the network
>>>>> level parameters are still needed to be flooded.
>>>>>> I mean the ANIMA interface is perhaps something like “northbound + part
>>>>> of southboundâ€.
>>>>>> Best Regards
>>>>>> Zongpeng Du
>>>>>>
>>>>>>
>>>>>> From: John Strassner [mailto:strazpdj@gmail.com]
>>>>>> Sent: Saturday, March 26, 2016 9:41 AM
>>>>>> To: Duzongpeng; John Strassner
>>>>>> Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang;
>>>>> anima@ietf.org; Jéferson Campos Nobre
>>>>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>>>>> beyond intent
>>>>>> Hi Zongpeng,
>>>>>>
>>>>>> I think we have a misunderstanding here.
>>>>>>
>>>>>>
>>>>>> But I do not support that the intent writer should look every detail in
>>>>> the catalog.
>>>>>> 
>>>>>> Most catalogs that I have seen have very few details. They instead tell
>>>>> the user what services or products are available.
>>>>>> 
>>>>>>
>>>>>> The intent writer may mainly care about what they want, but how to do
>>> it
>>>>> can be left to the autonomic network to deal with.
>>>>>> 
>>>>>> I agree completely.
>>>>>> 
>>>>>>
>>>>>> Perhaps several solutions exist, and autonomic management system choose
>>>>> one according to the abilities collected.
>>>>>> 
>>>>>> This is one of the reasons why I would like to see capabilities (and
>>>>> needs) advertised.
>>>>>> 
>>>>>>
>>>>>> ...
>>>>>> I mean in current status, some of the Autonomic Functions should be
>>>>> exposed to the network operator, and some of the Autonomic Functions
>>> can be
>>>>> hidden to the network operator
>>>>>> ...
>>>>>> 
>>>>>> How is this different than running a script, or invoking an API? There
>>>>> are plenty of southbound APIs (from a management function to a device).
>>>>> There are very few northbound APIs. That's what the operators that I
>>> have
>>>>> talked to want.
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>> On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng >>>> > wrote:
>>>>>> Hi John
>>>>>>
>>>>>> Some personal understanding here. If any misunderstanding, please
>>>>> correct me. Thanks.
>>>>>> 
>>>>>> I would suggest a slight modification, to enable users of
>>>>>> intent (and even intent writers) to NOT have to have
>>>>>> detailed implementation information.
>>>>>>
>>>>>> What if all of the Autonomic Functions, after they joined
>>>>>> the network, advertised both their capabilities as well as
>>>>>> their needs? We can put aside the latter (needs) as this
>>>>>> is more complex, but if the capabilities are advertised,
>>>>>> then the autonomic management system can assemble
>>>>>> those needs into a catalog. Then, the intent writer can
>>>>>> look at the catalog and write a declarative request to
>>>>>> use the functionality of the catalog.
>>>>>>
>>>>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>>>>> behind...).
>>>>>> < /jcs2>
>>>>>>
>>>>>> I agree that autonomic node should have that self-advertisement
>>> ability,
>>>>> and autonomic management system should have this catalog.
>>>>>> But I do not support that the intent writer should look every detail in
>>>>> the catalog.
>>>>>> The intent writer may mainly care about what they want, but how to do
>>> it
>>>>> can be left to the autonomic network to deal with.
>>>>>> Perhaps several solutions exist, and autonomic management system choose
>>>>> one according to the abilities collected.
>>>>>> In the ideal, the network operator may communicate with the autonomic
>>>>> network using network-level level intent and network-level report.
>>>>>> I mean in current status, some of the Autonomic Functions should be
>>>>> exposed to the network operator, and some of the Autonomic Functions
>>> can be
>>>>> hidden to the network operator
>>>>>> Perhaps in future, when the network nodes become more intelligent, the
>>>>> hidden part will increase.
>>>>>>
>>>>>> Best regards
>>>>>> Zongpeng Du
>>>>>>
>>>>>> From: Anima [mailto:anima-bounces@ietf.org>> anima-bounces@ietf.org>]
>>>>> On Behalf Of John Strassner
>>>>>> Sent: Thursday, March 24, 2016 10:41 AM
>>>>>> To: Michael Behringer (mbehring); John Strassner
>>>>>> Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org>>>> anima@ietf.org>; Jéferson Campos Nobre
>>>>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>>>>> beyond intent
>>>>>> Hi Michael,
>>>>>>
>>>>>> thanks for your reply. I do think that we largely agree, and would
>>> enjoy
>>>>> chatting during or (hopefully before) BA. Please see ...
>>> Also
>>>>> trying to unclutter, so I am eliding points that we agree on.
>>>>>>
>>>>>> While *CIEs don’t like magic, they do understand the concept of a
>>>>> default behaviour which they can influence if needed. Any reasonable
>>>>> network management approach tries to define “default†behaviour and
>>>>> “exceptionsâ€. The AN story fills the “default†part very nicely, and
>>> *CIEs
>>>>> get that.
>>>>>> 
>>>>>> Agreed!
>>>>>> 
>>>>>>
>>>>>> The “parameter†versus “policy†discussion: My view is that we
>>> shouldn’t
>>>>> try to define exactly *what* intent allows and doesn’t allow, since
>>> we’ve
>>>>> heard from many sides that the boundary is not clear, and it may be
>>>>> impossible to have a really clear, usable definition. So my point was:
>>>>> Let’s focus on the distribution model: Intent is flooded to all nodes.
>>>>> Let’s NOT do selective flooding, unicast Intent to specific nodes, etc,
>>>>> because that would blur the boundary even worse.
>>>>>> 
>>>>>> I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1
>>>>> was 2006, almost 10 years ago), we used a message bus and pub-sub. In
>>> v2,
>>>>> we flooded; in v3, we went back to pub-sub because we wanted to do some
>>>>> optimization and auditing. But I would be happy to try some type of
>>>>> controlled flooding again.
>>>>>> 
>>>>>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>>>>> owners define the content of what they want to push in their bit of the
>>>>> Intent. If we insist of full flooding, we encourage the right behaviour.
>>>>> And if someone really puts into his Intent “node A: do this; node B: do
>>> the
>>>>> other; node C: do the third†it is awkward enough for people to take a
>>> step
>>>>> back and think whether they’re doing the right thing ☺
>>>>>> 
>>>>>> I agree in principle with what you said. Plus, see above.
>>>>>> < /jcs2>
>>>>>> 
>>>>>> Do you mean that the intent writer will **know the difference**
>>>>>> between the fabric and each autonomic function? I hope not,
>>>>>> but perhaps I'm misunderstanding you...
>>>>>> < /jcs>
>>>>>> Not sure I understand your concern. Yes, in my model there is a section
>>>>> for the fabric, and sections for the various functions. Seems a natural
>>>>> structure?!? 
>>>>>> The intent **writer** would have to know implementation
>>>>>> details in order to place everything in a single file. I would
>>>>>> rather not require this, and instead, have the intent engine
>>>>>> (e.g., compiler and associated logic) do this.
>>>>>> < /jcs2>
>>>>>> “Function ownerâ€: Yes, I’m suggesting to structure Intent around
>>>>> Autonomic Functions. This is not optimal from a scientific point of
>>> view,
>>>>> but aligns very well with the way how things are organised. Probably we
>>>>> need to discuss that in BA. But refer to my YANG model sent a few days
>>> ago.
>>>>>> 
>>>>>> I would suggest a slight modification, to enable users of
>>>>>> intent (and even intent writers) to NOT have to have
>>>>>> detailed implementation information.
>>>>>>
>>>>>> What if all of the Autonomic Functions, after they joined
>>>>>> the network, advertised both their capabilities as well as
>>>>>> their needs? We can put aside the latter (needs) as this
>>>>>> is more complex, but if the capabilities are advertised,
>>>>>> then the autonomic management system can assemble
>>>>>> those needs into a catalog. Then, the intent writer can
>>>>>> look at the catalog and write a declarative request to
>>>>>> use the functionality of the catalog.
>>>>>>
>>>>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>>>>> behind...).
>>>>>> < /jcs2>
>>>>>>
>>>>>>
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <
>>>>> mbehring@cisco.com> wrote:
>>>>>> John,
>>>>>>
>>>>>> I think we’re on the same page. Couple of points, top posting here,
>>>>> otherwise it gets too messy.
>>>>>> While *CIEs don’t like magic, they do understand the concept of a
>>>>> default behaviour which they can influence if needed. Any reasonable
>>>>> network management approach tries to define “default†behaviour and
>>>>> “exceptionsâ€. The AN story fills the “default†part very nicely, and
>>> *CIEs
>>>>> get that.
>>>>>> The “parameter†versus “policy†discussion: My view is that we
>>> shouldn’t
>>>>> try to define exactly *what* intent allows and doesn’t allow, since
>>> we’ve
>>>>> heard from many sides that the boundary is not clear, and it may be
>>>>> impossible to have a really clear, usable definition. So my point was:
>>>>> Let’s focus on the distribution model: Intent is flooded to all nodes.
>>>>> Let’s NOT do selective flooding, unicast Intent to specific nodes, etc,
>>>>> because that would blur the boundary even worse.
>>>>>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>>>>> owners define the content of what they want to push in their bit of the
>>>>> Intent. If we insist of full flooding, we encourage the right behaviour.
>>>>> And if someone really puts into his Intent “node A: do this; node B: do
>>> the
>>>>> other; node C: do the third†it is awkward enough for people to take a
>>> step
>>>>> back and think whether they’re doing the right thing ☺
>>>>>> 
>>>>>> Do you mean that the intent writer will **know the difference**
>>>>>> between the fabric and each autonomic function? I hope not,
>>>>>> but perhaps I'm misunderstanding you...
>>>>>> 
>>>>>>
>>>>>> Not sure I understand your concern. Yes, in my model there is a section
>>>>> for the fabric, and sections for the various functions. Seems a natural
>>>>> structure?!?
>>>>>> “Function ownerâ€: Yes, I’m suggesting to structure Intent around
>>>>> Autonomic Functions. This is not optimal from a scientific point of
>>> view,
>>>>> but aligns very well with the way how things are organised. Probably we
>>>>> need to discuss that in BA. But refer to my YANG model sent a few days
>>> ago.
>>>>>> We should schedule some discussion time outside the ANIMA meeting
>>>>> window...
>>>>>> Michael
>>>>>>
>>>>>>
>>>>>> From: John Strassner [mailto:strazpdj@gmail.com>>>> strazpdj@gmail.com>]
>>>>>> Sent: 22 March 2016 21:32
>>>>>> To: Michael Behringer (mbehring) >>>> mbehring@cisco.com>>; John Strassner >>>> strazpdj@gmail.com>>
>>>>>> Cc: Duzongpeng >;
>>>>> anima@ietf.org; Laurent Ciavaglia <
>>>>> laurent.ciavaglia@nokia.com>;
>>>>> Jéferson Campos Nobre >>>> ;
>>>>> Sheng Jiang >
>>>>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>>>>> beyond intent
>>>>>> In general, I agree with Michael. I'd like to emphasize and expand on a
>>>>> couple of key points:
>>>>>>
>>>>>> First, ANIMA isn’t the only thing “influencing†or “steering†a
>>> network,
>>>>> at least not initially. There is still traditional config, plus all the
>>>>> existing and emerging SDN models. My view has always been that ANIMA
>>> sets a
>>>>> “default†policy in a network, and all other means are more specific and
>>>>> can override intent. This is described in RFC7575. Main point here:
>>> ANIMA
>>>>> Intent isn’t the only tool in a network.
>>>>>> 
>>>>>> Absolutely agree. Note that even in the Congress case I mentioned
>>>>>> earlier, it is NOT pushing config, and it is NOT viewed as the
>>>>>> "only" policy
>>>>>> 
>>>>>>
>>>>>>
>>>>>> Intent could perfectly well cover a high level policy such as “all
>>> nodes
>>>>> of type x do this; type y does thatâ€. (Which I think is what you mention
>>>>> below.) When we started with Intent, the vision was to keep it very high
>>>>> level, more along the line “do the right thing†☺
>>>>>> 
>>>>>> Agreed. Although to me, this means that we need to consider the types
>>> of
>>>>> actors that are using intent. For example, most of my CCIE and JCIE
>>> friends
>>>>> have no interest in using an intent language for pushing configuration.
>>> On
>>>>> the other hand, when I talk to operators and application developers,
>>> they
>>>>> would love to be able to use a high-level intent and have the intent
>>> engine
>>>>> take care of the details for them.
>>>>>> 
>>>>>>
>>>>>> It has become clear that in today’s networking, people will want to
>>> push
>>>>> more “parameters†than “policyâ€, and I think we’ll just have to
>>> acknowledge
>>>>> that.
>>>>>> 
>>>>>> Sorry Michael, I don't follow here. What do you mean
>>>>>> by "parameters"?
>>>>>> 
>>>>>>
>>>>>>
>>>>>> Personally I think it’ll pan out like this: Intent will have sections
>>>>> per autonomic function. An Intent parser will validate overall
>>> correctness
>>>>> (signature, etc), then segment the Intent into its bits: One part will
>>> be
>>>>> the autonomic fabric itself, then there will be bits for each autonomic
>>>>> function.
>>>>>> 
>>>>>> Do you mean that the intent writer will **know the difference**
>>>>>> between the fabric and each autonomic function? I hope not,
>>>>>> but perhaps I'm misunderstanding you...
>>>>>> 
>>>>>>
>>>>>>
>>>>>> So the Intent parser on a node will grab intent, see there is some
>>>>> Intent for autonomic function A, will see whether this is running
>>> locally,
>>>>> and if so, it’ll just pass that piece of Intent to that function.
>>>>>> 
>>>>>> Agreed.
>>>>>> 
>>>>>>
>>>>>>
>>>>>> In this way, a function owner defines both the intent itself, and how
>>> it
>>>>> is consumed by his/her function.
>>>>>> 
>>>>>> I'm not sure what you mean by "function owner". (Sorry to be
>>>>>> picky, I'm likely under-caffeinated.)
>>>>>>
>>>>>> The above seemed like an autonomous decision, made by the parser.
>>>>>> I don't think that the intent writer should be cognizant of how many or
>>>>>> what types of autonomic functions are present.
>>>>>> 
>>>>>>
>>>>>>
>>>>>> The downside is that there is no “control†and people can do what they
>>>>> want, in the worst case even push config, which is clearly NOT what
>>> Intent
>>>>> is meant to do.
>>>>>> 
>>>>>> And hence, the "HAL 2000" scenario. :-)
>>>>>> 
>>>>>>
>>>>>>
>>>>>> I’d hope that people don’t start writing Intent like “node 17: here is
>>>>> your CLI; node 23: here is your CLIâ€; node 37: here is your CLIâ€. But
>>> what
>>>>> I describe above COULD be abused in this way. RFC7575 says this is NOT
>>>>> Intent.
>>>>>> 
>>>>>> Absolutely agree!
>>>>>> 
>>>>>>
>>>>>>   regards,
>>>>>> John
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
>>>>> mbehring@cisco.com> wrote:
>>>>>> Hi Zongpeng,
>>>>>>
>>>>>> Yes, this is a recurring question and topic. Let me give my view on two
>>>>> levels:
>>>>>> First, ANIMA isn’t the only thing “influencing†or “steering†a
>>> network,
>>>>> at least not initially. There is still traditional config, plus all the
>>>>> existing and emerging SDN models. My view has always been that ANIMA
>>> sets a
>>>>> “default†policy in a network, and all other means are more specific and
>>>>> can override intent. This is described in RFC7575. Main point here:
>>> ANIMA
>>>>> Intent isn’t the only tool in a network.
>>>>>> Intent could perfectly well cover a high level policy such as “all
>>> nodes
>>>>> of type x do this; type y does thatâ€. (Which I think is what you mention
>>>>> below.) When we started with Intent, the vision was to keep it very high
>>>>> level, more along the line “do the right thing†☺  It has become clear
>>> that
>>>>> in today’s networking, people will want to push more “parameters†than
>>>>> “policyâ€, and I think we’ll just have to acknowledge that.
>>>>>> Personally I think it’ll pan out like this: Intent will have sections
>>>>> per autonomic function. An Intent parser will validate overall
>>> correctness
>>>>> (signature, etc), then segment the Intent into its bits: One part will
>>> be
>>>>> the autonomic fabric itself, then there will be bits for each autonomic
>>>>> function.
>>>>>> So the Intent parser on a node will grab intent, see there is some
>>>>> Intent for autonomic function A, will see whether this is running
>>> locally,
>>>>> and if so, it’ll just pass that piece of Intent to that function.
>>>>>> In this way, a function owner defines both the intent itself, and how
>>> it
>>>>> is consumed by his/her function.
>>>>>> The downside is that there is no “control†and people can do what they
>>>>> want, in the worst case even push config, which is clearly NOT what
>>> Intent
>>>>> is meant to do. I’d hope that people don’t start writing Intent like
>>> “node
>>>>> 17: here is your CLI; node 23: here is your CLIâ€; node 37: here is your
>>>>> CLIâ€. But what I describe above COULD be abused in this way. RFC7575
>>> says
>>>>> this is NOT Intent.
>>>>>> In other words: Let’s not be too prescriptive, and hope that folks will
>>>>> do more or less reasonable things.
>>>>>> I’ve been meaning to comment on the Intent drafts and possibly
>>>>> contribute, I’m just struggling with priorities at the moment (as you
>>> can
>>>>> see by my relative silence over the past weeks).
>>>>>> Michael
>>>>>>
>>>>>>
>>>>>>
>>>>>> From: Anima [mailto:anima-bounces@ietf.org>> anima-bounces@ietf.org>]
>>>>> On Behalf Of Duzongpeng
>>>>>> Sent: 16 March 2016 02:38
>>>>>> To: anima@ietf.org
>>>>>> Cc: Laurent Ciavaglia >>>> laurent.ciavaglia@nokia.com>>; Jéferson Campos Nobre <
>>> jcnobre@inf.ufrgs.br
>>>>> >; Sheng Jiang >> >>>> jiangsheng@huawei.com>>
>>>>>> Subject: [Anima] Does Autonomic network need some intervention beyond
>>>>> intent
>>>>>> Hello, everyone
>>>>>>
>>>>>>           I understand that Autonomic network should be able to work
>>> with
>>>>> as less configuration as possible. In the ideal case, an Autonomic
>>> network
>>>>> should be able to work well while only intent is needed from human
>>>>> operators.
>>>>>>           However, I wonder whether Autonomic network may need some
>>>>> intervention from human operators through some low-level policies beyond
>>>>> intents.
>>>>>>           For example, in the
>>>>> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it
>>> is
>>>>> suggested that the prefix lengths for the CSG, ASG, RSG (different
>>> roles in
>>>>> IP RAN) can be assigned as an "intent". These configuration parameters
>>> need
>>>>> to be distributed in the autonomic domain to influence the detail
>>>>> configurations on each autonomic node.
>>>>>>           As intent is described as an abstract, declarative, high-level
>>>>> policy used to operate an autonomic domain, such as an enterprise
>>> network.
>>>>> Perhaps we can rename the “intent†in the “prefix-management ID†to some
>>>>> other words.
>>>>>> Is “ASA parameters" ok here?
>>>>>>
>>>>>> I think some characteristics for this kind of low-level policies are
>>>>>>
>>>>>> 1.       Network level parameters, need to be distributed by GRASP and
>>>>> ACP
>>>>>> 2.       Relatively static compared to intent, perhaps only needed to
>>> be
>>>>> configured once
>>>>>> 3.       Unavoidable in current network environment, mostly for
>>>>> establishing the network infrastructure
>>>>>> 4.       Rarely need coordinations with others parameters, usually is
>>>>> closed related to one specific autonomic function
>>>>>> Welcome for comment. Your suggestions will be appreciated.
>>>>>>
>>>>>>
>>>>>> Best Regards
>>>>>> Zongpeng Du
>>>>>>
>>>>>> _______________________________________________
>>>>>> Anima mailing list
>>>>>> Anima@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Anima mailing list
>>>>>> Anima@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>>>
>>>>>
>>>>
>>>
>>
>

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------070605040304060404060501
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Hello,

    
    

    

    

      
Indeed not. An ASA will have some default behaviour that amounts to having
default settings for intent, but by definition intent comes from humans.
(One qualification: if the intent that reaches a given node is inconsistent,
then resolving that inconsistency might be autonomic.)

    

    

    Brian: could you explain what you mean here, especially "resolving
    that inconsistency might be autonomic"?

    I guess you mean that ASAs/nodes can autonomously detect
    inconsistency in intents and (try to) solve this inconsistency (on
    their own or via collaboration)...?

    

    


      

        

          
Of course, when we put Intent into the picture, there's a presumption
that Intent comes from some authority such as a NOC, but that's an
external semantic, not an assumption of the autonomic model.


        

        
Well, no - if the autonomic system is going to use intent, or at least
respond to it, then it must understand those external semantics.
This is even more important if the network is supposed to form
according to one or more intent rules.


      

      
Right, but what I meant is that the network will operate perfectly well
if no intent is sent, by using defaults. That will always happen to the
extent that intent distribution will take a finite time. Until the intent
reaches a node, it will operate on defaults.

    

    

    From collaborations with some operators on autonomic networking
    projects: default behavior without policy must equal to "do
    nothing". But operators' views might change/evolve...

    Also, and this is based on feedback from the field (e.g. 3GPP SON),
    turning on autonomic functions, even with well-thought parameter
    settings, turns out most of the times in chaotic behaviors and
    severe performance degradations.

    A key element to consider here is how to "gracefully" start/ramp-up
    an autonomic network and aspects of coordination (cf
    draft-ciavaglia-anima-coordination-01).

    

    

      
 Of course, the semantics of
intent must be well-defined, but it's an extra layer.



      

        
If intent is formulated at an abstract level, how is an autonomic
network going to consume intent that may have drastically different
syntactical structures, and likely, different semantics associated with
those syntactical structures?


      

      
Agreed. We need a model, syntax and semantics.



      

        

        

          
I think the point is that AN *adds* a new mode of communication
between self-managing nodes, that doesn't exist in a strictly
north/south scenario. That is meant to reduce the amount of
north/south communication and reduce the amount of
centralized work. So there's a trade-off - more autonomic
messaging means less north-south messaging.

        

      

    

    

    north-south is a matter of convention / roles.

    if we consider PBM / intent, for me it implies a "vertical"
    (authority) relationship.

    if we consider a more "flat" system of collaborative entities, then
    let's have a look at MAS / DAI approaches and not reinvent (a
    squarred) wheel.

    we may end up pushing more or less intelligence in the autonomic
    functions, but we also consider the scope of ANIMA to address
    "managed" networks (maybe not only this one), so it means some
    operators will expect certain behavior/performance from the neworks.
    Intent is one way to inform the network what is expected. 

    

    

      

        

          

        

        
There may indeed be a new mode of communication required.
However, if intent is not standardized, how does this work?


      

      
It doesn't. My only point is that intent is layered on top of an
underlying autonomic behaviour based on default behaviours. Actually,
that's why the question of intent was deferred in the WG charter.

    

    

    Your explanation is "valid", but I don't remember this was the
    reason used to consider intent FFS.

    Without rewriting history, we are facing now with the "heavy"
    design/conceptual discussions on the list, the effect of some
    decisions to exclude architectural analysis / investigations in the
    early phase of ANIMA...

    

    Laurent.

    

    

    

      

    Brian



      

        
I understand how unstructured and structured P2P networks
work. Let's assume this is a structured P2P network. In that case,
a DHT is used to assign ownership of each file to a peer. I fail to
understand how this works in an intent case. Files are passive,
intent is active. Please explain.

Regards,
John

On Sat, Mar 26, 2016 at 11:53 AM, Brian E Carpenter <
brian.e.carpenter@gmail.com> wrote:



        

          
Hi John,
On 27/03/2016 04:54, John Strassner wrote:


          

            

              
Correct, and there is also no presumption that individual nodes
have perfect knowledge of the topology. Of course the ACP
routing protocol will have perfect knowledge of the ACP topology,
but individual ASAs only know what they have gleaned from
discovery and what they have been told by Intent.

It is, I think, a very different world view than the SDN or SUPA or
NVO4 or NFV view.


            

            
Please note that I did NOT say anything about SDN, NFV, or SUPA
here. In fact, I'm confused as to why you both think I did.


          

          
I can't speak for Zongpeng, but I was triggered by the mention of
northbound and southbound APIs. There is no presumed direction in
Anima relationships, since in the limit a network is self-forming;
all relationships are peer-to-peer.

Of course, when we put Intent into the picture, there's a presumption
that Intent comes from some authority such as a NOC, but that's an
external semantic, not an assumption of the autonomic model.



          

            
I specifically said "There are plenty of southbound APIs (from a
management function to a device). There are very few northbound APIs."

Note the word "management function".

For the record, I don't think that either Phase 2 of NFV or the SDN 1.1
architecture have treated policy in any detail. Both architectures lack a
policy management component. I hope that ANIMA doesn't make the
same mistake.


          

          
Certainly, Intent needs a number of properties such as self-consistency
and stability that imply it is managed in some way.



          

            

              

                
However, in ANIMA network, perhaps we do not have a centralized
parsing node (controller), which can replace almost all the
communications between nodes.


              

            

            
Earlier I wrote about intent requiring a translation function. I never


          

          
said


          

            
that it was centralized or distributed. In the autonomic architectures I
have built, it was in fact a decentralized (agent-driven) function.
Furthermore, it did not "replace almost all the communications" - it
was used for compilation. There was, in fact, a separate distribution
mechanism (which I have also described), but that was limited to
"just" distribution.

I really don't understand what "replacing communications" means.


          

          
I think the point is that AN *adds* a new mode of communication between
self-managing nodes, that doesn't exist in a strictly north/south
scenario. That is meant to reduce the amount of north/south communication
and reduce the amount of centralized work. So there's a trade-off -
more autonomic messaging means less north-south messaging.

Best regards
   Brian



          

            
regards,
John

On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <
brian.e.carpenter@gmail.com> wrote:



            

              

                
However, in ANIMA network, perhaps we do not have a centralized parsing


              

              
node (controller), which can replace almost all the communications


            

          

          
between


          

            

              
nodes.

Correct, and there is also no presumption that individual nodes have
perfect
knowledge of the topology. Of course the ACP routing protocol will have
perfect
knowledge of the ACP topology, but individual ASAs only know what they


            

          

          
have


          

            

              
gleaned from discovery and what they have been told by Intent.

It is, I think, a very different world view than the SDN or SUPA or NVO4
or NFV view.

Regards
   Brian

On 26/03/2016 16:46, Duzongpeng wrote:


              

                
Hi John:

Thanks for your reply.

<jcs>
How is this different than running a script, or invoking an API? There


              

              
are plenty of southbound APIs (from a management function to a device).
There are very few northbound APIs. That's what the operators that I


            

          

          
have


          

            

              
talked to want.


              

                
</jcs>

I agree that intent belongs to northbound interface in the SDN network.


              

              
And ANIMA network perhaps can share the same northbound interface.


              

                
But, my understanding is that ANIMA network is a little different from


              

              
SDN network.


              

                
SDN controller will parse the intent, and transform to detailed


              

              
configurations of every device. It is a good model, and little
communications between nodes are needed.


              

                
Some network level parameters can be handled in SDN controller.

However, in ANIMA network, perhaps we do not have a centralized parsing


              

              
node (controller), which can replace almost all the communications


            

          

          
between


          

            

              
nodes.


              

                
So when a network operator is operating a ANIMA network, the network


              

              
level parameters are still needed to be flooded.


              

                
I mean the ANIMA interface is perhaps something like “northbound + part


              

              
of southboundâ€.


              

                
Best Regards
Zongpeng Du


From: John Strassner [mailto:strazpdj@gmail.com]
Sent: Saturday, March 26, 2016 9:41 AM
To: Duzongpeng; John Strassner
Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang;


              

              
anima@ietf.org; Jéferson Campos Nobre


              

                
Subject: Re: [Anima] Does Autonomic network need some intervention


              

              
beyond intent


              

                
Hi Zongpeng,

I think we have a misunderstanding here.


But I do not support that the intent writer should look every detail in


              

              
the catalog.


              

                
<jcs>
Most catalogs that I have seen have very few details. They instead tell


              

              
the user what services or products are available.


              

                
</jcs>

The intent writer may mainly care about what they want, but how to do


              

            

          

          
it


          

            

              
can be left to the autonomic network to deal with.


              

                
<jcs>
I agree completely.
</jcs>

Perhaps several solutions exist, and autonomic management system choose


              

              
one according to the abilities collected.


              

                
<jcs>
This is one of the reasons why I would like to see capabilities (and


              

              
needs) advertised.


              

                
</jcs>

...
I mean in current status, some of the Autonomic Functions should be


              

              
exposed to the network operator, and some of the Autonomic Functions


            

          

          
can be


          

            

              
hidden to the network operator


              

                
...
<jcs>
How is this different than running a script, or invoking an API? There


              

              
are plenty of southbound APIs (from a management function to a device).
There are very few northbound APIs. That's what the operators that I


            

          

          
have


          

            

              
talked to want.


              

                
regards,
John

On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng <duzongpeng@huawei.com


              

              
<mailto:duzongpeng@huawei.com>> wrote:


              

                
Hi John

Some personal understanding here. If any misunderstanding, please


              

              
correct me. Thanks.


              

                
<jcs2>
I would suggest a slight modification, to enable users of
intent (and even intent writers) to NOT have to have
detailed implementation information.

What if all of the Autonomic Functions, after they joined
the network, advertised both their capabilities as well as
their needs? We can put aside the latter (needs) as this
is more complex, but if the capabilities are advertised,
then the autonomic management system can assemble
those needs into a catalog. Then, the intent writer can
look at the catalog and write a declarative request to
use the functionality of the catalog.

I will look at your YANG model in a bit (sorry, I'm woefully
behind...).
< /jcs2>

I agree that autonomic node should have that self-advertisement


              

            

          

          
ability,


          

            

              
and autonomic management system should have this catalog.


              

                
But I do not support that the intent writer should look every detail in


              

              
the catalog.


              

                
The intent writer may mainly care about what they want, but how to do


              

            

          

          
it


          

            

              
can be left to the autonomic network to deal with.


              

                
Perhaps several solutions exist, and autonomic management system choose


              

              
one according to the abilities collected.


              

                
In the ideal, the network operator may communicate with the autonomic


              

              
network using network-level level intent and network-level report.


              

                
I mean in current status, some of the Autonomic Functions should be


              

              
exposed to the network operator, and some of the Autonomic Functions


            

          

          
can be


          

            

              
hidden to the network operator


              

                
Perhaps in future, when the network nodes become more intelligent, the


              

              
hidden part will increase.


              

                

Best regards
Zongpeng Du

From: Anima [mailto:anima-bounces@ietf.org<mailto:


              

            

          

          
anima-bounces@ietf.org>]


          

            

              
On Behalf Of John Strassner


              

                
Sent: Thursday, March 24, 2016 10:41 AM
To: Michael Behringer (mbehring); John Strassner
Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org<mailto:


              

              
anima@ietf.org>; Jéferson Campos Nobre


              

                
Subject: Re: [Anima] Does Autonomic network need some intervention


              

              
beyond intent


              

                
Hi Michael,

thanks for your reply. I do think that we largely agree, and would


              

            

          

          
enjoy


          

            

              
chatting during or (hopefully before) BA. Please see <jcs2>..</jcs2>.


            

          

          
Also


          

            

              
trying to unclutter, so I am eliding points that we agree on.


              

                

While *CIEs don’t like magic, they do understand the concept of a


              

              
default behaviour which they can influence if needed. Any reasonable
network management approach tries to define “default†behaviour and
“exceptionsâ€. The AN story fills the “default†part very nicely, and


            

          

          
*CIEs


          

            

              
get that.


              

                
<jcs2>
Agreed!
</jcs2>

The “parameter†versus “policy†discussion: My view is that we


              

            

          

          
shouldn’t


          

            

              
try to define exactly *what* intent allows and doesn’t allow, since


            

          

          
we’ve


          

            

              
heard from many sides that the boundary is not clear, and it may be
impossible to have a really clear, usable definition. So my point was:
Let’s focus on the distribution model: Intent is flooded to all nodes.
Let’s NOT do selective flooding, unicast Intent to specific nodes, etc,
because that would blur the boundary even worse.


              

                
<jcs2>
I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1


              

              
was 2006, almost 10 years ago), we used a message bus and pub-sub. In


            

          

          
v2,


          

            

              
we flooded; in v3, we went back to pub-sub because we wanted to do some
optimization and auditing. But I would be happy to try some type of
controlled flooding again.


              

                
</jcs2>
My suggestion: Intent is flooded to all nodes in a domain. The ASA


              

              
owners define the content of what they want to push in their bit of the
Intent. If we insist of full flooding, we encourage the right behaviour.
And if someone really puts into his Intent “node A: do this; node B: do


            

          

          
the


          

            

              
other; node C: do the third†it is awkward enough for people to take a


            

          

          
step


          

            

              
back and think whether they’re doing the right thing ☺


              

                
<jcs2>
I agree in principle with what you said. Plus, see above.
< /jcs2>
<jcs>
Do you mean that the intent writer will **know the difference**
between the fabric and each autonomic function? I hope not,
but perhaps I'm misunderstanding you...
< /jcs>
Not sure I understand your concern. Yes, in my model there is a section


              

              
for the fabric, and sections for the various functions. Seems a natural
structure?!? <jcs2>


              

                
The intent **writer** would have to know implementation
details in order to place everything in a single file. I would
rather not require this, and instead, have the intent engine
(e.g., compiler and associated logic) do this.
< /jcs2>
“Function ownerâ€: Yes, I’m suggesting to structure Intent around


              

              
Autonomic Functions. This is not optimal from a scientific point of


            

          

          
view,


          

            

              
but aligns very well with the way how things are organised. Probably we
need to discuss that in BA. But refer to my YANG model sent a few days


            

          

          
ago.


          

            

              

                
<jcs2>
I would suggest a slight modification, to enable users of
intent (and even intent writers) to NOT have to have
detailed implementation information.

What if all of the Autonomic Functions, after they joined
the network, advertised both their capabilities as well as
their needs? We can put aside the latter (needs) as this
is more complex, but if the capabilities are advertised,
then the autonomic management system can assemble
those needs into a catalog. Then, the intent writer can
look at the catalog and write a declarative request to
use the functionality of the catalog.

I will look at your YANG model in a bit (sorry, I'm woefully
behind...).
< /jcs2>


regards,
John

On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <


              

              
mbehring@cisco.com<mailto:mbehring@cisco.com>> wrote:


              

                
John,

I think we’re on the same page. Couple of points, top posting here,


              

              
otherwise it gets too messy.


              

                
While *CIEs don’t like magic, they do understand the concept of a


              

              
default behaviour which they can influence if needed. Any reasonable
network management approach tries to define “default†behaviour and
“exceptionsâ€. The AN story fills the “default†part very nicely, and


            

          

          
*CIEs


          

            

              
get that.


              

                
The “parameter†versus “policy†discussion: My view is that we


              

            

          

          
shouldn’t


          

            

              
try to define exactly *what* intent allows and doesn’t allow, since


            

          

          
we’ve


          

            

              
heard from many sides that the boundary is not clear, and it may be
impossible to have a really clear, usable definition. So my point was:
Let’s focus on the distribution model: Intent is flooded to all nodes.
Let’s NOT do selective flooding, unicast Intent to specific nodes, etc,
because that would blur the boundary even worse.


              

                
My suggestion: Intent is flooded to all nodes in a domain. The ASA


              

              
owners define the content of what they want to push in their bit of the
Intent. If we insist of full flooding, we encourage the right behaviour.
And if someone really puts into his Intent “node A: do this; node B: do


            

          

          
the


          

            

              
other; node C: do the third†it is awkward enough for people to take a


            

          

          
step


          

            

              
back and think whether they’re doing the right thing ☺


              

                
<jcs>
Do you mean that the intent writer will **know the difference**
between the fabric and each autonomic function? I hope not,
but perhaps I'm misunderstanding you...
</jcs>

Not sure I understand your concern. Yes, in my model there is a section


              

              
for the fabric, and sections for the various functions. Seems a natural
structure?!?


              

                
“Function ownerâ€: Yes, I’m suggesting to structure Intent around


              

              
Autonomic Functions. This is not optimal from a scientific point of


            

          

          
view,


          

            

              
but aligns very well with the way how things are organised. Probably we
need to discuss that in BA. But refer to my YANG model sent a few days


            

          

          
ago.


          

            

              

                
We should schedule some discussion time outside the ANIMA meeting


              

              
window...


              

                
Michael


From: John Strassner [mailto:strazpdj@gmail.com<mailto:


              

              
strazpdj@gmail.com>]


              

                
Sent: 22 March 2016 21:32
To: Michael Behringer (mbehring) <mbehring@cisco.com<mailto:


              

              
mbehring@cisco.com>>; John Strassner <strazpdj@gmail.com<mailto:
strazpdj@gmail.com>>


              

                
Cc: Duzongpeng <duzongpeng@huawei.com<mailto:duzongpeng@huawei.com>>;


              

              
anima@ietf.org<mailto:anima@ietf.org>; Laurent Ciavaglia <
laurent.ciavaglia@nokia.com<mailto:laurent.ciavaglia@nokia.com>>;
Jéferson Campos Nobre <jcnobre@inf.ufrgs.br<mailto:jcnobre@inf.ufrgs.br
;
Sheng Jiang <jiangsheng@huawei.com<mailto:jiangsheng@huawei.com>>


              

                
Subject: Re: [Anima] Does Autonomic network need some intervention


              

              
beyond intent


              

                
In general, I agree with Michael. I'd like to emphasize and expand on a


              

              
couple of key points:


              

                

First, ANIMA isn’t the only thing “influencing†or “steering†a


              

            

          

          
network,


          

            

              
at least not initially. There is still traditional config, plus all the
existing and emerging SDN models. My view has always been that ANIMA


            

          

          
sets a


          

            

              
“default†policy in a network, and all other means are more specific and
can override intent. This is described in RFC7575. Main point here:


            

          

          
ANIMA


          

            

              
Intent isn’t the only tool in a network.


              

                
<jcs>
Absolutely agree. Note that even in the Congress case I mentioned
earlier, it is NOT pushing config, and it is NOT viewed as the
"only" policy
</jcs>


Intent could perfectly well cover a high level policy such as “all


              

            

          

          
nodes


          

            

              
of type x do this; type y does thatâ€. (Which I think is what you mention
below.) When we started with Intent, the vision was to keep it very high
level, more along the line “do the right thing†☺


              

                
<jcs>
Agreed. Although to me, this means that we need to consider the types


              

            

          

          
of


          

            

              
actors that are using intent. For example, most of my CCIE and JCIE


            

          

          
friends


          

            

              
have no interest in using an intent language for pushing configuration.


            

          

          
On


          

            

              
the other hand, when I talk to operators and application developers,


            

          

          
they


          

            

              
would love to be able to use a high-level intent and have the intent


            

          

          
engine


          

            

              
take care of the details for them.


              

                
</jcs>

It has become clear that in today’s networking, people will want to


              

            

          

          
push


          

            

              
more “parameters†than “policyâ€, and I think we’ll just have to


            

          

          
acknowledge


          

            

              
that.


              

                
<jcs>
Sorry Michael, I don't follow here. What do you mean
by "parameters"?
</jcs>


Personally I think it’ll pan out like this: Intent will have sections


              

              
per autonomic function. An Intent parser will validate overall


            

          

          
correctness


          

            

              
(signature, etc), then segment the Intent into its bits: One part will


            

          

          
be


          

            

              
the autonomic fabric itself, then there will be bits for each autonomic
function.


              

                
<jcs>
Do you mean that the intent writer will **know the difference**
between the fabric and each autonomic function? I hope not,
but perhaps I'm misunderstanding you...
</jcs>


So the Intent parser on a node will grab intent, see there is some


              

              
Intent for autonomic function A, will see whether this is running


            

          

          
locally,


          

            

              
and if so, it’ll just pass that piece of Intent to that function.


              

                
<jcs>
Agreed.
</jcs>


In this way, a function owner defines both the intent itself, and how


              

            

          

          
it


          

            

              
is consumed by his/her function.


              

                
<jcs>
I'm not sure what you mean by "function owner". (Sorry to be
picky, I'm likely under-caffeinated.)

The above seemed like an autonomous decision, made by the parser.
I don't think that the intent writer should be cognizant of how many or
what types of autonomic functions are present.
</jcs>


The downside is that there is no “control†and people can do what they


              

              
want, in the worst case even push config, which is clearly NOT what


            

          

          
Intent


          

            

              
is meant to do.


              

                
<jcs>
And hence, the "HAL 2000" scenario. :-)
</jcs>


I’d hope that people don’t start writing Intent like “node 17: here is


              

              
your CLI; node 23: here is your CLIâ€; node 37: here is your CLIâ€. But


            

          

          
what


          

            

              
I describe above COULD be abused in this way. RFC7575 says this is NOT
Intent.


              

                
<jcs>
Absolutely agree!
</jcs>

 regards,
John



On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <


              

              
mbehring@cisco.com<mailto:mbehring@cisco.com>> wrote:


              

                
Hi Zongpeng,

Yes, this is a recurring question and topic. Let me give my view on two


              

              
levels:


              

                
First, ANIMA isn’t the only thing “influencing†or “steering†a


              

            

          

          
network,


          

            

              
at least not initially. There is still traditional config, plus all the
existing and emerging SDN models. My view has always been that ANIMA


            

          

          
sets a


          

            

              
“default†policy in a network, and all other means are more specific and
can override intent. This is described in RFC7575. Main point here:


            

          

          
ANIMA


          

            

              
Intent isn’t the only tool in a network.


              

                
Intent could perfectly well cover a high level policy such as “all


              

            

          

          
nodes


          

            

              
of type x do this; type y does thatâ€. (Which I think is what you mention
below.) When we started with Intent, the vision was to keep it very high
level, more along the line “do the right thing†☺  It has become clear


            

          

          
that


          

            

              
in today’s networking, people will want to push more “parameters†than
“policyâ€, and I think we’ll just have to acknowledge that.


              

                
Personally I think it’ll pan out like this: Intent will have sections


              

              
per autonomic function. An Intent parser will validate overall


            

          

          
correctness


          

            

              
(signature, etc), then segment the Intent into its bits: One part will


            

          

          
be


          

            

              
the autonomic fabric itself, then there will be bits for each autonomic
function.


              

                
So the Intent parser on a node will grab intent, see there is some


              

              
Intent for autonomic function A, will see whether this is running


            

          

          
locally,


          

            

              
and if so, it’ll just pass that piece of Intent to that function.


              

                
In this way, a function owner defines both the intent itself, and how


              

            

          

          
it


          

            

              
is consumed by his/her function.


              

                
The downside is that there is no “control†and people can do what they


              

              
want, in the worst case even push config, which is clearly NOT what


            

          

          
Intent


          

            

              
is meant to do. I’d hope that people don’t start writing Intent like


            

          

          
“node


          

            

              
17: here is your CLI; node 23: here is your CLIâ€; node 37: here is your
CLIâ€. But what I describe above COULD be abused in this way. RFC7575


            

          

          
says


          

            

              
this is NOT Intent.


              

                
In other words: Let’s not be too prescriptive, and hope that folks will


              

              
do more or less reasonable things.


              

                
I’ve been meaning to comment on the Intent drafts and possibly


              

              
contribute, I’m just struggling with priorities at the moment (as you


            

          

          
can


          

            

              
see by my relative silence over the past weeks).


              

                
Michael



From: Anima [mailto:anima-bounces@ietf.org<mailto:


              

            

          

          
anima-bounces@ietf.org>]


          

            

              
On Behalf Of Duzongpeng


              

                
Sent: 16 March 2016 02:38
To: anima@ietf.org<mailto:anima@ietf.org>
Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com<mailto:


              

              
laurent.ciavaglia@nokia.com>>; Jéferson Campos Nobre <


            

          

          
jcnobre@inf.ufrgs.br


          

            

              
<mailto:jcnobre@inf.ufrgs.br>>; Sheng Jiang <jiangsheng@huawei.com


            

          

          
<mailto:


          

            

              
jiangsheng@huawei.com>>


              

                
Subject: [Anima] Does Autonomic network need some intervention beyond


              

              
intent


              

                
Hello, everyone

         I understand that Autonomic network should be able to work


              

            

          

          
with


          

            

              
as less configuration as possible. In the ideal case, an Autonomic


            

          

          
network


          

            

              
should be able to work well while only intent is needed from human
operators.


              

                
         However, I wonder whether Autonomic network may need some


              

              
intervention from human operators through some low-level policies beyond
intents.


              

                
         For example, in the


              

              
https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it


            

          

          
is


          

            

              
suggested that the prefix lengths for the CSG, ASG, RSG (different


            

          

          
roles in


          

            

              
IP RAN) can be assigned as an "intent". These configuration parameters


            

          

          
need


          

            

              
to be distributed in the autonomic domain to influence the detail
configurations on each autonomic node.


              

                
         As intent is described as an abstract, declarative, high-level


              

              
policy used to operate an autonomic domain, such as an enterprise


            

          

          
network.


          

            

              
Perhaps we can rename the “intent†in the “prefix-management ID†to some
other words.


              

                
Is “ASA parameters" ok here?

I think some characteristics for this kind of low-level policies are

1.       Network level parameters, need to be distributed by GRASP and


              

              
ACP


              

                
2.       Relatively static compared to intent, perhaps only needed to


              

            

          

          
be


          

            

              
configured once


              

                
3.       Unavoidable in current network environment, mostly for


              

              
establishing the network infrastructure


              

                
4.       Rarely need coordinations with others parameters, usually is


              

              
closed related to one specific autonomic function


              

                
Welcome for comment. Your suggestions will be appreciated.


Best Regards
Zongpeng Du

_______________________________________________
Anima mailing list
Anima@ietf.org<mailto:Anima@ietf.org>
https://www.ietf.org/mailman/listinfo/anima



--
regards,
John



--
regards,
John



--
regards,
John



_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima



              

              



            

            



          

          



        

        



      

      



    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------070605040304060404060501--


From nobody Tue Mar 29 09:37:30 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1FDC812DACE for ; Tue, 29 Mar 2016 09:37:27 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XCukJk0RgIbc for ; Tue, 29 Mar 2016 09:37:17 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id BB46712DADF for ; Tue, 29 Mar 2016 09:30:18 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id 352C21D82096C; Tue, 29 Mar 2016 16:30:12 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2TGUEZb018557 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Tue, 29 Mar 2016 16:30:15 GMT
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2TGU75B022763 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 29 Mar 2016 18:30:14 +0200
Received: from [172.27.205.120] (135.239.27.38) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Tue, 29 Mar 2016 18:30:06 +0200
To: EXT Brian E Carpenter , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>     <56F60CE4.9040901@gmail.com>  <56F6DAA1.8000206@gmail.com>  <56F82FDD.7090405@gmail.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56FAAD88.9050902@alcatel-lucent.com>
Date: Tue, 29 Mar 2016 18:30:00 +0200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <56F82FDD.7090405@gmail.com>
Content-Type: multipart/alternative; boundary="------------020905070204040504040807"
X-Originating-IP: [135.239.27.38]
Archived-At: 
Cc: Laurent Ciavaglia , Duzongpeng , "Michael Behringer \(mbehring\)" , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 16:37:27 -0000

--------------020905070204040504040807
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit

Hello,


> Indeed not. An ASA will have some default behaviour that amounts to having
> default settings for intent, but by definition intent comes from humans.
> (One qualification: if the intent that reaches a given node is inconsistent,
> then resolving that inconsistency might be autonomic.)

Brian: could you explain what you mean here, especially "resolving that 
inconsistency might be autonomic"?
I guess you mean that ASAs/nodes can autonomously detect inconsistency 
in intents and (try to) solve this inconsistency (on their own or via 
collaboration)...?

>
>>> Of course, when we put Intent into the picture, there's a presumption
>>> that Intent comes from some authority such as a NOC, but that's an
>>> external semantic, not an assumption of the autonomic model.
>> Well, no - if the autonomic system is going to use intent, or at least
>> respond to it, then it must understand those external semantics.
>> This is even more important if the network is supposed to form
>> according to one or more intent rules.
> Right, but what I meant is that the network will operate perfectly well
> if no intent is sent, by using defaults. That will always happen to the
> extent that intent distribution will take a finite time. Until the intent
> reaches a node, it will operate on defaults.

 From collaborations with some operators on autonomic networking 
projects: default behavior without policy must equal to "do nothing". 
But operators' views might change/evolve...
Also, and this is based on feedback from the field (e.g. 3GPP SON), 
turning on autonomic functions, even with well-thought parameter 
settings, turns out most of the times in chaotic behaviors and severe 
performance degradations.
A key element to consider here is how to "gracefully" start/ramp-up an 
autonomic network and aspects of coordination (cf 
draft-ciavaglia-anima-coordination-01).

>   Of course, the semantics of
> intent must be well-defined, but it's an extra layer.
>
>> If intent is formulated at an abstract level, how is an autonomic
>> network going to consume intent that may have drastically different
>> syntactical structures, and likely, different semantics associated with
>> those syntactical structures?
> Agreed. We need a model, syntax and semantics.
>
>>> I think the point is that AN *adds* a new mode of communication
>>> between self-managing nodes, that doesn't exist in a strictly
>>> north/south scenario. That is meant to reduce the amount of
>>> north/south communication and reduce the amount of
>>> centralized work. So there's a trade-off - more autonomic
>>> messaging means less north-south messaging.

north-south is a matter of convention / roles.
if we consider PBM / intent, for me it implies a "vertical" (authority) 
relationship.
if we consider a more "flat" system of collaborative entities, then 
let's have a look at MAS / DAI approaches and not reinvent (a squared) 
wheel.
we may end up pushing more or less intelligence in the autonomic 
functions, but we also consider the scope of ANIMA to address "managed" 
networks (maybe not only this one), so it means some operators will 
expect certain behavior/performance from the networks. Intent is one way 
to inform the network what is expected.

>> There may indeed be a new mode of communication required.
>> However, if intent is not standardized, how does this work?
> It doesn't. My only point is that intent is layered on top of an
> underlying autonomic behaviour based on default behaviours. Actually,
> that's why the question of intent was deferred in the WG charter.

Your explanation is "valid", but I don't remember this was the reason 
used to consider intent FFS.
Without rewriting history, we are facing now with the "heavy" 
design/conceptual discussions on the list, the effect of some decisions 
to exclude architectural analysis / investigations in the early phase of 
ANIMA...

Laurent.


>
>      Brian
>
>> I understand how unstructured and structured P2P networks
>> work. Let's assume this is a structured P2P network. In that case,
>> a DHT is used to assign ownership of each file to a peer. I fail to
>> understand how this works in an intent case. Files are passive,
>> intent is active. Please explain.
>>
>> Regards,
>> John
>>
>> On Sat, Mar 26, 2016 at 11:53 AM, Brian E Carpenter <
>> brian.e.carpenter@gmail.com> wrote:
>>
>>> Hi John,
>>> On 27/03/2016 04:54, John Strassner wrote:
>>>>> Correct, and there is also no presumption that individual nodes
>>>>> have perfect knowledge of the topology. Of course the ACP
>>>>> routing protocol will have perfect knowledge of the ACP topology,
>>>>> but individual ASAs only know what they have gleaned from
>>>>> discovery and what they have been told by Intent.
>>>>>
>>>>> It is, I think, a very different world view than the SDN or SUPA or
>>>>> NVO4 or NFV view.
>>>> Please note that I did NOT say anything about SDN, NFV, or SUPA
>>>> here. In fact, I'm confused as to why you both think I did.
>>> I can't speak for Zongpeng, but I was triggered by the mention of
>>> northbound and southbound APIs. There is no presumed direction in
>>> Anima relationships, since in the limit a network is self-forming;
>>> all relationships are peer-to-peer.
>>>
>>> Of course, when we put Intent into the picture, there's a presumption
>>> that Intent comes from some authority such as a NOC, but that's an
>>> external semantic, not an assumption of the autonomic model.
>>>
>>>> I specifically said "There are plenty of southbound APIs (from a
>>>> management function to a device). There are very few northbound APIs."
>>>>
>>>> Note the word "management function".
>>>>
>>>> For the record, I don't think that either Phase 2 of NFV or the SDN 1.1
>>>> architecture have treated policy in any detail. Both architectures lack a
>>>> policy management component. I hope that ANIMA doesn't make the
>>>> same mistake.
>>> Certainly, Intent needs a number of properties such as self-consistency
>>> and stability that imply it is managed in some way.
>>>
>>>>>> However, in ANIMA network, perhaps we do not have a centralized
>>>>>> parsing node (controller), which can replace almost all the
>>>>>> communications between nodes.
>>>> Earlier I wrote about intent requiring a translation function. I never
>>> said
>>>> that it was centralized or distributed. In the autonomic architectures I
>>>> have built, it was in fact a decentralized (agent-driven) function.
>>>> Furthermore, it did not "replace almost all the communications" - it
>>>> was used for compilation. There was, in fact, a separate distribution
>>>> mechanism (which I have also described), but that was limited to
>>>> "just" distribution.
>>>>
>>>> I really don't understand what "replacing communications" means.
>>> I think the point is that AN *adds* a new mode of communication between
>>> self-managing nodes, that doesn't exist in a strictly north/south
>>> scenario. That is meant to reduce the amount of north/south communication
>>> and reduce the amount of centralized work. So there's a trade-off -
>>> more autonomic messaging means less north-south messaging.
>>>
>>> Best regards
>>>     Brian
>>>
>>>> regards,
>>>> John
>>>>
>>>> On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <
>>>> brian.e.carpenter@gmail.com> wrote:
>>>>
>>>>>> However, in ANIMA network, perhaps we do not have a centralized parsing
>>>>> node (controller), which can replace almost all the communications
>>> between
>>>>> nodes.
>>>>>
>>>>> Correct, and there is also no presumption that individual nodes have
>>>>> perfect
>>>>> knowledge of the topology. Of course the ACP routing protocol will have
>>>>> perfect
>>>>> knowledge of the ACP topology, but individual ASAs only know what they
>>> have
>>>>> gleaned from discovery and what they have been told by Intent.
>>>>>
>>>>> It is, I think, a very different world view than the SDN or SUPA or NVO4
>>>>> or NFV view.
>>>>>
>>>>> Regards
>>>>>     Brian
>>>>>
>>>>> On 26/03/2016 16:46, Duzongpeng wrote:
>>>>>> Hi John:
>>>>>>
>>>>>> Thanks for your reply.
>>>>>>
>>>>>> 
>>>>>> How is this different than running a script, or invoking an API? There
>>>>> are plenty of southbound APIs (from a management function to a device).
>>>>> There are very few northbound APIs. That's what the operators that I
>>> have
>>>>> talked to want.
>>>>>> 
>>>>>>
>>>>>> I agree that intent belongs to northbound interface in the SDN network.
>>>>> And ANIMA network perhaps can share the same northbound interface.
>>>>>> But, my understanding is that ANIMA network is a little different from
>>>>> SDN network.
>>>>>> SDN controller will parse the intent, and transform to detailed
>>>>> configurations of every device. It is a good model, and little
>>>>> communications between nodes are needed.
>>>>>> Some network level parameters can be handled in SDN controller.
>>>>>>
>>>>>> However, in ANIMA network, perhaps we do not have a centralized parsing
>>>>> node (controller), which can replace almost all the communications
>>> between
>>>>> nodes.
>>>>>> So when a network operator is operating a ANIMA network, the network
>>>>> level parameters are still needed to be flooded.
>>>>>> I mean the ANIMA interface is perhaps something like “northbound + part
>>>>> of southboundâ€.
>>>>>> Best Regards
>>>>>> Zongpeng Du
>>>>>>
>>>>>>
>>>>>> From: John Strassner [mailto:strazpdj@gmail.com]
>>>>>> Sent: Saturday, March 26, 2016 9:41 AM
>>>>>> To: Duzongpeng; John Strassner
>>>>>> Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang;
>>>>> anima@ietf.org; Jéferson Campos Nobre
>>>>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>>>>> beyond intent
>>>>>> Hi Zongpeng,
>>>>>>
>>>>>> I think we have a misunderstanding here.
>>>>>>
>>>>>>
>>>>>> But I do not support that the intent writer should look every detail in
>>>>> the catalog.
>>>>>> 
>>>>>> Most catalogs that I have seen have very few details. They instead tell
>>>>> the user what services or products are available.
>>>>>> 
>>>>>>
>>>>>> The intent writer may mainly care about what they want, but how to do
>>> it
>>>>> can be left to the autonomic network to deal with.
>>>>>> 
>>>>>> I agree completely.
>>>>>> 
>>>>>>
>>>>>> Perhaps several solutions exist, and autonomic management system choose
>>>>> one according to the abilities collected.
>>>>>> 
>>>>>> This is one of the reasons why I would like to see capabilities (and
>>>>> needs) advertised.
>>>>>> 
>>>>>>
>>>>>> ...
>>>>>> I mean in current status, some of the Autonomic Functions should be
>>>>> exposed to the network operator, and some of the Autonomic Functions
>>> can be
>>>>> hidden to the network operator
>>>>>> ...
>>>>>> 
>>>>>> How is this different than running a script, or invoking an API? There
>>>>> are plenty of southbound APIs (from a management function to a device).
>>>>> There are very few northbound APIs. That's what the operators that I
>>> have
>>>>> talked to want.
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>> On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng >>>> > wrote:
>>>>>> Hi John
>>>>>>
>>>>>> Some personal understanding here. If any misunderstanding, please
>>>>> correct me. Thanks.
>>>>>> 
>>>>>> I would suggest a slight modification, to enable users of
>>>>>> intent (and even intent writers) to NOT have to have
>>>>>> detailed implementation information.
>>>>>>
>>>>>> What if all of the Autonomic Functions, after they joined
>>>>>> the network, advertised both their capabilities as well as
>>>>>> their needs? We can put aside the latter (needs) as this
>>>>>> is more complex, but if the capabilities are advertised,
>>>>>> then the autonomic management system can assemble
>>>>>> those needs into a catalog. Then, the intent writer can
>>>>>> look at the catalog and write a declarative request to
>>>>>> use the functionality of the catalog.
>>>>>>
>>>>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>>>>> behind...).
>>>>>> < /jcs2>
>>>>>>
>>>>>> I agree that autonomic node should have that self-advertisement
>>> ability,
>>>>> and autonomic management system should have this catalog.
>>>>>> But I do not support that the intent writer should look every detail in
>>>>> the catalog.
>>>>>> The intent writer may mainly care about what they want, but how to do
>>> it
>>>>> can be left to the autonomic network to deal with.
>>>>>> Perhaps several solutions exist, and autonomic management system choose
>>>>> one according to the abilities collected.
>>>>>> In the ideal, the network operator may communicate with the autonomic
>>>>> network using network-level level intent and network-level report.
>>>>>> I mean in current status, some of the Autonomic Functions should be
>>>>> exposed to the network operator, and some of the Autonomic Functions
>>> can be
>>>>> hidden to the network operator
>>>>>> Perhaps in future, when the network nodes become more intelligent, the
>>>>> hidden part will increase.
>>>>>>
>>>>>> Best regards
>>>>>> Zongpeng Du
>>>>>>
>>>>>> From: Anima [mailto:anima-bounces@ietf.org>> anima-bounces@ietf.org>]
>>>>> On Behalf Of John Strassner
>>>>>> Sent: Thursday, March 24, 2016 10:41 AM
>>>>>> To: Michael Behringer (mbehring); John Strassner
>>>>>> Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org>>>> anima@ietf.org>; Jéferson Campos Nobre
>>>>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>>>>> beyond intent
>>>>>> Hi Michael,
>>>>>>
>>>>>> thanks for your reply. I do think that we largely agree, and would
>>> enjoy
>>>>> chatting during or (hopefully before) BA. Please see ...
>>> Also
>>>>> trying to unclutter, so I am eliding points that we agree on.
>>>>>>
>>>>>> While *CIEs don’t like magic, they do understand the concept of a
>>>>> default behaviour which they can influence if needed. Any reasonable
>>>>> network management approach tries to define “default†behaviour and
>>>>> “exceptionsâ€. The AN story fills the “default†part very nicely, and
>>> *CIEs
>>>>> get that.
>>>>>> 
>>>>>> Agreed!
>>>>>> 
>>>>>>
>>>>>> The “parameter†versus “policy†discussion: My view is that we
>>> shouldn’t
>>>>> try to define exactly *what* intent allows and doesn’t allow, since
>>> we’ve
>>>>> heard from many sides that the boundary is not clear, and it may be
>>>>> impossible to have a really clear, usable definition. So my point was:
>>>>> Let’s focus on the distribution model: Intent is flooded to all nodes.
>>>>> Let’s NOT do selective flooding, unicast Intent to specific nodes, etc,
>>>>> because that would blur the boundary even worse.
>>>>>> 
>>>>>> I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1
>>>>> was 2006, almost 10 years ago), we used a message bus and pub-sub. In
>>> v2,
>>>>> we flooded; in v3, we went back to pub-sub because we wanted to do some
>>>>> optimization and auditing. But I would be happy to try some type of
>>>>> controlled flooding again.
>>>>>> 
>>>>>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>>>>> owners define the content of what they want to push in their bit of the
>>>>> Intent. If we insist of full flooding, we encourage the right behaviour.
>>>>> And if someone really puts into his Intent “node A: do this; node B: do
>>> the
>>>>> other; node C: do the third†it is awkward enough for people to take a
>>> step
>>>>> back and think whether they’re doing the right thing ☺
>>>>>> 
>>>>>> I agree in principle with what you said. Plus, see above.
>>>>>> < /jcs2>
>>>>>> 
>>>>>> Do you mean that the intent writer will **know the difference**
>>>>>> between the fabric and each autonomic function? I hope not,
>>>>>> but perhaps I'm misunderstanding you...
>>>>>> < /jcs>
>>>>>> Not sure I understand your concern. Yes, in my model there is a section
>>>>> for the fabric, and sections for the various functions. Seems a natural
>>>>> structure?!? 
>>>>>> The intent **writer** would have to know implementation
>>>>>> details in order to place everything in a single file. I would
>>>>>> rather not require this, and instead, have the intent engine
>>>>>> (e.g., compiler and associated logic) do this.
>>>>>> < /jcs2>
>>>>>> “Function ownerâ€: Yes, I’m suggesting to structure Intent around
>>>>> Autonomic Functions. This is not optimal from a scientific point of
>>> view,
>>>>> but aligns very well with the way how things are organised. Probably we
>>>>> need to discuss that in BA. But refer to my YANG model sent a few days
>>> ago.
>>>>>> 
>>>>>> I would suggest a slight modification, to enable users of
>>>>>> intent (and even intent writers) to NOT have to have
>>>>>> detailed implementation information.
>>>>>>
>>>>>> What if all of the Autonomic Functions, after they joined
>>>>>> the network, advertised both their capabilities as well as
>>>>>> their needs? We can put aside the latter (needs) as this
>>>>>> is more complex, but if the capabilities are advertised,
>>>>>> then the autonomic management system can assemble
>>>>>> those needs into a catalog. Then, the intent writer can
>>>>>> look at the catalog and write a declarative request to
>>>>>> use the functionality of the catalog.
>>>>>>
>>>>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>>>>> behind...).
>>>>>> < /jcs2>
>>>>>>
>>>>>>
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <
>>>>> mbehring@cisco.com> wrote:
>>>>>> John,
>>>>>>
>>>>>> I think we’re on the same page. Couple of points, top posting here,
>>>>> otherwise it gets too messy.
>>>>>> While *CIEs don’t like magic, they do understand the concept of a
>>>>> default behaviour which they can influence if needed. Any reasonable
>>>>> network management approach tries to define “default†behaviour and
>>>>> “exceptionsâ€. The AN story fills the “default†part very nicely, and
>>> *CIEs
>>>>> get that.
>>>>>> The “parameter†versus “policy†discussion: My view is that we
>>> shouldn’t
>>>>> try to define exactly *what* intent allows and doesn’t allow, since
>>> we’ve
>>>>> heard from many sides that the boundary is not clear, and it may be
>>>>> impossible to have a really clear, usable definition. So my point was:
>>>>> Let’s focus on the distribution model: Intent is flooded to all nodes.
>>>>> Let’s NOT do selective flooding, unicast Intent to specific nodes, etc,
>>>>> because that would blur the boundary even worse.
>>>>>> My suggestion: Intent is flooded to all nodes in a domain. The ASA
>>>>> owners define the content of what they want to push in their bit of the
>>>>> Intent. If we insist of full flooding, we encourage the right behaviour.
>>>>> And if someone really puts into his Intent “node A: do this; node B: do
>>> the
>>>>> other; node C: do the third†it is awkward enough for people to take a
>>> step
>>>>> back and think whether they’re doing the right thing ☺
>>>>>> 
>>>>>> Do you mean that the intent writer will **know the difference**
>>>>>> between the fabric and each autonomic function? I hope not,
>>>>>> but perhaps I'm misunderstanding you...
>>>>>> 
>>>>>>
>>>>>> Not sure I understand your concern. Yes, in my model there is a section
>>>>> for the fabric, and sections for the various functions. Seems a natural
>>>>> structure?!?
>>>>>> “Function ownerâ€: Yes, I’m suggesting to structure Intent around
>>>>> Autonomic Functions. This is not optimal from a scientific point of
>>> view,
>>>>> but aligns very well with the way how things are organised. Probably we
>>>>> need to discuss that in BA. But refer to my YANG model sent a few days
>>> ago.
>>>>>> We should schedule some discussion time outside the ANIMA meeting
>>>>> window...
>>>>>> Michael
>>>>>>
>>>>>>
>>>>>> From: John Strassner [mailto:strazpdj@gmail.com>>>> strazpdj@gmail.com>]
>>>>>> Sent: 22 March 2016 21:32
>>>>>> To: Michael Behringer (mbehring) >>>> mbehring@cisco.com>>; John Strassner >>>> strazpdj@gmail.com>>
>>>>>> Cc: Duzongpeng >;
>>>>> anima@ietf.org; Laurent Ciavaglia <
>>>>> laurent.ciavaglia@nokia.com>;
>>>>> Jéferson Campos Nobre >>>> ;
>>>>> Sheng Jiang >
>>>>>> Subject: Re: [Anima] Does Autonomic network need some intervention
>>>>> beyond intent
>>>>>> In general, I agree with Michael. I'd like to emphasize and expand on a
>>>>> couple of key points:
>>>>>>
>>>>>> First, ANIMA isn’t the only thing “influencing†or “steering†a
>>> network,
>>>>> at least not initially. There is still traditional config, plus all the
>>>>> existing and emerging SDN models. My view has always been that ANIMA
>>> sets a
>>>>> “default†policy in a network, and all other means are more specific and
>>>>> can override intent. This is described in RFC7575. Main point here:
>>> ANIMA
>>>>> Intent isn’t the only tool in a network.
>>>>>> 
>>>>>> Absolutely agree. Note that even in the Congress case I mentioned
>>>>>> earlier, it is NOT pushing config, and it is NOT viewed as the
>>>>>> "only" policy
>>>>>> 
>>>>>>
>>>>>>
>>>>>> Intent could perfectly well cover a high level policy such as “all
>>> nodes
>>>>> of type x do this; type y does thatâ€. (Which I think is what you mention
>>>>> below.) When we started with Intent, the vision was to keep it very high
>>>>> level, more along the line “do the right thing†☺
>>>>>> 
>>>>>> Agreed. Although to me, this means that we need to consider the types
>>> of
>>>>> actors that are using intent. For example, most of my CCIE and JCIE
>>> friends
>>>>> have no interest in using an intent language for pushing configuration.
>>> On
>>>>> the other hand, when I talk to operators and application developers,
>>> they
>>>>> would love to be able to use a high-level intent and have the intent
>>> engine
>>>>> take care of the details for them.
>>>>>> 
>>>>>>
>>>>>> It has become clear that in today’s networking, people will want to
>>> push
>>>>> more “parameters†than “policyâ€, and I think we’ll just have to
>>> acknowledge
>>>>> that.
>>>>>> 
>>>>>> Sorry Michael, I don't follow here. What do you mean
>>>>>> by "parameters"?
>>>>>> 
>>>>>>
>>>>>>
>>>>>> Personally I think it’ll pan out like this: Intent will have sections
>>>>> per autonomic function. An Intent parser will validate overall
>>> correctness
>>>>> (signature, etc), then segment the Intent into its bits: One part will
>>> be
>>>>> the autonomic fabric itself, then there will be bits for each autonomic
>>>>> function.
>>>>>> 
>>>>>> Do you mean that the intent writer will **know the difference**
>>>>>> between the fabric and each autonomic function? I hope not,
>>>>>> but perhaps I'm misunderstanding you...
>>>>>> 
>>>>>>
>>>>>>
>>>>>> So the Intent parser on a node will grab intent, see there is some
>>>>> Intent for autonomic function A, will see whether this is running
>>> locally,
>>>>> and if so, it’ll just pass that piece of Intent to that function.
>>>>>> 
>>>>>> Agreed.
>>>>>> 
>>>>>>
>>>>>>
>>>>>> In this way, a function owner defines both the intent itself, and how
>>> it
>>>>> is consumed by his/her function.
>>>>>> 
>>>>>> I'm not sure what you mean by "function owner". (Sorry to be
>>>>>> picky, I'm likely under-caffeinated.)
>>>>>>
>>>>>> The above seemed like an autonomous decision, made by the parser.
>>>>>> I don't think that the intent writer should be cognizant of how many or
>>>>>> what types of autonomic functions are present.
>>>>>> 
>>>>>>
>>>>>>
>>>>>> The downside is that there is no “control†and people can do what they
>>>>> want, in the worst case even push config, which is clearly NOT what
>>> Intent
>>>>> is meant to do.
>>>>>> 
>>>>>> And hence, the "HAL 2000" scenario. :-)
>>>>>> 
>>>>>>
>>>>>>
>>>>>> I’d hope that people don’t start writing Intent like “node 17: here is
>>>>> your CLI; node 23: here is your CLIâ€; node 37: here is your CLIâ€. But
>>> what
>>>>> I describe above COULD be abused in this way. RFC7575 says this is NOT
>>>>> Intent.
>>>>>> 
>>>>>> Absolutely agree!
>>>>>> 
>>>>>>
>>>>>>   regards,
>>>>>> John
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
>>>>> mbehring@cisco.com> wrote:
>>>>>> Hi Zongpeng,
>>>>>>
>>>>>> Yes, this is a recurring question and topic. Let me give my view on two
>>>>> levels:
>>>>>> First, ANIMA isn’t the only thing “influencing†or “steering†a
>>> network,
>>>>> at least not initially. There is still traditional config, plus all the
>>>>> existing and emerging SDN models. My view has always been that ANIMA
>>> sets a
>>>>> “default†policy in a network, and all other means are more specific and
>>>>> can override intent. This is described in RFC7575. Main point here:
>>> ANIMA
>>>>> Intent isn’t the only tool in a network.
>>>>>> Intent could perfectly well cover a high level policy such as “all
>>> nodes
>>>>> of type x do this; type y does thatâ€. (Which I think is what you mention
>>>>> below.) When we started with Intent, the vision was to keep it very high
>>>>> level, more along the line “do the right thing†☺  It has become clear
>>> that
>>>>> in today’s networking, people will want to push more “parameters†than
>>>>> “policyâ€, and I think we’ll just have to acknowledge that.
>>>>>> Personally I think it’ll pan out like this: Intent will have sections
>>>>> per autonomic function. An Intent parser will validate overall
>>> correctness
>>>>> (signature, etc), then segment the Intent into its bits: One part will
>>> be
>>>>> the autonomic fabric itself, then there will be bits for each autonomic
>>>>> function.
>>>>>> So the Intent parser on a node will grab intent, see there is some
>>>>> Intent for autonomic function A, will see whether this is running
>>> locally,
>>>>> and if so, it’ll just pass that piece of Intent to that function.
>>>>>> In this way, a function owner defines both the intent itself, and how
>>> it
>>>>> is consumed by his/her function.
>>>>>> The downside is that there is no “control†and people can do what they
>>>>> want, in the worst case even push config, which is clearly NOT what
>>> Intent
>>>>> is meant to do. I’d hope that people don’t start writing Intent like
>>> “node
>>>>> 17: here is your CLI; node 23: here is your CLIâ€; node 37: here is your
>>>>> CLIâ€. But what I describe above COULD be abused in this way. RFC7575
>>> says
>>>>> this is NOT Intent.
>>>>>> In other words: Let’s not be too prescriptive, and hope that folks will
>>>>> do more or less reasonable things.
>>>>>> I’ve been meaning to comment on the Intent drafts and possibly
>>>>> contribute, I’m just struggling with priorities at the moment (as you
>>> can
>>>>> see by my relative silence over the past weeks).
>>>>>> Michael
>>>>>>
>>>>>>
>>>>>>
>>>>>> From: Anima [mailto:anima-bounces@ietf.org>> anima-bounces@ietf.org>]
>>>>> On Behalf Of Duzongpeng
>>>>>> Sent: 16 March 2016 02:38
>>>>>> To: anima@ietf.org
>>>>>> Cc: Laurent Ciavaglia >>>> laurent.ciavaglia@nokia.com>>; Jéferson Campos Nobre <
>>> jcnobre@inf.ufrgs.br
>>>>> >; Sheng Jiang >> >>>> jiangsheng@huawei.com>>
>>>>>> Subject: [Anima] Does Autonomic network need some intervention beyond
>>>>> intent
>>>>>> Hello, everyone
>>>>>>
>>>>>>           I understand that Autonomic network should be able to work
>>> with
>>>>> as less configuration as possible. In the ideal case, an Autonomic
>>> network
>>>>> should be able to work well while only intent is needed from human
>>>>> operators.
>>>>>>           However, I wonder whether Autonomic network may need some
>>>>> intervention from human operators through some low-level policies beyond
>>>>> intents.
>>>>>>           For example, in the
>>>>> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it
>>> is
>>>>> suggested that the prefix lengths for the CSG, ASG, RSG (different
>>> roles in
>>>>> IP RAN) can be assigned as an "intent". These configuration parameters
>>> need
>>>>> to be distributed in the autonomic domain to influence the detail
>>>>> configurations on each autonomic node.
>>>>>>           As intent is described as an abstract, declarative, high-level
>>>>> policy used to operate an autonomic domain, such as an enterprise
>>> network.
>>>>> Perhaps we can rename the “intent†in the “prefix-management ID†to some
>>>>> other words.
>>>>>> Is “ASA parameters" ok here?
>>>>>>
>>>>>> I think some characteristics for this kind of low-level policies are
>>>>>>
>>>>>> 1.       Network level parameters, need to be distributed by GRASP and
>>>>> ACP
>>>>>> 2.       Relatively static compared to intent, perhaps only needed to
>>> be
>>>>> configured once
>>>>>> 3.       Unavoidable in current network environment, mostly for
>>>>> establishing the network infrastructure
>>>>>> 4.       Rarely need coordinations with others parameters, usually is
>>>>> closed related to one specific autonomic function
>>>>>> Welcome for comment. Your suggestions will be appreciated.
>>>>>>
>>>>>>
>>>>>> Best Regards
>>>>>> Zongpeng Du
>>>>>>
>>>>>> _______________________________________________
>>>>>> Anima mailing list
>>>>>> Anima@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> regards,
>>>>>> John
>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Anima mailing list
>>>>>> Anima@ietf.org
>>>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>>>
>>>>>
>>>>
>>>
>>
>

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------020905070204040504040807
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Hello,

    
    

    

    

      
Indeed not. An ASA will have some default behaviour that amounts to having
default settings for intent, but by definition intent comes from humans.
(One qualification: if the intent that reaches a given node is inconsistent,
then resolving that inconsistency might be autonomic.)

    

    

    Brian: could you explain what you mean here, especially "resolving
    that inconsistency might be autonomic"?

    I guess you mean that ASAs/nodes can autonomously detect
    inconsistency in intents and (try to) solve this inconsistency (on
    their own or via collaboration)...?

    

    


      

        

          
Of course, when we put Intent into the picture, there's a presumption
that Intent comes from some authority such as a NOC, but that's an
external semantic, not an assumption of the autonomic model.


        

        
Well, no - if the autonomic system is going to use intent, or at least
respond to it, then it must understand those external semantics.
This is even more important if the network is supposed to form
according to one or more intent rules.


      

      
Right, but what I meant is that the network will operate perfectly well
if no intent is sent, by using defaults. That will always happen to the
extent that intent distribution will take a finite time. Until the intent
reaches a node, it will operate on defaults.

    

    

    From collaborations with some operators on autonomic networking
    projects: default behavior without policy must equal to "do
    nothing". But operators' views might change/evolve...

    Also, and this is based on feedback from the field (e.g. 3GPP SON),
    turning on autonomic functions, even with well-thought parameter
    settings, turns out most of the times in chaotic behaviors and
    severe performance degradations.

    A key element to consider here is how to "gracefully" start/ramp-up
    an autonomic network and aspects of coordination (cf
    draft-ciavaglia-anima-coordination-01).

    

    

      
 Of course, the semantics of
intent must be well-defined, but it's an extra layer.



      

        
If intent is formulated at an abstract level, how is an autonomic
network going to consume intent that may have drastically different
syntactical structures, and likely, different semantics associated with
those syntactical structures?


      

      
Agreed. We need a model, syntax and semantics.



      

        

        

          
I think the point is that AN *adds* a new mode of communication
between self-managing nodes, that doesn't exist in a strictly
north/south scenario. That is meant to reduce the amount of
north/south communication and reduce the amount of
centralized work. So there's a trade-off - more autonomic
messaging means less north-south messaging.

        

      

    

    

    north-south is a matter of convention / roles.

    if we consider PBM / intent, for me it implies a "vertical"
    (authority) relationship.

    if we consider a more "flat" system of collaborative entities, then
    let's have a look at MAS / DAI approaches and not reinvent (a
    squared) wheel.

    we may end up pushing more or less intelligence in the autonomic
    functions, but we also consider the scope of ANIMA to address
    "managed" networks (maybe not only this one), so it means some
    operators will expect certain behavior/performance from the
    networks. Intent is one way to inform the network what is expected.
    

    

    

      

        

          

        

        
There may indeed be a new mode of communication required.
However, if intent is not standardized, how does this work?


      

      
It doesn't. My only point is that intent is layered on top of an
underlying autonomic behaviour based on default behaviours. Actually,
that's why the question of intent was deferred in the WG charter.

    

    

    Your explanation is "valid", but I don't remember this was the
    reason used to consider intent FFS.

    Without rewriting history, we are facing now with the "heavy"
    design/conceptual discussions on the list, the effect of some
    decisions to exclude architectural analysis / investigations in the
    early phase of ANIMA...

    

    Laurent.

    

    

    

      

    Brian



      

        
I understand how unstructured and structured P2P networks
work. Let's assume this is a structured P2P network. In that case,
a DHT is used to assign ownership of each file to a peer. I fail to
understand how this works in an intent case. Files are passive,
intent is active. Please explain.

Regards,
John

On Sat, Mar 26, 2016 at 11:53 AM, Brian E Carpenter <
brian.e.carpenter@gmail.com> wrote:



        

          
Hi John,
On 27/03/2016 04:54, John Strassner wrote:


          

            

              
Correct, and there is also no presumption that individual nodes
have perfect knowledge of the topology. Of course the ACP
routing protocol will have perfect knowledge of the ACP topology,
but individual ASAs only know what they have gleaned from
discovery and what they have been told by Intent.

It is, I think, a very different world view than the SDN or SUPA or
NVO4 or NFV view.


            

            
Please note that I did NOT say anything about SDN, NFV, or SUPA
here. In fact, I'm confused as to why you both think I did.


          

          
I can't speak for Zongpeng, but I was triggered by the mention of
northbound and southbound APIs. There is no presumed direction in
Anima relationships, since in the limit a network is self-forming;
all relationships are peer-to-peer.

Of course, when we put Intent into the picture, there's a presumption
that Intent comes from some authority such as a NOC, but that's an
external semantic, not an assumption of the autonomic model.



          

            
I specifically said "There are plenty of southbound APIs (from a
management function to a device). There are very few northbound APIs."

Note the word "management function".

For the record, I don't think that either Phase 2 of NFV or the SDN 1.1
architecture have treated policy in any detail. Both architectures lack a
policy management component. I hope that ANIMA doesn't make the
same mistake.


          

          
Certainly, Intent needs a number of properties such as self-consistency
and stability that imply it is managed in some way.



          

            

              

                
However, in ANIMA network, perhaps we do not have a centralized
parsing node (controller), which can replace almost all the
communications between nodes.


              

            

            
Earlier I wrote about intent requiring a translation function. I never


          

          
said


          

            
that it was centralized or distributed. In the autonomic architectures I
have built, it was in fact a decentralized (agent-driven) function.
Furthermore, it did not "replace almost all the communications" - it
was used for compilation. There was, in fact, a separate distribution
mechanism (which I have also described), but that was limited to
"just" distribution.

I really don't understand what "replacing communications" means.


          

          
I think the point is that AN *adds* a new mode of communication between
self-managing nodes, that doesn't exist in a strictly north/south
scenario. That is meant to reduce the amount of north/south communication
and reduce the amount of centralized work. So there's a trade-off -
more autonomic messaging means less north-south messaging.

Best regards
   Brian



          

            
regards,
John

On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <
brian.e.carpenter@gmail.com> wrote:



            

              

                
However, in ANIMA network, perhaps we do not have a centralized parsing


              

              
node (controller), which can replace almost all the communications


            

          

          
between


          

            

              
nodes.

Correct, and there is also no presumption that individual nodes have
perfect
knowledge of the topology. Of course the ACP routing protocol will have
perfect
knowledge of the ACP topology, but individual ASAs only know what they


            

          

          
have


          

            

              
gleaned from discovery and what they have been told by Intent.

It is, I think, a very different world view than the SDN or SUPA or NVO4
or NFV view.

Regards
   Brian

On 26/03/2016 16:46, Duzongpeng wrote:


              

                
Hi John:

Thanks for your reply.

<jcs>
How is this different than running a script, or invoking an API? There


              

              
are plenty of southbound APIs (from a management function to a device).
There are very few northbound APIs. That's what the operators that I


            

          

          
have


          

            

              
talked to want.


              

                
</jcs>

I agree that intent belongs to northbound interface in the SDN network.


              

              
And ANIMA network perhaps can share the same northbound interface.


              

                
But, my understanding is that ANIMA network is a little different from


              

              
SDN network.


              

                
SDN controller will parse the intent, and transform to detailed


              

              
configurations of every device. It is a good model, and little
communications between nodes are needed.


              

                
Some network level parameters can be handled in SDN controller.

However, in ANIMA network, perhaps we do not have a centralized parsing


              

              
node (controller), which can replace almost all the communications


            

          

          
between


          

            

              
nodes.


              

                
So when a network operator is operating a ANIMA network, the network


              

              
level parameters are still needed to be flooded.


              

                
I mean the ANIMA interface is perhaps something like “northbound + part


              

              
of southboundâ€.


              

                
Best Regards
Zongpeng Du


From: John Strassner [mailto:strazpdj@gmail.com]
Sent: Saturday, March 26, 2016 9:41 AM
To: Duzongpeng; John Strassner
Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang;


              

              
anima@ietf.org; Jéferson Campos Nobre


              

                
Subject: Re: [Anima] Does Autonomic network need some intervention


              

              
beyond intent


              

                
Hi Zongpeng,

I think we have a misunderstanding here.


But I do not support that the intent writer should look every detail in


              

              
the catalog.


              

                
<jcs>
Most catalogs that I have seen have very few details. They instead tell


              

              
the user what services or products are available.


              

                
</jcs>

The intent writer may mainly care about what they want, but how to do


              

            

          

          
it


          

            

              
can be left to the autonomic network to deal with.


              

                
<jcs>
I agree completely.
</jcs>

Perhaps several solutions exist, and autonomic management system choose


              

              
one according to the abilities collected.


              

                
<jcs>
This is one of the reasons why I would like to see capabilities (and


              

              
needs) advertised.


              

                
</jcs>

...
I mean in current status, some of the Autonomic Functions should be


              

              
exposed to the network operator, and some of the Autonomic Functions


            

          

          
can be


          

            

              
hidden to the network operator


              

                
...
<jcs>
How is this different than running a script, or invoking an API? There


              

              
are plenty of southbound APIs (from a management function to a device).
There are very few northbound APIs. That's what the operators that I


            

          

          
have


          

            

              
talked to want.


              

                
regards,
John

On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng <duzongpeng@huawei.com


              

              
<mailto:duzongpeng@huawei.com>> wrote:


              

                
Hi John

Some personal understanding here. If any misunderstanding, please


              

              
correct me. Thanks.


              

                
<jcs2>
I would suggest a slight modification, to enable users of
intent (and even intent writers) to NOT have to have
detailed implementation information.

What if all of the Autonomic Functions, after they joined
the network, advertised both their capabilities as well as
their needs? We can put aside the latter (needs) as this
is more complex, but if the capabilities are advertised,
then the autonomic management system can assemble
those needs into a catalog. Then, the intent writer can
look at the catalog and write a declarative request to
use the functionality of the catalog.

I will look at your YANG model in a bit (sorry, I'm woefully
behind...).
< /jcs2>

I agree that autonomic node should have that self-advertisement


              

            

          

          
ability,


          

            

              
and autonomic management system should have this catalog.


              

                
But I do not support that the intent writer should look every detail in


              

              
the catalog.


              

                
The intent writer may mainly care about what they want, but how to do


              

            

          

          
it


          

            

              
can be left to the autonomic network to deal with.


              

                
Perhaps several solutions exist, and autonomic management system choose


              

              
one according to the abilities collected.


              

                
In the ideal, the network operator may communicate with the autonomic


              

              
network using network-level level intent and network-level report.


              

                
I mean in current status, some of the Autonomic Functions should be


              

              
exposed to the network operator, and some of the Autonomic Functions


            

          

          
can be


          

            

              
hidden to the network operator


              

                
Perhaps in future, when the network nodes become more intelligent, the


              

              
hidden part will increase.


              

                

Best regards
Zongpeng Du

From: Anima [mailto:anima-bounces@ietf.org<mailto:


              

            

          

          
anima-bounces@ietf.org>]


          

            

              
On Behalf Of John Strassner


              

                
Sent: Thursday, March 24, 2016 10:41 AM
To: Michael Behringer (mbehring); John Strassner
Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org<mailto:


              

              
anima@ietf.org>; Jéferson Campos Nobre


              

                
Subject: Re: [Anima] Does Autonomic network need some intervention


              

              
beyond intent


              

                
Hi Michael,

thanks for your reply. I do think that we largely agree, and would


              

            

          

          
enjoy


          

            

              
chatting during or (hopefully before) BA. Please see <jcs2>..</jcs2>.


            

          

          
Also


          

            

              
trying to unclutter, so I am eliding points that we agree on.


              

                

While *CIEs don’t like magic, they do understand the concept of a


              

              
default behaviour which they can influence if needed. Any reasonable
network management approach tries to define “default†behaviour and
“exceptionsâ€. The AN story fills the “default†part very nicely, and


            

          

          
*CIEs


          

            

              
get that.


              

                
<jcs2>
Agreed!
</jcs2>

The “parameter†versus “policy†discussion: My view is that we


              

            

          

          
shouldn’t


          

            

              
try to define exactly *what* intent allows and doesn’t allow, since


            

          

          
we’ve


          

            

              
heard from many sides that the boundary is not clear, and it may be
impossible to have a really clear, usable definition. So my point was:
Let’s focus on the distribution model: Intent is flooded to all nodes.
Let’s NOT do selective flooding, unicast Intent to specific nodes, etc,
because that would blur the boundary even worse.


              

                
<jcs2>
I like flooding intent to all nodes. In FOCALE (v1-3, and remember, v1


              

              
was 2006, almost 10 years ago), we used a message bus and pub-sub. In


            

          

          
v2,


          

            

              
we flooded; in v3, we went back to pub-sub because we wanted to do some
optimization and auditing. But I would be happy to try some type of
controlled flooding again.


              

                
</jcs2>
My suggestion: Intent is flooded to all nodes in a domain. The ASA


              

              
owners define the content of what they want to push in their bit of the
Intent. If we insist of full flooding, we encourage the right behaviour.
And if someone really puts into his Intent “node A: do this; node B: do


            

          

          
the


          

            

              
other; node C: do the third†it is awkward enough for people to take a


            

          

          
step


          

            

              
back and think whether they’re doing the right thing ☺


              

                
<jcs2>
I agree in principle with what you said. Plus, see above.
< /jcs2>
<jcs>
Do you mean that the intent writer will **know the difference**
between the fabric and each autonomic function? I hope not,
but perhaps I'm misunderstanding you...
< /jcs>
Not sure I understand your concern. Yes, in my model there is a section


              

              
for the fabric, and sections for the various functions. Seems a natural
structure?!? <jcs2>


              

                
The intent **writer** would have to know implementation
details in order to place everything in a single file. I would
rather not require this, and instead, have the intent engine
(e.g., compiler and associated logic) do this.
< /jcs2>
“Function ownerâ€: Yes, I’m suggesting to structure Intent around


              

              
Autonomic Functions. This is not optimal from a scientific point of


            

          

          
view,


          

            

              
but aligns very well with the way how things are organised. Probably we
need to discuss that in BA. But refer to my YANG model sent a few days


            

          

          
ago.


          

            

              

                
<jcs2>
I would suggest a slight modification, to enable users of
intent (and even intent writers) to NOT have to have
detailed implementation information.

What if all of the Autonomic Functions, after they joined
the network, advertised both their capabilities as well as
their needs? We can put aside the latter (needs) as this
is more complex, but if the capabilities are advertised,
then the autonomic management system can assemble
those needs into a catalog. Then, the intent writer can
look at the catalog and write a declarative request to
use the functionality of the catalog.

I will look at your YANG model in a bit (sorry, I'm woefully
behind...).
< /jcs2>


regards,
John

On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <


              

              
mbehring@cisco.com<mailto:mbehring@cisco.com>> wrote:


              

                
John,

I think we’re on the same page. Couple of points, top posting here,


              

              
otherwise it gets too messy.


              

                
While *CIEs don’t like magic, they do understand the concept of a


              

              
default behaviour which they can influence if needed. Any reasonable
network management approach tries to define “default†behaviour and
“exceptionsâ€. The AN story fills the “default†part very nicely, and


            

          

          
*CIEs


          

            

              
get that.


              

                
The “parameter†versus “policy†discussion: My view is that we


              

            

          

          
shouldn’t


          

            

              
try to define exactly *what* intent allows and doesn’t allow, since


            

          

          
we’ve


          

            

              
heard from many sides that the boundary is not clear, and it may be
impossible to have a really clear, usable definition. So my point was:
Let’s focus on the distribution model: Intent is flooded to all nodes.
Let’s NOT do selective flooding, unicast Intent to specific nodes, etc,
because that would blur the boundary even worse.


              

                
My suggestion: Intent is flooded to all nodes in a domain. The ASA


              

              
owners define the content of what they want to push in their bit of the
Intent. If we insist of full flooding, we encourage the right behaviour.
And if someone really puts into his Intent “node A: do this; node B: do


            

          

          
the


          

            

              
other; node C: do the third†it is awkward enough for people to take a


            

          

          
step


          

            

              
back and think whether they’re doing the right thing ☺


              

                
<jcs>
Do you mean that the intent writer will **know the difference**
between the fabric and each autonomic function? I hope not,
but perhaps I'm misunderstanding you...
</jcs>

Not sure I understand your concern. Yes, in my model there is a section


              

              
for the fabric, and sections for the various functions. Seems a natural
structure?!?


              

                
“Function ownerâ€: Yes, I’m suggesting to structure Intent around


              

              
Autonomic Functions. This is not optimal from a scientific point of


            

          

          
view,


          

            

              
but aligns very well with the way how things are organised. Probably we
need to discuss that in BA. But refer to my YANG model sent a few days


            

          

          
ago.


          

            

              

                
We should schedule some discussion time outside the ANIMA meeting


              

              
window...


              

                
Michael


From: John Strassner [mailto:strazpdj@gmail.com<mailto:


              

              
strazpdj@gmail.com>]


              

                
Sent: 22 March 2016 21:32
To: Michael Behringer (mbehring) <mbehring@cisco.com<mailto:


              

              
mbehring@cisco.com>>; John Strassner <strazpdj@gmail.com<mailto:
strazpdj@gmail.com>>


              

                
Cc: Duzongpeng <duzongpeng@huawei.com<mailto:duzongpeng@huawei.com>>;


              

              
anima@ietf.org<mailto:anima@ietf.org>; Laurent Ciavaglia <
laurent.ciavaglia@nokia.com<mailto:laurent.ciavaglia@nokia.com>>;
Jéferson Campos Nobre <jcnobre@inf.ufrgs.br<mailto:jcnobre@inf.ufrgs.br
;
Sheng Jiang <jiangsheng@huawei.com<mailto:jiangsheng@huawei.com>>


              

                
Subject: Re: [Anima] Does Autonomic network need some intervention


              

              
beyond intent


              

                
In general, I agree with Michael. I'd like to emphasize and expand on a


              

              
couple of key points:


              

                

First, ANIMA isn’t the only thing “influencing†or “steering†a


              

            

          

          
network,


          

            

              
at least not initially. There is still traditional config, plus all the
existing and emerging SDN models. My view has always been that ANIMA


            

          

          
sets a


          

            

              
“default†policy in a network, and all other means are more specific and
can override intent. This is described in RFC7575. Main point here:


            

          

          
ANIMA


          

            

              
Intent isn’t the only tool in a network.


              

                
<jcs>
Absolutely agree. Note that even in the Congress case I mentioned
earlier, it is NOT pushing config, and it is NOT viewed as the
"only" policy
</jcs>


Intent could perfectly well cover a high level policy such as “all


              

            

          

          
nodes


          

            

              
of type x do this; type y does thatâ€. (Which I think is what you mention
below.) When we started with Intent, the vision was to keep it very high
level, more along the line “do the right thing†☺


              

                
<jcs>
Agreed. Although to me, this means that we need to consider the types


              

            

          

          
of


          

            

              
actors that are using intent. For example, most of my CCIE and JCIE


            

          

          
friends


          

            

              
have no interest in using an intent language for pushing configuration.


            

          

          
On


          

            

              
the other hand, when I talk to operators and application developers,


            

          

          
they


          

            

              
would love to be able to use a high-level intent and have the intent


            

          

          
engine


          

            

              
take care of the details for them.


              

                
</jcs>

It has become clear that in today’s networking, people will want to


              

            

          

          
push


          

            

              
more “parameters†than “policyâ€, and I think we’ll just have to


            

          

          
acknowledge


          

            

              
that.


              

                
<jcs>
Sorry Michael, I don't follow here. What do you mean
by "parameters"?
</jcs>


Personally I think it’ll pan out like this: Intent will have sections


              

              
per autonomic function. An Intent parser will validate overall


            

          

          
correctness


          

            

              
(signature, etc), then segment the Intent into its bits: One part will


            

          

          
be


          

            

              
the autonomic fabric itself, then there will be bits for each autonomic
function.


              

                
<jcs>
Do you mean that the intent writer will **know the difference**
between the fabric and each autonomic function? I hope not,
but perhaps I'm misunderstanding you...
</jcs>


So the Intent parser on a node will grab intent, see there is some


              

              
Intent for autonomic function A, will see whether this is running


            

          

          
locally,


          

            

              
and if so, it’ll just pass that piece of Intent to that function.


              

                
<jcs>
Agreed.
</jcs>


In this way, a function owner defines both the intent itself, and how


              

            

          

          
it


          

            

              
is consumed by his/her function.


              

                
<jcs>
I'm not sure what you mean by "function owner". (Sorry to be
picky, I'm likely under-caffeinated.)

The above seemed like an autonomous decision, made by the parser.
I don't think that the intent writer should be cognizant of how many or
what types of autonomic functions are present.
</jcs>


The downside is that there is no “control†and people can do what they


              

              
want, in the worst case even push config, which is clearly NOT what


            

          

          
Intent


          

            

              
is meant to do.


              

                
<jcs>
And hence, the "HAL 2000" scenario. :-)
</jcs>


I’d hope that people don’t start writing Intent like “node 17: here is


              

              
your CLI; node 23: here is your CLIâ€; node 37: here is your CLIâ€. But


            

          

          
what


          

            

              
I describe above COULD be abused in this way. RFC7575 says this is NOT
Intent.


              

                
<jcs>
Absolutely agree!
</jcs>

 regards,
John



On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <


              

              
mbehring@cisco.com<mailto:mbehring@cisco.com>> wrote:


              

                
Hi Zongpeng,

Yes, this is a recurring question and topic. Let me give my view on two


              

              
levels:


              

                
First, ANIMA isn’t the only thing “influencing†or “steering†a


              

            

          

          
network,


          

            

              
at least not initially. There is still traditional config, plus all the
existing and emerging SDN models. My view has always been that ANIMA


            

          

          
sets a


          

            

              
“default†policy in a network, and all other means are more specific and
can override intent. This is described in RFC7575. Main point here:


            

          

          
ANIMA


          

            

              
Intent isn’t the only tool in a network.


              

                
Intent could perfectly well cover a high level policy such as “all


              

            

          

          
nodes


          

            

              
of type x do this; type y does thatâ€. (Which I think is what you mention
below.) When we started with Intent, the vision was to keep it very high
level, more along the line “do the right thing†☺  It has become clear


            

          

          
that


          

            

              
in today’s networking, people will want to push more “parameters†than
“policyâ€, and I think we’ll just have to acknowledge that.


              

                
Personally I think it’ll pan out like this: Intent will have sections


              

              
per autonomic function. An Intent parser will validate overall


            

          

          
correctness


          

            

              
(signature, etc), then segment the Intent into its bits: One part will


            

          

          
be


          

            

              
the autonomic fabric itself, then there will be bits for each autonomic
function.


              

                
So the Intent parser on a node will grab intent, see there is some


              

              
Intent for autonomic function A, will see whether this is running


            

          

          
locally,


          

            

              
and if so, it’ll just pass that piece of Intent to that function.


              

                
In this way, a function owner defines both the intent itself, and how


              

            

          

          
it


          

            

              
is consumed by his/her function.


              

                
The downside is that there is no “control†and people can do what they


              

              
want, in the worst case even push config, which is clearly NOT what


            

          

          
Intent


          

            

              
is meant to do. I’d hope that people don’t start writing Intent like


            

          

          
“node


          

            

              
17: here is your CLI; node 23: here is your CLIâ€; node 37: here is your
CLIâ€. But what I describe above COULD be abused in this way. RFC7575


            

          

          
says


          

            

              
this is NOT Intent.


              

                
In other words: Let’s not be too prescriptive, and hope that folks will


              

              
do more or less reasonable things.


              

                
I’ve been meaning to comment on the Intent drafts and possibly


              

              
contribute, I’m just struggling with priorities at the moment (as you


            

          

          
can


          

            

              
see by my relative silence over the past weeks).


              

                
Michael



From: Anima [mailto:anima-bounces@ietf.org<mailto:


              

            

          

          
anima-bounces@ietf.org>]


          

            

              
On Behalf Of Duzongpeng


              

                
Sent: 16 March 2016 02:38
To: anima@ietf.org<mailto:anima@ietf.org>
Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com<mailto:


              

              
laurent.ciavaglia@nokia.com>>; Jéferson Campos Nobre <


            

          

          
jcnobre@inf.ufrgs.br


          

            

              
<mailto:jcnobre@inf.ufrgs.br>>; Sheng Jiang <jiangsheng@huawei.com


            

          

          
<mailto:


          

            

              
jiangsheng@huawei.com>>


              

                
Subject: [Anima] Does Autonomic network need some intervention beyond


              

              
intent


              

                
Hello, everyone

         I understand that Autonomic network should be able to work


              

            

          

          
with


          

            

              
as less configuration as possible. In the ideal case, an Autonomic


            

          

          
network


          

            

              
should be able to work well while only intent is needed from human
operators.


              

                
         However, I wonder whether Autonomic network may need some


              

              
intervention from human operators through some low-level policies beyond
intents.


              

                
         For example, in the


              

              
https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it


            

          

          
is


          

            

              
suggested that the prefix lengths for the CSG, ASG, RSG (different


            

          

          
roles in


          

            

              
IP RAN) can be assigned as an "intent". These configuration parameters


            

          

          
need


          

            

              
to be distributed in the autonomic domain to influence the detail
configurations on each autonomic node.


              

                
         As intent is described as an abstract, declarative, high-level


              

              
policy used to operate an autonomic domain, such as an enterprise


            

          

          
network.


          

            

              
Perhaps we can rename the “intent†in the “prefix-management ID†to some
other words.


              

                
Is “ASA parameters" ok here?

I think some characteristics for this kind of low-level policies are

1.       Network level parameters, need to be distributed by GRASP and


              

              
ACP


              

                
2.       Relatively static compared to intent, perhaps only needed to


              

            

          

          
be


          

            

              
configured once


              

                
3.       Unavoidable in current network environment, mostly for


              

              
establishing the network infrastructure


              

                
4.       Rarely need coordinations with others parameters, usually is


              

              
closed related to one specific autonomic function


              

                
Welcome for comment. Your suggestions will be appreciated.


Best Regards
Zongpeng Du

_______________________________________________
Anima mailing list
Anima@ietf.org<mailto:Anima@ietf.org>
https://www.ietf.org/mailman/listinfo/anima



--
regards,
John



--
regards,
John



--
regards,
John



_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima



              

              



            

            



          

          



        

        



      

      



    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------020905070204040504040807--


From nobody Tue Mar 29 09:44:03 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0262A12DA82 for ; Tue, 29 Mar 2016 09:44:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.52
X-Spam-Level: 
X-Spam-Status: No, score=-14.52 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_KAM_HTML_FONT_INVALID=0.01, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4_xVQaFWC3eQ for ; Tue, 29 Mar 2016 09:43:57 -0700 (PDT)
Received: from rcdn-iport-4.cisco.com (rcdn-iport-4.cisco.com [173.37.86.75]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9968912D9EC for ; Tue, 29 Mar 2016 09:34:48 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=104380; q=dns/txt; s=iport; t=1459269288; x=1460478888; h=from:to:cc:subject:date:message-id:references: in-reply-to:mime-version; bh=RfPHPiqNEVyUxPI1ZICm3Qz6OHV18an1ZcwyoO9wF0g=; b=Z/wXheAiqGqnhMxz32/pljwXH4fUXZULlfG/iq2rFmeP5FAOMdLljMyR OVNciv/kQRAPf3YFfuyS4hZ8YQdZkjziETaIAFcf23RmxZd751ybzAeRN ExwEyZKwoqrYp4b/0ktUaV1FLtJ9mo7DGOqBbxiedntA3jGTEtP9f8qQh k=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0ATAgBerfpW/51dJa1dgmJMU30GryWLU?= =?us-ascii?q?QENgW0DFwEJhWwCHIEYOBQBAQEBAQEBZCeEQQEBAQQBAQEXCQpBCwwEAgEIEQQ?= =?us-ascii?q?BASEBBgMCAgIfBgsUCQgCBAENBQiICgMSDq9li3ENhGUBAQEBAQEBAQEBAQEBA?= =?us-ascii?q?QEBAQEBAQEVhh2DRX+CP4FfAQEzFwmCSoJWBZJpJYQtMQGFcYYggW6PFIc7h1M?= =?us-ascii?q?BDw8BAUKDZWwBhwMHFwcYfgEBAQ?=
X-IronPort-AV: E=Sophos; i="5.24,411,1454976000"; d="scan'208,217"; a="87763003"
Received: from rcdn-core-6.cisco.com ([173.37.93.157]) by rcdn-iport-4.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 29 Mar 2016 16:34:47 +0000
Received: from XCH-ALN-010.cisco.com (xch-aln-010.cisco.com [173.36.7.20]) by rcdn-core-6.cisco.com (8.14.5/8.14.5) with ESMTP id u2TGYkP9021695 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Tue, 29 Mar 2016 16:34:47 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-010.cisco.com (173.36.7.20) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Tue, 29 Mar 2016 11:34:46 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Tue, 29 Mar 2016 11:34:45 -0500
From: "Michael Behringer (mbehring)" 
To: Laurent Ciavaglia , Brian E Carpenter , John Strassner , Duzongpeng 
Thread-Topic: Another use case for Intent: "Freeze network enrolment"
Thread-Index: AQHRidMDO/a05lCyz0ONkyiIPytX+J9wk74w
Date: Tue, 29 Mar 2016 16:34:45 +0000
Message-ID: <58704b9ce2cb43c0b07d9aac8b284187@XCH-RCD-006.cisco.com>
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56FAA4B7.9010305@alcatel-lucent.com>
In-Reply-To: <56FAA4B7.9010305@alcatel-lucent.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: multipart/alternative; boundary="_000_58704b9ce2cb43c0b07d9aac8b284187XCHRCD006ciscocom_"
MIME-Version: 1.0
Archived-At: 
Cc: =?utf-8?B?SsOpZmVyc29uIENhbXBvcyBOb2JyZQ==?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 16:44:02 -0000

--_000_58704b9ce2cb43c0b07d9aac8b284187XCHRCD006ciscocom_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_58704b9ce2cb43c0b07d9aac8b284187XCHRCD006ciscocom_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_58704b9ce2cb43c0b07d9aac8b284187XCHRCD006ciscocom_--


From nobody Tue Mar 29 09:45:24 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AB15112DA41 for ; Tue, 29 Mar 2016 09:45:23 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2-CZVfrLky0q for ; Tue, 29 Mar 2016 09:45:22 -0700 (PDT)
Received: from alln-iport-5.cisco.com (alln-iport-5.cisco.com [173.37.142.92]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CD5A112D99B for ; Tue, 29 Mar 2016 09:36:01 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=626; q=dns/txt; s=iport; t=1459269361; x=1460478961; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=Pm7i0vclc47K6bZT46nGYlAoiLKKb5Zq1Sfw/0MMSKM=; b=NuHTJA8zybkj8pUdMRmFkLfcH7YSCcTla6ftjREBfYA1NZbe4jP7aZ7G Jzd+wafYZ1pGenR6hP+Osr1ASbPbv8eQlf1Em4clVSYTNIO+qN8lXUjpa ik+Zy0fLLJPzdia/dih+1nmlXE0lF5v8Jkm+eenp5mQ91PrHBYfeI+4Nm o=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0APAgCnrvpW/4QNJK1dgy6BUAa6eAENg?= =?us-ascii?q?XCGDQKBNTgUAQEBAQEBAWQnhEEBAQEEOj8MBAIBCBEEAQEfCQcyFAkIAgQOBQi?= =?us-ascii?q?IH8BXAQEBAQEBAQEBAQEBAQEBAQEBAQEBFYYdhESKEwEEl2wBjX+BV409jw4BH?= =?us-ascii?q?gEBQoNlbIdBfgEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,411,1454976000"; d="scan'208";a="253066436"
Received: from alln-core-10.cisco.com ([173.36.13.132]) by alln-iport-5.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 29 Mar 2016 16:35:49 +0000
Received: from XCH-ALN-008.cisco.com (xch-aln-008.cisco.com [173.36.7.18]) by alln-core-10.cisco.com (8.14.5/8.14.5) with ESMTP id u2TGZnOW028992 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Tue, 29 Mar 2016 16:35:49 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-008.cisco.com (173.36.7.18) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Tue, 29 Mar 2016 11:35:48 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Tue, 29 Mar 2016 11:35:48 -0500
From: "Michael Behringer (mbehring)" 
To: "Joel M. Halpern" 
Thread-Topic: [Anima] Another use case for Intent: "Freeze network enrolment"
Thread-Index: AQHRhgPpzIwvl+WeaEyrb+KXVUZG1J9qK3OAgAES5wCABORlsIAAxzYA//+uO0CAAFgggP//taOQ
Date: Tue, 29 Mar 2016 16:35:48 +0000
Message-ID: 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FAA188.2050901@joelhalpern.com> <9c96c405709e4920bcb0e9c7483d1f90@XCH-RCD-006.cisco.com> <56FAA6DD.30403@joelhalpern.com>
In-Reply-To: <56FAA6DD.30403@joelhalpern.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Cc: Laurent Ciavaglia , "anima@ietf.org" 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 16:45:23 -0000

> -----Original Message-----
> From: Joel M. Halpern [mailto:jmh@joelhalpern.com]
> Sent: 29 March 2016 18:02
> To: Michael Behringer (mbehring) 
> Cc: Laurent Ciavaglia ; anima@ietf.org
> Subject: Re: [Anima] Another use case for Intent: "Freeze network
> enrolment"
>=20
> I can buy the low change rate.  In part because if we find that flooding =
is
> awkward, it is easy enough to add a new mechanism later.
>=20
> But I find the assumption that being high level will somehow make intents
> small to be very unlikely.

Do you have an example?=20

Michael


From nobody Tue Mar 29 09:54:30 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 64CD612DCFF for ; Tue, 29 Mar 2016 09:54:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.702
X-Spam-Level: 
X-Spam-Status: No, score=-2.702 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HuNrUqpdafLo for ; Tue, 29 Mar 2016 09:54:26 -0700 (PDT)
Received: from mailb2.tigertech.net (mailb2.tigertech.net [208.80.4.154]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 8E16912D9B2 for ; Tue, 29 Mar 2016 09:45:51 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by mailb2.tigertech.net (Postfix) with ESMTP id 5B061680037; Tue, 29 Mar 2016 09:45:51 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1459269951; bh=o8bun9BzEMhoa8dUJo/USLBO3MPf9qmVFlASTc3UDCA=; h=Subject:To:References:Cc:From:Date:In-Reply-To:From; b=IEQg6j6jvuhE9UOgw5Mc+dsYYnxCGvgxTpKDD0JC4sfaUxHaddMzhLydcngzG57eO WDQucs1kTS+Uz4m8qS2aCTlXgh51ZB10c59xOwONWjIpExmDRkf5alumolkl2HS55K lhsb2ieRfiR8Figp0VTzy2IrdABzr0ouokS+lsns=
X-Virus-Scanned: Debian amavisd-new at b2.tigertech.net
Received: from Joels-MacBook-Pro.local (209-255-163-147.ip.mcleodusa.net [209.255.163.147]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by mailb2.tigertech.net (Postfix) with ESMTPSA id C64681C00D7; Tue, 29 Mar 2016 09:45:50 -0700 (PDT)
To: "Michael Behringer (mbehring)" 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FAA188.2050901@joelhalpern.com> <9c96c405709e4920bcb0e9c7483d1f90@XCH-RCD-006.cisco.com> <56FAA6DD.30403@joelhalpern.com> 
From: "Joel M. Halpern" 
Message-ID: <56FAB136.5010803@joelhalpern.com>
Date: Tue, 29 Mar 2016 12:45:42 -0400
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.1
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Laurent Ciavaglia , "anima@ietf.org" 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 16:54:28 -0000

There are two drivers here.

First, when I look at the history of configs, they started small, and 
then got large, evne when the number of interfaces and neighbors did not 
grow that fast (and the neighbor count went up much more than we expected.)

Second, I look at the growth of topics on which people have policy.

The result is that I would strongly prefer a design which coped with a a 
large number of roles, ASAs, policy parameters, and other aspects.

No, I can't prove it.  But looking at the hsitory of protocol design, I 
find that we have made a mess when we assume things will be small, and 
generally done reasonably when we assume things will grow beyond our 
expectations.

Yours,
Joel


On 3/29/16 12:35 PM, Michael Behringer (mbehring) wrote:
>> -----Original Message-----
>> From: Joel M. Halpern [mailto:jmh@joelhalpern.com]
>> Sent: 29 March 2016 18:02
>> To: Michael Behringer (mbehring) 
>> Cc: Laurent Ciavaglia ; anima@ietf.org
>> Subject: Re: [Anima] Another use case for Intent: "Freeze network
>> enrolment"
>>
>> I can buy the low change rate.  In part because if we find that flooding is
>> awkward, it is easy enough to add a new mechanism later.
>>
>> But I find the assumption that being high level will somehow make intents
>> small to be very unlikely.
>
> Do you have an example?
>
> Michael
>


From nobody Tue Mar 29 12:26:51 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 44B4712DF98 for ; Tue, 29 Mar 2016 12:26:50 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.911
X-Spam-Level: 
X-Spam-Status: No, score=-1.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tSLdGeT1Cv08 for ; Tue, 29 Mar 2016 12:26:48 -0700 (PDT)
Received: from tuna.sandelman.ca (tuna.sandelman.ca [209.87.249.19]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9E19912E204 for ; Tue, 29 Mar 2016 11:52:56 -0700 (PDT)
Received: from sandelman.ca (obiwan.sandelman.ca [IPv6:2607:f0b0:f:2::247]) by tuna.sandelman.ca (Postfix) with ESMTP id 7449E203AD for ; Tue, 29 Mar 2016 14:56:00 -0400 (EDT)
Received: from obiwan.sandelman.ca (localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id 7338863755 for ; Tue, 29 Mar 2016 14:52:55 -0400 (EDT)
From: Michael Richardson 
To: "anima\@ietf.org" 
In-Reply-To: <58704b9ce2cb43c0b07d9aac8b284187@XCH-RCD-006.cisco.com>
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56FAA4B7.9010305@alcatel-lucent.com> <58704b9ce2cb43c0b07d9aac8b284187@XCH-RCD-006.cisco.com>
X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2
X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m
Archived-At: 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 19:26:50 -0000

--=-=-=
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable


see inline

Michael Behringer (mbehring)  wrote:
    > When I say =E2=80=9Cfreeze=E2=80=9D I mean: All the nodes that the op=
erator expected
    > to join, have joined; the next ones will come in 6 months. Until then,
    > I want to disable the enrolment function.

Okay, so tell the registrar to stop approving new nodes.

Does that need to be network-wide?
If you really mean that you want to turn off the new node join helpers,
then let's be clear about that.

Maybe the registrar should emit an intent that says, "Willing to join nodes"
when "unfrozen", and that is what causes nodes that can perform the helper
function to enable it.

{it would be great if you'd edit the trail of the message to remove the
content which isn't relevant. Google and the mail archive would thank you}

=2D-
Michael Richardson , Sandelman Software Works
 -=3D IPv6 IoT consulting =3D-




--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBVvrPBICLcPvd0N1lAQLX6ggAlxmMft2NZp6yb9XegAXBUacFDhm/SiT+
/UonQFUhRqZNI1XIDSCjJof+EzJgzi0BqsUNyFQ6VDldh3N6BZEQ94wYwnHmZV2O
bm0b9jn4Hr/xLONuly7cqz0cKCcukpnELSf8/QAIXEw89F0VWM0W1OY3jhem5U2u
kg7xaYKX4drLzIskfkWlifbt3JofBk5CUsg+mUzKOrbm6NBptjWjWJadLutIAqHM
ul781EZ6wcWdZXPMAA+BsXU22YS3IeAp3mm5CkGSRxEYQGu525OYpsDxCgfhXb/A
sni3bk9K8EIK0EYdIsdMAB6WYJtwAB9OGj6eXGImBQOSgAvjqAgi3w==
=wJhJ
-----END PGP SIGNATURE-----
--=-=-=--


From nobody Tue Mar 29 12:54:13 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8F52D12DDF9 for ; Tue, 29 Mar 2016 12:54:11 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JXtAxuwY1D9R for ; Tue, 29 Mar 2016 12:54:10 -0700 (PDT)
Received: from mail-pa0-x235.google.com (mail-pa0-x235.google.com [IPv6:2607:f8b0:400e:c03::235]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id ABDF212E081 for ; Tue, 29 Mar 2016 12:17:35 -0700 (PDT)
Received: by mail-pa0-x235.google.com with SMTP id zm5so21380909pac.0 for ; Tue, 29 Mar 2016 12:17:35 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=M7pNHwU/NigPVd0UgY7ysGkPcTp/ZxElO6530mWfsbM=; b=N/IEslqLxTTobKENbCHeI+RE+MbSWv3Af9Rgei1zkz/572W7Ypz6w1MnW8sTIzlVdm LenlbA7ozGrLqNOKB9S/xcgRKmP2GqSvk2riBoTpN39UUlOUaWo/Q3RXzL+0w3SrM+y4 qUhFOsNHdCdIYRrz6W3yRAzN+WjeZxEAvG1fLF4l8/aORaJ5wN72LlKQ7oQ4a02ajGAK AFSVoN2hmuI7KfFrwDR7VCAxG7Xj8pE6vAuis3DtBPHI2AOjGyckoP5aUXJKyNEwdeh2 VYabQ3U1tYB6WV2kPlmzta8xFNb+roh9ZntQNl2TMRPwTX+BIGZl0EuQdJs/YWB9+FlU jSMg==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=M7pNHwU/NigPVd0UgY7ysGkPcTp/ZxElO6530mWfsbM=; b=H3SBoCOxAXzMY83Bh1kbnJ0vIWzDf2A+5GtnBl+yDo/xCWL6NAcQmYLZUFOB1kOK4v FCslJvhJbwjv1scHs0bS0XnoPDMlViCQphUMDHnRzhTS9Dgz96Oz+eVMOBSKtnctkejz S9Zb5nWVFpyRllrswgv7GwZmTHyoCU5M61oMdH7sJoiGPk1ieSQha0YqRu7pZl2W2Jjk PHcIdUr7VY9NiBm+uUXwriNLmyhsb56SpqNqTrURcp5ZbLreGnFbvQpwYYDbsU5SmJ9c f2FEDVC2exafzl1f0N/TqtnlyJ93HIivSRDOsT6AFhudkPaBTEU0kn8DKjpOXjW4TJb+ acgw==
X-Gm-Message-State: AD7BkJLs/2UVWufEMec0wrDfsfYD0NZlIb3OmaZvkknR+hZ3rASsHkDmMeZEBO5sTT5X5w==
X-Received: by 10.66.167.145 with SMTP id zo17mr6047363pab.73.1459279055325; Tue, 29 Mar 2016 12:17:35 -0700 (PDT)
Received: from ?IPv6:2406:e007:5ec7:1:28cc:dc4c:9703:6781? ([2406:e007:5ec7:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id ql1sm156975pac.24.2016.03.29.12.17.32 (version=TLSv1/SSLv3 cipher=OTHER); Tue, 29 Mar 2016 12:17:34 -0700 (PDT)
To: "Michael Behringer (mbehring)" , "anima@ietf.org" 
References: <77a801fd616449f0bc4c8a664286b9cc@XCH-RCD-006.cisco.com> <56F5DE90.6010509@gmail.com> <405bab40b6694cc5893f2cd89ef8e913@XCH-RCD-006.cisco.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56FAD4D0.3080009@gmail.com>
Date: Wed, 30 Mar 2016 08:17:36 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <405bab40b6694cc5893f2cd89ef8e913@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: Re: [Anima] 7.2 Specific ASAs [was Reference Model Issues]
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 19:54:11 -0000

On 29/03/2016 22:02, Michael Behringer (mbehring) wrote:
>> -----Original Message-----
>> From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
>> Sent: 26 March 2016 01:58
>> To: Michael Behringer (mbehring) ; anima@ietf.org
>> Subject: 7.2 Specific ASAs [was Reference Model Issues]
>>
>> On 25/03/2016 06:52, Michael Behringer (mbehring) wrote:
>> ...
>>> 7.2 Specific ASAs [for the Enrolment Process]
>>> - should explain MUST implement
>>
>> And that they MUST be present in all nodes by default. In particular the
>> Proxy ASA MUST discover whether it needs to be active (see discussion
>> yesterday). I suppose the same applies to the Registrar ASA but possibly that
>> needs to be configured "on" by the NOC?
> 
> Yes, agree, that's what I meant. Switching the proxy ASA off if not needed is an optimisation, in my view. (Also, once enrolled, you can switch off the enrolment ASA of course). 
> 
> The registrar is effectively a policy definition point (PDP), thus MUST be set up manually. 

Yes, that seems reasonable in a "professionally managed network" which is the
Anima target. I wouldn't exclude an election process in a completely unmanaged
network, but that's out of scope here.

   Brian

> 
> That still needs to be explained better in the draft, was what I was trying to say :-) 
> 
> Michael
> 


From nobody Tue Mar 29 13:06:21 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CCB0C12E225 for ; Tue, 29 Mar 2016 13:06:19 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uOtcyGYLu92o for ; Tue, 29 Mar 2016 13:06:16 -0700 (PDT)
Received: from mail-pf0-x22c.google.com (mail-pf0-x22c.google.com [IPv6:2607:f8b0:400e:c00::22c]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9C44B12E1D6 for ; Tue, 29 Mar 2016 12:29:36 -0700 (PDT)
Received: by mail-pf0-x22c.google.com with SMTP id e128so1154443pfe.3 for ; Tue, 29 Mar 2016 12:29:36 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=moPSVRKI1C1zMKL7tVbi0YiqSbLPFDDLhG5WLGmECNg=; b=gW0zqdRSwlEP4qJYp6eKCSc+U6+4u999DLnsITniwLbWTtrLA9XQ6Uv45suiT3uliY ONfPo8AlL4NB7uNq3CKdmkLNpXA4f25zwVmrEwTvKB0IbPtDkCuk6RSMiaWqcuIhQAcs ebYfV3x+42feL8lTkLTTfH0T47Po3JEaEzHRHXrPlPQTZ25oizgRAU1Dj1gPnQYGEuw5 Rz7P8zID8shwQ3Q7lOy6YBWWRfovAi41BDWBt6MXqDe11yvtZUxK3cTAeVb3B/anUxdK lokYuG5FiSFsuYJNRzE0JxrAKm0oQmALGqKE9i41T4IrGcLoTz7Cwzk7Ndn48R6f1d40 soow==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=moPSVRKI1C1zMKL7tVbi0YiqSbLPFDDLhG5WLGmECNg=; b=gMLGf9tLJXgAum4RN5duzn6I6eIVE4PjOJMiPr6rZGXcXio3pWJ/ajeVOFgEKRbOs7 0sEJjU697VnzpyhssJNbUZ2QBSDLQfqkGJDsKwcHOqfJ7el2s1qt0viGArydgoh32ppq cSbnz7KPuPiqqSbzHooFUA3m9geF9NGa+IqKasML3q/D7EdnCGblr6iuSxb2Ch5fHxqT HPlE9ifTFR8CnQ/49eaEMJRc37z2l1El0L7OjXGfVCzvJsoI/H1v4OfSXR5fdRxQyDpY EomBTrCm7hsg+yxq1JrxNPXcQIjbwqaROVXj66lCNIUWGUEj5qa3uHuEuhGJNxjih+0+ B0kg==
X-Gm-Message-State: AD7BkJJkHoqifMK4MF2E4CVTFkZoLTQR23E3yboazasWqLb7m5Wl4qeOyh1C4QCQt5HR9Q==
X-Received: by 10.98.74.209 with SMTP id c78mr6188174pfj.90.1459279776157; Tue, 29 Mar 2016 12:29:36 -0700 (PDT)
Received: from ?IPv6:2406:e007:5ec7:1:28cc:dc4c:9703:6781? ([2406:e007:5ec7:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id q20sm179032pfi.63.2016.03.29.12.29.31 (version=TLSv1/SSLv3 cipher=OTHER); Tue, 29 Mar 2016 12:29:34 -0700 (PDT)
To: Laurent Ciavaglia , John Strassner 
References:  <701d351267344c21b3ea053aa5b15728@XCH-RCD-006.cisco.com>  <8368235337ea4adfab5c428d57c3bfb2@XCH-RCD-006.cisco.com>     <56F60CE4.9040901@gmail.com>  <56F6DAA1.8000206@gmail.com>  <56F82FDD.7090405@gmail.com> <56FAAD7B.6080305@alcatel-lucent.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56FAD79E.2020504@gmail.com>
Date: Wed, 30 Mar 2016 08:29:34 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56FAAD7B.6080305@alcatel-lucent.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Archived-At: 
Cc: Duzongpeng , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "Michael Behringer \(mbehring\)" , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Does Autonomic network need some intervention beyond intent
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 20:06:19 -0000

Hi Laurent, comments in line:
On 30/03/2016 05:29, Laurent Ciavaglia wrote:
> Hello,
>=20
>=20
>> Indeed not. An ASA will have some default behaviour that amounts to ha=
ving
>> default settings for intent, but by definition intent comes from human=
s.
>> (One qualification: if the intent that reaches a given node is inconsi=
stent,
>> then resolving that inconsistency might be autonomic.)
>=20
> Brian: could you explain what you mean here, especially "resolving that=
 inconsistency might be autonomic"?
> I guess you mean that ASAs/nodes can autonomously detect inconsistency =
in intents and (try to) solve this inconsistency (on
> their own or via collaboration)...?

Yes. Actually if we allow more than one source of intent, this is essenti=
al.

>=20
>>
>>>> Of course, when we put Intent into the picture, there's a presumptio=
n
>>>> that Intent comes from some authority such as a NOC, but that's an
>>>> external semantic, not an assumption of the autonomic model.
>>> Well, no - if the autonomic system is going to use intent, or at leas=
t
>>> respond to it, then it must understand those external semantics.
>>> This is even more important if the network is supposed to form
>>> according to one or more intent rules.
>> Right, but what I meant is that the network will operate perfectly wel=
l
>> if no intent is sent, by using defaults. That will always happen to th=
e
>> extent that intent distribution will take a finite time. Until the int=
ent
>> reaches a node, it will operate on defaults.
>=20
> From collaborations with some operators on autonomic networking project=
s: default behavior without policy must equal to "do
> nothing". But operators' views might change/evolve...

That's certainly a possibility. I'm not sure it is always wise though.
For example, if a network is fragmented by a natural disaster, the fragme=
nts
might not receive Intent for several days.

> Also, and this is based on feedback from the field (e.g. 3GPP SON), tur=
ning on autonomic functions, even with well-thought
> parameter settings, turns out most of the times in chaotic behaviors an=
d severe performance degradations.
> A key element to consider here is how to "gracefully" start/ramp-up an =
autonomic network and aspects of coordination (cf
> draft-ciavaglia-anima-coordination-01).

Agreed. Obviously the defaults must be conservative and fail-safe.

>=20
>>   Of course, the semantics of
>> intent must be well-defined, but it's an extra layer.
>>
>>> If intent is formulated at an abstract level, how is an autonomic
>>> network going to consume intent that may have drastically different
>>> syntactical structures, and likely, different semantics associated wi=
th
>>> those syntactical structures?
>> Agreed. We need a model, syntax and semantics.
>>
>>>> I think the point is that AN *adds* a new mode of communication
>>>> between self-managing nodes, that doesn't exist in a strictly
>>>> north/south scenario. That is meant to reduce the amount of
>>>> north/south communication and reduce the amount of
>>>> centralized work. So there's a trade-off - more autonomic
>>>> messaging means less north-south messaging.
>=20
> north-south is a matter of convention / roles.
> if we consider PBM / intent, for me it implies a "vertical" (authority)=
 relationship.
> if we consider a more "flat" system of collaborative entities, then let=
's have a look at MAS / DAI approaches and not reinvent
> (a squarred) wheel.
> we may end up pushing more or less intelligence in the autonomic functi=
ons, but we also consider the scope of ANIMA to address
> "managed" networks (maybe not only this one), so it means some operator=
s will expect certain behavior/performance from the
> neworks. Intent is one way to inform the network what is expected.

Sure. But always remember the disaster-recovery scenario where we are for=
ced
back to defaults.

>=20
>>> There may indeed be a new mode of communication required.
>>> However, if intent is not standardized, how does this work?
>> It doesn't. My only point is that intent is layered on top of an
>> underlying autonomic behaviour based on default behaviours. Actually,
>> that's why the question of intent was deferred in the WG charter.
>=20
> Your explanation is "valid", but I don't remember this was the reason u=
sed to consider intent FFS.

OK, maybe that was just in my brain. Another aspect that Intent was incom=
pletely
understood.

> Without rewriting history, we are facing now with the "heavy" design/co=
nceptual discussions on the list, the effect of some
> decisions to exclude architectural analysis / investigations in the ear=
ly phase of ANIMA...

I maintain that these are issues of ASA design, not infrastructure design=
=2E We always
knew that the infrastructure has to distribute Intent.

Regards
   Brian

> Laurent.
>=20
>=20
>>
>>      Brian
>>
>>> I understand how unstructured and structured P2P networks
>>> work. Let's assume this is a structured P2P network. In that case,
>>> a DHT is used to assign ownership of each file to a peer. I fail to
>>> understand how this works in an intent case. Files are passive,
>>> intent is active. Please explain.
>>>
>>> Regards,
>>> John
>>>
>>> On Sat, Mar 26, 2016 at 11:53 AM, Brian E Carpenter <
>>> brian.e.carpenter@gmail.com> wrote:
>>>
>>>> Hi John,
>>>> On 27/03/2016 04:54, John Strassner wrote:
>>>>>> Correct, and there is also no presumption that individual nodes
>>>>>> have perfect knowledge of the topology. Of course the ACP
>>>>>> routing protocol will have perfect knowledge of the ACP topology,
>>>>>> but individual ASAs only know what they have gleaned from
>>>>>> discovery and what they have been told by Intent.
>>>>>>
>>>>>> It is, I think, a very different world view than the SDN or SUPA o=
r
>>>>>> NVO4 or NFV view.
>>>>> Please note that I did NOT say anything about SDN, NFV, or SUPA
>>>>> here. In fact, I'm confused as to why you both think I did.
>>>> I can't speak for Zongpeng, but I was triggered by the mention of
>>>> northbound and southbound APIs. There is no presumed direction in
>>>> Anima relationships, since in the limit a network is self-forming;
>>>> all relationships are peer-to-peer.
>>>>
>>>> Of course, when we put Intent into the picture, there's a presumptio=
n
>>>> that Intent comes from some authority such as a NOC, but that's an
>>>> external semantic, not an assumption of the autonomic model.
>>>>
>>>>> I specifically said "There are plenty of southbound APIs (from a
>>>>> management function to a device). There are very few northbound API=
s."
>>>>>
>>>>> Note the word "management function".
>>>>>
>>>>> For the record, I don't think that either Phase 2 of NFV or the SDN=
 1.1
>>>>> architecture have treated policy in any detail. Both architectures =
lack a
>>>>> policy management component. I hope that ANIMA doesn't make the
>>>>> same mistake.
>>>> Certainly, Intent needs a number of properties such as self-consiste=
ncy
>>>> and stability that imply it is managed in some way.
>>>>
>>>>>>> However, in ANIMA network, perhaps we do not have a centralized
>>>>>>> parsing node (controller), which can replace almost all the
>>>>>>> communications between nodes.
>>>>> Earlier I wrote about intent requiring a translation function. I ne=
ver
>>>> said
>>>>> that it was centralized or distributed. In the autonomic architectu=
res I
>>>>> have built, it was in fact a decentralized (agent-driven) function.=

>>>>> Furthermore, it did not "replace almost all the communications" - i=
t
>>>>> was used for compilation. There was, in fact, a separate distributi=
on
>>>>> mechanism (which I have also described), but that was limited to
>>>>> "just" distribution.
>>>>>
>>>>> I really don't understand what "replacing communications" means.
>>>> I think the point is that AN *adds* a new mode of communication betw=
een
>>>> self-managing nodes, that doesn't exist in a strictly north/south
>>>> scenario. That is meant to reduce the amount of north/south communic=
ation
>>>> and reduce the amount of centralized work. So there's a trade-off -
>>>> more autonomic messaging means less north-south messaging.
>>>>
>>>> Best regards
>>>>     Brian
>>>>
>>>>> regards,
>>>>> John
>>>>>
>>>>> On Fri, Mar 25, 2016 at 9:15 PM, Brian E Carpenter <
>>>>> brian.e.carpenter@gmail.com> wrote:
>>>>>
>>>>>>> However, in ANIMA network, perhaps we do not have a centralized p=
arsing
>>>>>> node (controller), which can replace almost all the communications=

>>>> between
>>>>>> nodes.
>>>>>>
>>>>>> Correct, and there is also no presumption that individual nodes ha=
ve
>>>>>> perfect
>>>>>> knowledge of the topology. Of course the ACP routing protocol will=
 have
>>>>>> perfect
>>>>>> knowledge of the ACP topology, but individual ASAs only know what =
they
>>>> have
>>>>>> gleaned from discovery and what they have been told by Intent.
>>>>>>
>>>>>> It is, I think, a very different world view than the SDN or SUPA o=
r NVO4
>>>>>> or NFV view.
>>>>>>
>>>>>> Regards
>>>>>>     Brian
>>>>>>
>>>>>> On 26/03/2016 16:46, Duzongpeng wrote:
>>>>>>> Hi John:
>>>>>>>
>>>>>>> Thanks for your reply.
>>>>>>>
>>>>>>> 
>>>>>>> How is this different than running a script, or invoking an API? =
There
>>>>>> are plenty of southbound APIs (from a management function to a dev=
ice).
>>>>>> There are very few northbound APIs. That's what the operators that=
 I
>>>> have
>>>>>> talked to want.
>>>>>>> 
>>>>>>>
>>>>>>> I agree that intent belongs to northbound interface in the SDN ne=
twork.
>>>>>> And ANIMA network perhaps can share the same northbound interface.=

>>>>>>> But, my understanding is that ANIMA network is a little different=
 from
>>>>>> SDN network.
>>>>>>> SDN controller will parse the intent, and transform to detailed
>>>>>> configurations of every device. It is a good model, and little
>>>>>> communications between nodes are needed.
>>>>>>> Some network level parameters can be handled in SDN controller.
>>>>>>>
>>>>>>> However, in ANIMA network, perhaps we do not have a centralized p=
arsing
>>>>>> node (controller), which can replace almost all the communications=

>>>> between
>>>>>> nodes.
>>>>>>> So when a network operator is operating a ANIMA network, the netw=
ork
>>>>>> level parameters are still needed to be flooded.
>>>>>>> I mean the ANIMA interface is perhaps something like =E2=80=9Cnor=
thbound + part
>>>>>> of southbound=E2=80=9D.
>>>>>>> Best Regards
>>>>>>> Zongpeng Du
>>>>>>>
>>>>>>>
>>>>>>> From: John Strassner [mailto:strazpdj@gmail.com]
>>>>>>> Sent: Saturday, March 26, 2016 9:41 AM
>>>>>>> To: Duzongpeng; John Strassner
>>>>>>> Cc: Michael Behringer (mbehring); Laurent Ciavaglia; Sheng Jiang;=

>>>>>> anima@ietf.org; J=C3=A9ferson Campos Nobre
>>>>>>> Subject: Re: [Anima] Does Autonomic network need some interventio=
n
>>>>>> beyond intent
>>>>>>> Hi Zongpeng,
>>>>>>>
>>>>>>> I think we have a misunderstanding here.
>>>>>>>
>>>>>>>
>>>>>>> But I do not support that the intent writer should look every det=
ail in
>>>>>> the catalog.
>>>>>>> 
>>>>>>> Most catalogs that I have seen have very few details. They instea=
d tell
>>>>>> the user what services or products are available.
>>>>>>> 
>>>>>>>
>>>>>>> The intent writer may mainly care about what they want, but how t=
o do
>>>> it
>>>>>> can be left to the autonomic network to deal with.
>>>>>>> 
>>>>>>> I agree completely.
>>>>>>> 
>>>>>>>
>>>>>>> Perhaps several solutions exist, and autonomic management system =
choose
>>>>>> one according to the abilities collected.
>>>>>>> 
>>>>>>> This is one of the reasons why I would like to see capabilities (=
and
>>>>>> needs) advertised.
>>>>>>> 
>>>>>>>
>>>>>>> ...
>>>>>>> I mean in current status, some of the Autonomic Functions should =
be
>>>>>> exposed to the network operator, and some of the Autonomic Functio=
ns
>>>> can be
>>>>>> hidden to the network operator
>>>>>>> ...
>>>>>>> 
>>>>>>> How is this different than running a script, or invoking an API? =
There
>>>>>> are plenty of southbound APIs (from a management function to a dev=
ice).
>>>>>> There are very few northbound APIs. That's what the operators that=
 I
>>>> have
>>>>>> talked to want.
>>>>>>> regards,
>>>>>>> John
>>>>>>>
>>>>>>> On Fri, Mar 25, 2016 at 2:31 AM, Duzongpeng >>>>> > wrote:
>>>>>>> Hi John
>>>>>>>
>>>>>>> Some personal understanding here. If any misunderstanding, please=

>>>>>> correct me. Thanks.
>>>>>>> 
>>>>>>> I would suggest a slight modification, to enable users of
>>>>>>> intent (and even intent writers) to NOT have to have
>>>>>>> detailed implementation information.
>>>>>>>
>>>>>>> What if all of the Autonomic Functions, after they joined
>>>>>>> the network, advertised both their capabilities as well as
>>>>>>> their needs? We can put aside the latter (needs) as this
>>>>>>> is more complex, but if the capabilities are advertised,
>>>>>>> then the autonomic management system can assemble
>>>>>>> those needs into a catalog. Then, the intent writer can
>>>>>>> look at the catalog and write a declarative request to
>>>>>>> use the functionality of the catalog.
>>>>>>>
>>>>>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>>>>>> behind...).
>>>>>>> < /jcs2>
>>>>>>>
>>>>>>> I agree that autonomic node should have that self-advertisement
>>>> ability,
>>>>>> and autonomic management system should have this catalog.
>>>>>>> But I do not support that the intent writer should look every det=
ail in
>>>>>> the catalog.
>>>>>>> The intent writer may mainly care about what they want, but how t=
o do
>>>> it
>>>>>> can be left to the autonomic network to deal with.
>>>>>>> Perhaps several solutions exist, and autonomic management system =
choose
>>>>>> one according to the abilities collected.
>>>>>>> In the ideal, the network operator may communicate with the auton=
omic
>>>>>> network using network-level level intent and network-level report.=

>>>>>>> I mean in current status, some of the Autonomic Functions should =
be
>>>>>> exposed to the network operator, and some of the Autonomic Functio=
ns
>>>> can be
>>>>>> hidden to the network operator
>>>>>>> Perhaps in future, when the network nodes become more intelligent=
, the
>>>>>> hidden part will increase.
>>>>>>>
>>>>>>> Best regards
>>>>>>> Zongpeng Du
>>>>>>>
>>>>>>> From: Anima [mailto:anima-bounces@ietf.org>>> anima-bounces@ietf.org>]
>>>>>> On Behalf Of John Strassner
>>>>>>> Sent: Thursday, March 24, 2016 10:41 AM
>>>>>>> To: Michael Behringer (mbehring); John Strassner
>>>>>>> Cc: Duzongpeng; Laurent Ciavaglia; Sheng Jiang; anima@ietf.org>>>>> anima@ietf.org>; J=C3=A9ferson Campos Nobre
>>>>>>> Subject: Re: [Anima] Does Autonomic network need some interventio=
n
>>>>>> beyond intent
>>>>>>> Hi Michael,
>>>>>>>
>>>>>>> thanks for your reply. I do think that we largely agree, and woul=
d
>>>> enjoy
>>>>>> chatting during or (hopefully before) BA. Please see ...
>>>> Also
>>>>>> trying to unclutter, so I am eliding points that we agree on.
>>>>>>>
>>>>>>> While *CIEs don=E2=80=99t like magic, they do understand the conc=
ept of a
>>>>>> default behaviour which they can influence if needed. Any reasonab=
le
>>>>>> network management approach tries to define =E2=80=9Cdefault=E2=80=
=9D behaviour and
>>>>>> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefa=
ult=E2=80=9D part very nicely, and
>>>> *CIEs
>>>>>> get that.
>>>>>>> 
>>>>>>> Agreed!
>>>>>>> 
>>>>>>>
>>>>>>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D d=
iscussion: My view is that we
>>>> shouldn=E2=80=99t
>>>>>> try to define exactly *what* intent allows and doesn=E2=80=99t all=
ow, since
>>>> we=E2=80=99ve
>>>>>> heard from many sides that the boundary is not clear, and it may b=
e
>>>>>> impossible to have a really clear, usable definition. So my point =
was:
>>>>>> Let=E2=80=99s focus on the distribution model: Intent is flooded t=
o all nodes.
>>>>>> Let=E2=80=99s NOT do selective flooding, unicast Intent to specifi=
c nodes, etc,
>>>>>> because that would blur the boundary even worse.
>>>>>>> 
>>>>>>> I like flooding intent to all nodes. In FOCALE (v1-3, and remembe=
r, v1
>>>>>> was 2006, almost 10 years ago), we used a message bus and pub-sub.=
 In
>>>> v2,
>>>>>> we flooded; in v3, we went back to pub-sub because we wanted to do=
 some
>>>>>> optimization and auditing. But I would be happy to try some type o=
f
>>>>>> controlled flooding again.
>>>>>>> 
>>>>>>> My suggestion: Intent is flooded to all nodes in a domain. The AS=
A
>>>>>> owners define the content of what they want to push in their bit o=
f the
>>>>>> Intent. If we insist of full flooding, we encourage the right beha=
viour.
>>>>>> And if someone really puts into his Intent =E2=80=9Cnode A: do thi=
s; node B: do
>>>> the
>>>>>> other; node C: do the third=E2=80=9D it is awkward enough for peop=
le to take a
>>>> step
>>>>>> back and think whether they=E2=80=99re doing the right thing =E2=98=
=BA
>>>>>>> 
>>>>>>> I agree in principle with what you said. Plus, see above.
>>>>>>> < /jcs2>
>>>>>>> 
>>>>>>> Do you mean that the intent writer will **know the difference**
>>>>>>> between the fabric and each autonomic function? I hope not,
>>>>>>> but perhaps I'm misunderstanding you...
>>>>>>> < /jcs>
>>>>>>> Not sure I understand your concern. Yes, in my model there is a s=
ection
>>>>>> for the fabric, and sections for the various functions. Seems a na=
tural
>>>>>> structure?!? 
>>>>>>> The intent **writer** would have to know implementation
>>>>>>> details in order to place everything in a single file. I would
>>>>>>> rather not require this, and instead, have the intent engine
>>>>>>> (e.g., compiler and associated logic) do this.
>>>>>>> < /jcs2>
>>>>>>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to =
structure Intent around
>>>>>> Autonomic Functions. This is not optimal from a scientific point o=
f
>>>> view,
>>>>>> but aligns very well with the way how things are organised. Probab=
ly we
>>>>>> need to discuss that in BA. But refer to my YANG model sent a few =
days
>>>> ago.
>>>>>>> 
>>>>>>> I would suggest a slight modification, to enable users of
>>>>>>> intent (and even intent writers) to NOT have to have
>>>>>>> detailed implementation information.
>>>>>>>
>>>>>>> What if all of the Autonomic Functions, after they joined
>>>>>>> the network, advertised both their capabilities as well as
>>>>>>> their needs? We can put aside the latter (needs) as this
>>>>>>> is more complex, but if the capabilities are advertised,
>>>>>>> then the autonomic management system can assemble
>>>>>>> those needs into a catalog. Then, the intent writer can
>>>>>>> look at the catalog and write a declarative request to
>>>>>>> use the functionality of the catalog.
>>>>>>>
>>>>>>> I will look at your YANG model in a bit (sorry, I'm woefully
>>>>>>> behind...).
>>>>>>> < /jcs2>
>>>>>>>
>>>>>>>
>>>>>>> regards,
>>>>>>> John
>>>>>>>
>>>>>>> On Wed, Mar 23, 2016 at 3:57 AM, Michael Behringer (mbehring) <
>>>>>> mbehring@cisco.com> wrote:
>>>>>>> John,
>>>>>>>
>>>>>>> I think we=E2=80=99re on the same page. Couple of points, top pos=
ting here,
>>>>>> otherwise it gets too messy.
>>>>>>> While *CIEs don=E2=80=99t like magic, they do understand the conc=
ept of a
>>>>>> default behaviour which they can influence if needed. Any reasonab=
le
>>>>>> network management approach tries to define =E2=80=9Cdefault=E2=80=
=9D behaviour and
>>>>>> =E2=80=9Cexceptions=E2=80=9D. The AN story fills the =E2=80=9Cdefa=
ult=E2=80=9D part very nicely, and
>>>> *CIEs
>>>>>> get that.
>>>>>>> The =E2=80=9Cparameter=E2=80=9D versus =E2=80=9Cpolicy=E2=80=9D d=
iscussion: My view is that we
>>>> shouldn=E2=80=99t
>>>>>> try to define exactly *what* intent allows and doesn=E2=80=99t all=
ow, since
>>>> we=E2=80=99ve
>>>>>> heard from many sides that the boundary is not clear, and it may b=
e
>>>>>> impossible to have a really clear, usable definition. So my point =
was:
>>>>>> Let=E2=80=99s focus on the distribution model: Intent is flooded t=
o all nodes.
>>>>>> Let=E2=80=99s NOT do selective flooding, unicast Intent to specifi=
c nodes, etc,
>>>>>> because that would blur the boundary even worse.
>>>>>>> My suggestion: Intent is flooded to all nodes in a domain. The AS=
A
>>>>>> owners define the content of what they want to push in their bit o=
f the
>>>>>> Intent. If we insist of full flooding, we encourage the right beha=
viour.
>>>>>> And if someone really puts into his Intent =E2=80=9Cnode A: do thi=
s; node B: do
>>>> the
>>>>>> other; node C: do the third=E2=80=9D it is awkward enough for peop=
le to take a
>>>> step
>>>>>> back and think whether they=E2=80=99re doing the right thing =E2=98=
=BA
>>>>>>> 
>>>>>>> Do you mean that the intent writer will **know the difference**
>>>>>>> between the fabric and each autonomic function? I hope not,
>>>>>>> but perhaps I'm misunderstanding you...
>>>>>>> 
>>>>>>>
>>>>>>> Not sure I understand your concern. Yes, in my model there is a s=
ection
>>>>>> for the fabric, and sections for the various functions. Seems a na=
tural
>>>>>> structure?!?
>>>>>>> =E2=80=9CFunction owner=E2=80=9D: Yes, I=E2=80=99m suggesting to =
structure Intent around
>>>>>> Autonomic Functions. This is not optimal from a scientific point o=
f
>>>> view,
>>>>>> but aligns very well with the way how things are organised. Probab=
ly we
>>>>>> need to discuss that in BA. But refer to my YANG model sent a few =
days
>>>> ago.
>>>>>>> We should schedule some discussion time outside the ANIMA meeting=

>>>>>> window...
>>>>>>> Michael
>>>>>>>
>>>>>>>
>>>>>>> From: John Strassner [mailto:strazpdj@gmail.com>>>>> strazpdj@gmail.com>]
>>>>>>> Sent: 22 March 2016 21:32
>>>>>>> To: Michael Behringer (mbehring) >>>>> mbehring@cisco.com>>; John Strassner >>>>> strazpdj@gmail.com>>
>>>>>>> Cc: Duzongpeng >;
>>>>>> anima@ietf.org; Laurent Ciavaglia <
>>>>>> laurent.ciavaglia@nokia.com>;
>>>>>> J=C3=A9ferson Campos Nobre >>>>> ;
>>>>>> Sheng Jiang >
>>>>>>> Subject: Re: [Anima] Does Autonomic network need some interventio=
n
>>>>>> beyond intent
>>>>>>> In general, I agree with Michael. I'd like to emphasize and expan=
d on a
>>>>>> couple of key points:
>>>>>>>
>>>>>>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=
=80=9D or =E2=80=9Csteering=E2=80=9D a
>>>> network,
>>>>>> at least not initially. There is still traditional config, plus al=
l the
>>>>>> existing and emerging SDN models. My view has always been that ANI=
MA
>>>> sets a
>>>>>> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means=
 are more specific and
>>>>>> can override intent. This is described in RFC7575. Main point here=
:
>>>> ANIMA
>>>>>> Intent isn=E2=80=99t the only tool in a network.
>>>>>>> 
>>>>>>> Absolutely agree. Note that even in the Congress case I mentioned=

>>>>>>> earlier, it is NOT pushing config, and it is NOT viewed as the
>>>>>>> "only" policy
>>>>>>> 
>>>>>>>
>>>>>>>
>>>>>>> Intent could perfectly well cover a high level policy such as =E2=
=80=9Call
>>>> nodes
>>>>>> of type x do this; type y does that=E2=80=9D. (Which I think is wh=
at you mention
>>>>>> below.) When we started with Intent, the vision was to keep it ver=
y high
>>>>>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=
=98=BA
>>>>>>> 
>>>>>>> Agreed. Although to me, this means that we need to consider the t=
ypes
>>>> of
>>>>>> actors that are using intent. For example, most of my CCIE and JCI=
E
>>>> friends
>>>>>> have no interest in using an intent language for pushing configura=
tion.
>>>> On
>>>>>> the other hand, when I talk to operators and application developer=
s,
>>>> they
>>>>>> would love to be able to use a high-level intent and have the inte=
nt
>>>> engine
>>>>>> take care of the details for them.
>>>>>>> 
>>>>>>>
>>>>>>> It has become clear that in today=E2=80=99s networking, people wi=
ll want to
>>>> push
>>>>>> more =E2=80=9Cparameters=E2=80=9D than =E2=80=9Cpolicy=E2=80=9D, a=
nd I think we=E2=80=99ll just have to
>>>> acknowledge
>>>>>> that.
>>>>>>> 
>>>>>>> Sorry Michael, I don't follow here. What do you mean
>>>>>>> by "parameters"?
>>>>>>> 
>>>>>>>
>>>>>>>
>>>>>>> Personally I think it=E2=80=99ll pan out like this: Intent will h=
ave sections
>>>>>> per autonomic function. An Intent parser will validate overall
>>>> correctness
>>>>>> (signature, etc), then segment the Intent into its bits: One part =
will
>>>> be
>>>>>> the autonomic fabric itself, then there will be bits for each auto=
nomic
>>>>>> function.
>>>>>>> 
>>>>>>> Do you mean that the intent writer will **know the difference**
>>>>>>> between the fabric and each autonomic function? I hope not,
>>>>>>> but perhaps I'm misunderstanding you...
>>>>>>> 
>>>>>>>
>>>>>>>
>>>>>>> So the Intent parser on a node will grab intent, see there is som=
e
>>>>>> Intent for autonomic function A, will see whether this is running
>>>> locally,
>>>>>> and if so, it=E2=80=99ll just pass that piece of Intent to that fu=
nction.
>>>>>>> 
>>>>>>> Agreed.
>>>>>>> 
>>>>>>>
>>>>>>>
>>>>>>> In this way, a function owner defines both the intent itself, and=
 how
>>>> it
>>>>>> is consumed by his/her function.
>>>>>>> 
>>>>>>> I'm not sure what you mean by "function owner". (Sorry to be
>>>>>>> picky, I'm likely under-caffeinated.)
>>>>>>>
>>>>>>> The above seemed like an autonomous decision, made by the parser.=

>>>>>>> I don't think that the intent writer should be cognizant of how m=
any or
>>>>>>> what types of autonomic functions are present.
>>>>>>> 
>>>>>>>
>>>>>>>
>>>>>>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and pe=
ople can do what they
>>>>>> want, in the worst case even push config, which is clearly NOT wha=
t
>>>> Intent
>>>>>> is meant to do.
>>>>>>> 
>>>>>>> And hence, the "HAL 2000" scenario. :-)
>>>>>>> 
>>>>>>>
>>>>>>>
>>>>>>> I=E2=80=99d hope that people don=E2=80=99t start writing Intent l=
ike =E2=80=9Cnode 17: here is
>>>>>> your CLI; node 23: here is your CLI=E2=80=9D; node 37: here is you=
r CLI=E2=80=9D. But
>>>> what
>>>>>> I describe above COULD be abused in this way. RFC7575 says this is=
 NOT
>>>>>> Intent.
>>>>>>> 
>>>>>>> Absolutely agree!
>>>>>>> 
>>>>>>>
>>>>>>>   regards,
>>>>>>> John
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Wed, Mar 16, 2016 at 1:16 AM, Michael Behringer (mbehring) <
>>>>>> mbehring@cisco.com> wrote:
>>>>>>> Hi Zongpeng,
>>>>>>>
>>>>>>> Yes, this is a recurring question and topic. Let me give my view =
on two
>>>>>> levels:
>>>>>>> First, ANIMA isn=E2=80=99t the only thing =E2=80=9Cinfluencing=E2=
=80=9D or =E2=80=9Csteering=E2=80=9D a
>>>> network,
>>>>>> at least not initially. There is still traditional config, plus al=
l the
>>>>>> existing and emerging SDN models. My view has always been that ANI=
MA
>>>> sets a
>>>>>> =E2=80=9Cdefault=E2=80=9D policy in a network, and all other means=
 are more specific and
>>>>>> can override intent. This is described in RFC7575. Main point here=
:
>>>> ANIMA
>>>>>> Intent isn=E2=80=99t the only tool in a network.
>>>>>>> Intent could perfectly well cover a high level policy such as =E2=
=80=9Call
>>>> nodes
>>>>>> of type x do this; type y does that=E2=80=9D. (Which I think is wh=
at you mention
>>>>>> below.) When we started with Intent, the vision was to keep it ver=
y high
>>>>>> level, more along the line =E2=80=9Cdo the right thing=E2=80=9D =E2=
=98=BA  It has become clear
>>>> that
>>>>>> in today=E2=80=99s networking, people will want to push more =E2=80=
=9Cparameters=E2=80=9D than
>>>>>> =E2=80=9Cpolicy=E2=80=9D, and I think we=E2=80=99ll just have to a=
cknowledge that.
>>>>>>> Personally I think it=E2=80=99ll pan out like this: Intent will h=
ave sections
>>>>>> per autonomic function. An Intent parser will validate overall
>>>> correctness
>>>>>> (signature, etc), then segment the Intent into its bits: One part =
will
>>>> be
>>>>>> the autonomic fabric itself, then there will be bits for each auto=
nomic
>>>>>> function.
>>>>>>> So the Intent parser on a node will grab intent, see there is som=
e
>>>>>> Intent for autonomic function A, will see whether this is running
>>>> locally,
>>>>>> and if so, it=E2=80=99ll just pass that piece of Intent to that fu=
nction.
>>>>>>> In this way, a function owner defines both the intent itself, and=
 how
>>>> it
>>>>>> is consumed by his/her function.
>>>>>>> The downside is that there is no =E2=80=9Ccontrol=E2=80=9D and pe=
ople can do what they
>>>>>> want, in the worst case even push config, which is clearly NOT wha=
t
>>>> Intent
>>>>>> is meant to do. I=E2=80=99d hope that people don=E2=80=99t start w=
riting Intent like
>>>> =E2=80=9Cnode
>>>>>> 17: here is your CLI; node 23: here is your CLI=E2=80=9D; node 37:=
 here is your
>>>>>> CLI=E2=80=9D. But what I describe above COULD be abused in this wa=
y. RFC7575
>>>> says
>>>>>> this is NOT Intent.
>>>>>>> In other words: Let=E2=80=99s not be too prescriptive, and hope t=
hat folks will
>>>>>> do more or less reasonable things.
>>>>>>> I=E2=80=99ve been meaning to comment on the Intent drafts and pos=
sibly
>>>>>> contribute, I=E2=80=99m just struggling with priorities at the mom=
ent (as you
>>>> can
>>>>>> see by my relative silence over the past weeks).
>>>>>>> Michael
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> From: Anima [mailto:anima-bounces@ietf.org>>> anima-bounces@ietf.org>]
>>>>>> On Behalf Of Duzongpeng
>>>>>>> Sent: 16 March 2016 02:38
>>>>>>> To: anima@ietf.org
>>>>>>> Cc: Laurent Ciavaglia >>>>> laurent.ciavaglia@nokia.com>>; J=C3=A9ferson Campos Nobre <
>>>> jcnobre@inf.ufrgs.br
>>>>>> >; Sheng Jiang >>> >>>>> jiangsheng@huawei.com>>
>>>>>>> Subject: [Anima] Does Autonomic network need some intervention be=
yond
>>>>>> intent
>>>>>>> Hello, everyone
>>>>>>>
>>>>>>>           I understand that Autonomic network should be able to w=
ork
>>>> with
>>>>>> as less configuration as possible. In the ideal case, an Autonomic=

>>>> network
>>>>>> should be able to work well while only intent is needed from human=

>>>>>> operators.
>>>>>>>           However, I wonder whether Autonomic network may need so=
me
>>>>>> intervention from human operators through some low-level policies =
beyond
>>>>>> intents.
>>>>>>>           For example, in the
>>>>>> https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,=
 it
>>>> is
>>>>>> suggested that the prefix lengths for the CSG, ASG, RSG (different=

>>>> roles in
>>>>>> IP RAN) can be assigned as an "intent". These configuration parame=
ters
>>>> need
>>>>>> to be distributed in the autonomic domain to influence the detail
>>>>>> configurations on each autonomic node.
>>>>>>>           As intent is described as an abstract, declarative, hig=
h-level
>>>>>> policy used to operate an autonomic domain, such as an enterprise
>>>> network.
>>>>>> Perhaps we can rename the =E2=80=9Cintent=E2=80=9D in the =E2=80=9C=
prefix-management ID=E2=80=9D to some
>>>>>> other words.
>>>>>>> Is =E2=80=9CASA parameters" ok here?
>>>>>>>
>>>>>>> I think some characteristics for this kind of low-level policies =
are
>>>>>>>
>>>>>>> 1.       Network level parameters, need to be distributed by GRAS=
P and
>>>>>> ACP
>>>>>>> 2.       Relatively static compared to intent, perhaps only neede=
d to
>>>> be
>>>>>> configured once
>>>>>>> 3.       Unavoidable in current network environment, mostly for
>>>>>> establishing the network infrastructure
>>>>>>> 4.       Rarely need coordinations with others parameters, usuall=
y is
>>>>>> closed related to one specific autonomic function
>>>>>>> Welcome for comment. Your suggestions will be appreciated.
>>>>>>>
>>>>>>>
>>>>>>> Best Regards
>>>>>>> Zongpeng Du
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Anima mailing list
>>>>>>> Anima@ietf.org
>>>>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --=20
>>>>>>> regards,
>>>>>>> John
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --=20
>>>>>>> regards,
>>>>>>> John
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --=20
>>>>>>> regards,
>>>>>>> John
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Anima mailing list
>>>>>>> Anima@ietf.org
>>>>>>> https://www.ietf.org/mailman/listinfo/anima
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>=20


From nobody Tue Mar 29 15:40:46 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7ECEE12D9B7 for ; Tue, 29 Mar 2016 15:40:45 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KGLOhT1xZAIL for ; Tue, 29 Mar 2016 15:40:43 -0700 (PDT)
Received: from mail-pa0-x230.google.com (mail-pa0-x230.google.com [IPv6:2607:f8b0:400e:c03::230]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0768D12DAAC for ; Tue, 29 Mar 2016 15:40:42 -0700 (PDT)
Received: by mail-pa0-x230.google.com with SMTP id fe3so24648320pab.1 for ; Tue, 29 Mar 2016 15:40:42 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=DKWL67rkCc/aAgwpxpsQ/2Z8G5B/oeiMxKAIoXAACe0=; b=QbUHphV7xk1bmb9wh+bLdcAwnlM5gCKMVZ6TPqwNMiYIeMTl64vbLrkqQIfC0vcIjk UYxnQ1Os6t/UBCmKiZ51j+mQ5PsNAzugVa6HU4F+wTi5QFW9ajiwTHQbSwspo2fwV27W 8pHyyngE45xXNR+anPVsUR5drmiQ3VBT9+OaB3uVHcVNVwCWVIoYwQjbmSpNW6usEqQh UgnGi3qHpUlgVk1jLqkdcb+KeG1iTd9gDbJrcjdCk3/hrBvMhC9DWCPFP95OajKxdmhz xiKr6ko9/W8kZI2PEehwdsDgFnsoGniTnxXy6sLL3dpFAWSgJuJgWuall6aiErKU/6me zWYA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=DKWL67rkCc/aAgwpxpsQ/2Z8G5B/oeiMxKAIoXAACe0=; b=jEFXqObyJlS3CNMmdoZpUf/BIblJeRL6HayuQvlfee/q6iOLwqEROxgjwLw5tLOJLe VD/ngBX0VRDxeJaLboWVLVjRy+aromeS0gUZC1VK2fE9/vhihgNbScb2Tp6m1A2fmMEi BiAHhdBmHgus0GG+9eYkFRbU2XPwWv/6oXDfI0Thc5EVlf2Tm55agskovuK1RgjA1BWa Aq5IkbiXljdcaAP95kjrdvefzUo216KUx9oJ4eD/6LClU8Hvhrr8TamUHi/qNYpz0kG/ oHPYofhaQS0rNT+Zd+lqr9RH/ioB+E9ksv6jTEviceTJZO/mOT6mvuob7snu0Qqm7lK7 WyTg==
X-Gm-Message-State: AD7BkJKHFFJ9GLfFpx6VfL5jqdT1alY+eGHTG+rAAh25BJSYGr5wHdchfqETFzrgdELGgA==
X-Received: by 10.66.121.97 with SMTP id lj1mr7361274pab.51.1459291241660; Tue, 29 Mar 2016 15:40:41 -0700 (PDT)
Received: from ?IPv6:2001:df0:0:2006:c0da:ac17:5f6d:8e76? ([2001:df0:0:2006:c0da:ac17:5f6d:8e76]) by smtp.gmail.com with ESMTPSA id m89sm724608pfi.12.2016.03.29.15.40.37 (version=TLSv1/SSLv3 cipher=OTHER); Tue, 29 Mar 2016 15:40:40 -0700 (PDT)
To: "Michael Behringer (mbehring)" , John Strassner , Duzongpeng 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com> 
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56FB0468.9010404@gmail.com>
Date: Wed, 30 Mar 2016 11:40:40 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: 
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 22:40:45 -0000

On 29/03/2016 21:58, Michael Behringer (mbehring) wrote:
>> I just thought that if the NOC wants to send out some sort of urgent
>> instruction to all nodes, it is by its nature a partial update of Intent. Of course
>> a partial update can be achieved by sending everything. That's a trade-off of
>> resource vs complexity, I suppose.
> 
> Anything *can* be partial. My claim is that if Intent really is a high level policy, it should be so stable over time that the optimisation to support partial updates just isn't required. 
> 
> In my proposal on "content distribution" (https://mailarchive.ietf.org/arch/msg/anima/KNIglx9abDMEejwEZu9hE4nDA08) I suggest a content type; I suggest to start with only one content type supported ("Intent version 0") and get experience with that. If it turns out that we need to slice it, we could define new content types, potentially much more granular, so that we could later support effectively a partial update. If we do this job well, the distribution protocol doesn't need to change at all. 
> 
> So my proposal is really: Let's start as simple as possible, but leave space to grow if we need to. 

Without disrespect to the subsequent arguments, I think this is a key
point. At the moment we're designing the ANI, not the Intent system,
so what I'm looking for in this discussion is the features that the ANI
has to support - which seem to be
(a) ability to distribute the complete Intent to all nodes
(b) ability to distribute partial Intent to all (interested) nodes later on

BTW I put "interested" in parenthesis because it seems dangerous to
assume that a node's interests don't change. As ASAs start or stop
in a given node, its interest in Intent will change. So unless we
construct a quite elaborate pub/sub system, flooding is needed,
even for partial intent. That's what GRASP supports, of course.

    Brian


From nobody Tue Mar 29 16:20:46 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EAA0F12DB19 for ; Tue, 29 Mar 2016 16:20:44 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qAGJlWnmUekJ for ; Tue, 29 Mar 2016 16:20:41 -0700 (PDT)
Received: from mail-pa0-x230.google.com (mail-pa0-x230.google.com [IPv6:2607:f8b0:400e:c03::230]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D36CB12DB25 for ; Tue, 29 Mar 2016 16:20:41 -0700 (PDT)
Received: by mail-pa0-x230.google.com with SMTP id zm5so25352438pac.0 for ; Tue, 29 Mar 2016 16:20:41 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=ofM+J6AoV13ZFVHgXdwJq58/5gEAjbvyVyLxaThuZQs=; b=bzIgRlclkI0YtY2RZimpF+iq/1BvMw34NuzEMg9TD6qU2R+Y33/dGSWf/Ya5txS5w1 //DLBxAy1r7IJls0jE6aHPKk/FnZoM6BBtMWvQJd2WnzO8pb2YFK8tWBbGnGDOxYtUzM 2ts8VZ9q7M9108l3uMpuWApbrBZk/sI4XToM8iMTHeMN1+8RX6P7Ys2x0kTcaFhWbrJ3 jbN23/zAxcrvbBsfoLCh+D1uSgtUJ1/jRgPuZ2HeDBBPTP7Z2JyPSBn2DdNq+DjGWc9p AQ+31BDVIb+nsc25FCiPMAWTz/vtmKhciU4Yuj94tDsOUFYJx2vmfHmq0Y3q0p5bPrki DiYw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=ofM+J6AoV13ZFVHgXdwJq58/5gEAjbvyVyLxaThuZQs=; b=Fjp2sRvoNaSG7XzkyFC6l4ml0lOq5XgVRtCHlUIbkQPTjdvYITd7l4ej+G4S4cWI+P NRRy9gJr/j0zCFZ0GFFPCItRG44e+J7encSoVNqUi0JSMg+wGYOkaMeG3IA1Z/UTxuxJ 2hRysN7dQHLAA639hURcTd1Tfa24LrQYhhMTa3Sl7u4LlVuQC2WTdIJmzW7oa6cLqdie AtE5GMK7IMV9m0giat0XNL/jjLNU2d2d7bxpa5c1fpZQpjotzNHyVJ++ow1GDitefpBp /9ZEBsTViKsq9j6t9Ufwka6Uwh3ANOooFNFw74S3g7Q7J0FqcwbZXRs7N9qCfYzp0awK Y7fg==
X-Gm-Message-State: AD7BkJI63la+nFY/4y1px92ZE74dhAg/TYJw6p5ospbiTXUMhOV0lfa8B0vcvcdQ5tjD/A==
X-Received: by 10.66.248.198 with SMTP id yo6mr7655323pac.54.1459293641415; Tue, 29 Mar 2016 16:20:41 -0700 (PDT)
Received: from ?IPv6:2001:df0:0:2006:c0da:ac17:5f6d:8e76? ([2001:df0:0:2006:c0da:ac17:5f6d:8e76]) by smtp.gmail.com with ESMTPSA id h2sm770804pfd.91.2016.03.29.16.20.36 (version=TLSv1/SSLv3 cipher=OTHER); Tue, 29 Mar 2016 16:20:39 -0700 (PDT)
To: "Joel M. Halpern" , "Michael Behringer (mbehring)" , John Strassner , Duzongpeng 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FAA188.2050901@joelhalpern.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56FB0DC6.1030701@gmail.com>
Date: Wed, 30 Mar 2016 12:20:38 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56FAA188.2050901@joelhalpern.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 23:20:45 -0000

On 30/03/2016 04:38, Joel M. Halpern wrote:
> The notion of distributing all intents as a single blob seems like a really bad model.  Policy always decomposes into pieces. 
> Some of those pieces are related.  I think that assuming that the total policy (intent) is small is asking for trouble.
> 
> This is also why I worry a bit about flooding to everything.

There's a technical issue there, which is that GRASP flooding uses
multicast UDP (what else could it reasonably use, anyway?). So if
the chunk to be flooded gets too big, we would have to chop it up.

    Brian

> 
> Yours,
> Joel
> 
> On 3/29/16 4:58 AM, Michael Behringer (mbehring) wrote:
>>> I just thought that if the NOC wants to send out some sort of urgent
>>> instruction to all nodes, it is by its nature a partial update of Intent. Of course
>>> a partial update can be achieved by sending everything. That's a trade-off of
>>> resource vs complexity, I suppose.
>>
>> Anything *can* be partial. My claim is that if Intent really is a high level policy, it should be so stable over time that the
>> optimisation to support partial updates just isn't required.
>>
>> In my proposal on "content distribution" (https://mailarchive.ietf.org/arch/msg/anima/KNIglx9abDMEejwEZu9hE4nDA08) I suggest a
>> content type; I suggest to start with only one content type supported ("Intent version 0") and get experience with that. If it
>> turns out that we need to slice it, we could define new content types, potentially much more granular, so that we could later
>> support effectively a partial update. If we do this job well, the distribution protocol doesn't need to change at all.
>>
>> So my proposal is really: Let's start as simple as possible, but leave space to grow if we need to.
>>
>> Michael
>>
>> _______________________________________________
>> Anima mailing list
>> Anima@ietf.org
>> https://www.ietf.org/mailman/listinfo/anima
>>
> .
> 


From nobody Tue Mar 29 16:55:16 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 411CB12DB67 for ; Tue, 29 Mar 2016 16:55:15 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.91
X-Spam-Level: 
X-Spam-Status: No, score=-6.91 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_KAM_HTML_FONT_INVALID=0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id klr801WinRCO for ; Tue, 29 Mar 2016 16:55:10 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9DE8112DB0C for ; Tue, 29 Mar 2016 16:55:09 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id 3F614D8979977; Tue, 29 Mar 2016 23:55:03 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2TNt6WX012121 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Tue, 29 Mar 2016 23:55:07 GMT
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2TNsnA1007111 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 30 Mar 2016 01:55:04 +0200
Received: from [135.224.216.185] (135.239.27.41) by FR712WXCHHUB03.zeu.alcatel-lucent.com (135.239.2.74) with Microsoft SMTP Server (TLS) id 14.3.195.1; Wed, 30 Mar 2016 01:54:59 +0200
To: "EXT Michael Behringer (mbehring)" , Laurent Ciavaglia , Brian E Carpenter , John Strassner , Duzongpeng 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56FAA4B7.9010305@alcatel-lucent.com> <58704b9ce2cb43c0b07d9aac8b284187@XCH-RCD-006.cisco.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56FB15D2.10200@alcatel-lucent.com>
Date: Wed, 30 Mar 2016 01:54:58 +0200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <58704b9ce2cb43c0b07d9aac8b284187@XCH-RCD-006.cisco.com>
Content-Type: multipart/alternative; boundary="------------020903040400010904040802"
X-Originating-IP: [135.239.27.41]
Archived-At: 
Cc: =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Tue, 29 Mar 2016 23:55:15 -0000

--------------020903040400010904040802
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit

Thanks Michael (B.), I was wrong in my understanding.

Laurent.

On 29/03/2016 18:34, EXT Michael Behringer (mbehring) wrote:
>
> When I say “freeze†I mean: All the nodes that the operator expected 
> to join, have joined; the next ones will come in 6 months. Until then, 
> I want to disable the enrolment function.
>
> Obviously, it **could** be understood as: for today, there are no more 
> installations of new devices expected; tomorrow at 9 I expect an 
> installation, so I unfreeze the network, once that device has enrolled 
> I freeze again until 10, when the next one it expected. That would not 
> be the right thing.
>
> I guess that’s exactly the same for all examples... It depends how you 
> use them.
>
> Michael
>
> *From:*Laurent Ciavaglia [mailto:laurent.ciavaglia@nokia.com]
> *Sent:* 29 March 2016 17:52
> *To:* Michael Behringer (mbehring) ; Brian E 
> Carpenter ; John Strassner 
> ; Duzongpeng 
> *Cc:* Laurent Ciavaglia ; Jéferson Campos 
> Nobre ; anima@ietf.org; Sheng Jiang 
> 
> *Subject:* Re: Another use case for Intent: "Freeze network enrolment"
>
> Michael,
>
> It seems to me that "Freeze network enrollment" is a valid declarative 
> form of policy (intent), yet it seems to be a counter-example to the 
> argument of long-lasting intent, i.e. the operator will want to 
> roll-back to "normal" behavior as soon as the issue that raised the 
> "freeze" intent creation is cleared...
> (I am not saying that the case is not a good one to investigate).
>
> Am I over-/mis-interpreting?
>
> Best regards, Laurent.
>
> On 24/03/2016 14:49, EXT Michael Behringer (mbehring) wrote:
>
>     Maybe we should combine two threads here. I think there is a
>     perfect use case for Intent I  mentioned in another mail, which
>     would be even applicable to what we do today:
>
>     "Freeze network enrolment".
>
>     Since any node can be a proxy node for bootstrap, it makes perfect
>     sense to flood this to all nodes.
>
>     Local override possible via traditional config / netconf / etc.
>
>     No parameters needed. Can't get any simpler.
>
>     Michael
>
>         -----Original Message-----
>
>         From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]
>
>         Sent: 24 March 2016 05:53
>
>         To: John Strassner 
>         ; Duzongpeng
>
>          
>
>         Cc: Laurent Ciavaglia 
>         ; Jéferson Campos Nobre
>
>          ;
>         anima@ietf.org ; Michael Behringer
>         (mbehring)
>
>          ; Sheng Jiang
>          
>
>         Subject: Re: [Anima] Does Autonomic network need some intervention
>
>         beyond intent
>
>         On 24/03/2016 16:49, John Strassner wrote:
>
>             I am confused as to why we would consider
>             prefix-management as an
>
>             autonomic intent.
>
>         It's the other way round. Prefix management was chosen as an
>         autonomic
>
>         use case, so a follow-up question is: what intent is
>         appropriate for that use
>
>         case?
>
>         And the suggested answer is: the ISP's intent is that certain
>         types of edge
>
>         network should get certain sizes of prefix. How the prefixes
>         are actually
>
>         divided up and assigned is left to the autonomic nodes.
>
>             To me, this breaks a fundamental rule of what is intent by
>             requiring
>
>             the intent writer to understand too many details about the
>             target of
>
>             the intent. The analogy (at least in the ONF and TMF) was
>             specifying
>
>             to a taxi company that you want a taxi, but you need to
>             take a certain
>
>             route, because you (the customer) know traffic patterns
>             better than
>
>             the taxi driver.
>
>             Surely we could pick a more appropriate example?
>
>         By all means describe other use cases and what sort of intent
>         they need.
>
>             Brian
>
>             regards,
>
>             John
>
>             On Wed, Mar 16, 2016 at 5:41 AM, Duzongpeng
>
>           wrote:
>
>                 Hi, Michael,
>
>                          Thanks for the reply. I can understand that
>                 intent is not in
>
>                 the current milestones, and you are working on several
>                 milestones drafts
>
>         now.
>
>                          You are welcome to contribute to the intent
>                 draft at any
>
>                 time, and thanks in advance.
>
>                          According to your statements, in my
>                 understanding, we can
>
>                 keep regarding the “intent†example in
>                 “prefix-management ID†as an
>
>                 Autonomic intent.
>
>                          It’s semantics are just like “all nodes of
>                 type x do this;
>
>                 type y does thatâ€, so it is a high level policy.
>
>                          It really helps a lot. Thanks.
>
>                          I generally agree with your statement, while
>                 some other
>
>                 personal understandings exist (shown below). But
>                 perhaps they can be
>
>         discussed later.
>
>                 Best
>
>                 Regards
>
>                 Zongpeng Du
>
>                 *From:* Michael Behringer (mbehring)
>                 [mailto:mbehring@cisco.com]
>
>                 *Sent:* Wednesday, March 16, 2016 4:17 PM
>
>                 *To:* Duzongpeng; anima@ietf.org 
>
>                 *Cc:* Laurent Ciavaglia; Jéferson Campos Nobre; Sheng
>                 Jiang
>
>                 *Subject:* RE: [Anima] Does Autonomic network need
>                 some intervention
>
>                 beyond intent
>
>                 Hi Zongpeng,
>
>                 Yes, this is a recurring question and topic. Let me
>                 give my view on
>
>                 two
>
>                 levels:
>
>                 First, ANIMA isn’t the only thing “influencing†or
>                 “steering†a
>
>                 network, at least not initially. There is still
>                 traditional config,
>
>                 plus all the existing and emerging SDN models. My view
>                 has always
>
>                 been that ANIMA sets a “default†policy in a network,
>                 and all other
>
>                 means are more specific and can override intent. This
>                 is described in
>
>                 RFC7575. Main point here: ANIMA Intent isn’t the only
>                 tool in a network.
>
>                 [duzongpeng] I agree that intent can coexist with
>                 traditional config.
>
>                 I generally agree with the policy priorities, but I
>                 have some other
>
>                 concerns.
>
>                 Some default policies/parameters set through “intentâ€
>                 perhaps are
>
>                 fundamental for the uplayer services.
>
>                 It should be prevented that human operator uses some
>                 CLIs to change
>
>                 them by mistake, and the uplayer services are interrupted.
>
>                 At least, some warning should be reported to the human
>                 operator.
>
>                 [duzongpeng]
>
>                 Intent could perfectly well cover a high level policy
>                 such as “all
>
>                 nodes of type x do this; type y does thatâ€. (Which I
>                 think is what
>
>                 you mention
>
>                 below.) When we started with Intent, the vision was to
>                 keep it very
>
>                 high level, more along the line “do the right thingâ€
>                 J  It has become
>
>                 clear that in today’s networking, people will want to
>                 push more
>
>                 “parameters†than “policyâ€, and I think we’ll just
>                 have to acknowledge
>
>         that.
>
>                 Personally I think it’ll pan out like this: Intent
>                 will have sections
>
>                 per autonomic function. An Intent parser will validate
>                 overall
>
>                 correctness (signature, etc), then segment the Intent
>                 into its bits:
>
>                 One part will be the autonomic fabric itself, then
>                 there will be bits
>
>                 for each autonomic function.
>
>                 So the Intent parser on a node will grab intent, see
>                 there is some
>
>                 Intent for autonomic function A, will see whether this
>                 is running
>
>                 locally, and if so, it’ll just pass that piece of
>                 Intent to that function.
>
>                 In this way, a function owner defines both the intent
>                 itself, and how
>
>                 it is consumed by his/her function.
>
>                 [duzongpeng]Agree.[duzongpeng]
>
>                 The downside is that there is no “control†an//d
>                 people can do what
>
>                 they want, in the worst case even push config, which
>                 is clearly NOT
>
>                 what Intent is meant to do. I’d hope that people don’t
>                 start writing
>
>                 Intent like “node
>
>                 17: here is your CLI; node 23: here is your CLIâ€; node
>                 37: here is
>
>                 your CLIâ€. But what I describe above COULD be abused
>                 in this way.
>
>                 RFC7575 says this is NOT Intent.
>
>                 [duzongpeng]Agree. It is hard to give an exact boundary.
>
>                 However, I think we have two choices here.
>
>                 The first one is to define a policy first in every
>                 ASA, but with some
>
>                 open parameters that can be configured. What we need
>                 to do is just to
>
>                 distribute the network-level parameters for that
>                 function. This way
>
>                 is suitable to some relatively static functions.
>
>                 The second is more advanced, in the ideal case, the
>                 human operator
>
>                 can communicate with network using natural language or
>                 something
>
>                 similar. This way is more flexible, but harder to realize.
>
>                 [duzongpeng]
>
>                 In other words: Let’s not be too prescriptive, and
>                 hope that folks
>
>                 will do more or less reasonable things.
>
>                 [duzongpeng]Agree.[duzongpeng]
>
>                 I’ve been meaning to comment on the Intent drafts and
>                 possibly
>
>                 contribute, I’m just struggling with priorities at the
>                 moment (as you
>
>                 can see by my relative silence over the past weeks).
>
>                 Michael
>
>                 *From:* Anima [mailto:anima-bounces@ietf.org
>
>                 
>                 ] *On Behalf Of *Duzongpeng
>
>                 *Sent:* 16 March 2016 02:38
>
>                 *To:* anima@ietf.org 
>
>                 *Cc:* Laurent Ciavaglia 
>                 ; Jéferson
>
>                 Campos Nobre 
>                 ; Sheng Jiang
>
>                  
>
>                 *Subject:* [Anima] Does Autonomic network need some
>                 intervention
>
>                 beyond intent
>
>                 Hello, everyone
>
>                          I understand that Autonomic network should be
>                 able to work
>
>                 with as less configuration as possible. In the ideal
>                 case, an
>
>                 Autonomic network should be able to work well while
>                 only intent is
>
>                 needed from human operators.
>
>                          However, I wonder whether Autonomic network
>                 may need some
>
>                 intervention from human operators through some
>                 low-level policies
>
>                 beyond intents.
>
>                          For example, in the
>
>                 https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00,
>                 it
>
>                 is suggested that the prefix lengths for the CSG, ASG,
>                 RSG (different
>
>                 roles in IP RAN) can be assigned as an "intent". These
>                 configuration
>
>                 parameters need to be distributed in the autonomic
>                 domain to
>
>                 influence the detail configurations on each autonomic
>                 node.
>
>                          As intent is described as an abstract,
>                 declarative,
>
>                 high-level policy used to operate an autonomic domain,
>                 such as an
>
>         enterprise network.
>
>                 Perhaps we can rename the “intent†in the
>                 “prefix-management ID†to
>
>                 some other words.
>
>                 Is “ASA parameters" ok here?
>
>                 I think some characteristics for this kind of
>                 low-level policies are
>
>                 1.       Network level parameters, need to be
>                 distributed by GRASP and
>
>         ACP
>
>                 2.       Relatively static compared to intent, perhaps
>                 only needed to be
>
>                 configured once
>
>                 3.       Unavoidable in current network environment,
>                 mostly for
>
>                 establishing the network infrastructure
>
>                 4.       Rarely need coordinations with others
>                 parameters, usually is
>
>                 closed related to one specific autonomic function
>
>                 Welcome for comment. Your suggestions will be appreciated.
>
>                 Best Regards
>
>                 Zongpeng Du
>
>                 _______________________________________________
>
>                 Anima mailing list
>
>                 Anima@ietf.org 
>
>                 https://www.ietf.org/mailman/listinfo/anima
>
>             _______________________________________________
>
>             Anima mailing list
>
>             Anima@ietf.org 
>
>             https://www.ietf.org/mailman/listinfo/anima
>
> -- 
>
> Laurent Ciavaglia
>
> Senior Research Manager
>
> Bell Labs, Nokia
>
> +33 160 402 636
>
> Route de Villejust | 91620 Nozay | France
>
> LinkedIn: laurentciavaglia 
>

-- 

Laurent Ciavaglia

Senior Research Manager

Bell Labs, Nokia

+33 160 402 636

Route de Villejust | 91620 Nozay | France

LinkedIn: laurentciavaglia 


--------------020903040400010904040802
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Thanks Michael (B.), I was wrong in my
      understanding.

      

      Laurent.

    

    
On 29/03/2016 18:34, EXT Michael
      Behringer (mbehring) wrote:

    

    

      
      
      
      

        
When
              I say “freeze†I mean: All the nodes that the operator
              expected to join, have joined; the next ones will come in
              6 months. Until then, I want to disable the enrolment
              function. 

        
Obviously,
              it *could* be
              understood as: for today, there are no more installations
              of new devices expected; tomorrow at 9 I expect an
              installation, so I unfreeze the network, once that device
              has enrolled I freeze again until 10, when the next one it
              expected. That would not be the right thing.
              

        
I
              guess that’s exactly the same for all examples... It
              depends how you use them.
              

        
Michael

        
 

        

          

            

              

                From: Laurent Ciavaglia
                    [mailto:laurent.ciavaglia@nokia.com] 

                    Sent:
                    29 March 2016 17:52

                    To:
                    Michael Behringer (mbehring)
                    <mbehring@cisco.com>; Brian E Carpenter
                    <brian.e.carpenter@gmail.com>; John Strassner
                    <strazpdj@gmail.com>; Duzongpeng
                    <duzongpeng@huawei.com>

                    Cc:
                    Laurent Ciavaglia
                    <laurent.ciavaglia@nokia.com>; Jéferson Campos
                    Nobre <jcnobre@inf.ufrgs.br>; anima@ietf.org;
                    Sheng Jiang <jiangsheng@huawei.com>

                    Subject:
                    Re: Another use case for Intent: "Freeze network
                    enrolment"

            

          

          
 

          
Michael,

                

                It seems to me that "Freeze network enrollment" is a
                valid declarative form of policy (intent), yet it seems
                to be a counter-example to the argument of long-lasting
                intent, i.e. the operator will want to roll-back to
                "normal" behavior as soon as the issue that raised the
                "freeze" intent creation is cleared...

                (I am not saying that the case is not a good one to
                investigate).

                

                Am I over-/mis-interpreting?

                

                Best regards, Laurent.

                

              

          

            
On
                  24/03/2016 14:49, EXT Michael Behringer (mbehring)
                  wrote:

          

          

            
Maybe we should combine two threads here. I think there is a perfect use case for Intent I  mentioned in another mail, which would be even applicable to what we do today: 

            
 

            
"Freeze network enrolment". 

            
 

            
Since any node can be a proxy node for bootstrap, it makes perfect sense to flood this to all nodes. 

            
 

            
Local override possible via traditional config / netconf / etc. 

            
 

            
No parameters needed. Can't get any simpler. 

            
 

            
Michael

            
 

            
 

            

              
-----Original Message-----

              
From: Brian E Carpenter [mailto:brian.e.carpenter@gmail.com]

              
Sent: 24 March 2016 05:53

              
To: John Strassner <strazpdj@gmail.com>; Duzongpeng

              
<duzongpeng@huawei.com>

              
Cc: Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; Jéferson Campos Nobre

              
<jcnobre@inf.ufrgs.br>; anima@ietf.org; Michael Behringer (mbehring)

              
<mbehring@cisco.com>; Sheng Jiang <jiangsheng@huawei.com>

              
Subject: Re: [Anima] Does Autonomic network need some intervention

              
beyond intent

              
 

              
On 24/03/2016 16:49, John Strassner wrote:

              

                
I am confused as to why we would consider prefix-management as an

                
autonomic intent.

              

              
 

              
It's the other way round. Prefix management was chosen as an autonomic

              
use case, so a follow-up question is: what intent is appropriate for that use

              
case?

              
 

              
And the suggested answer is: the ISP's intent is that certain types of edge

              
network should get certain sizes of prefix. How the prefixes are actually

              
divided up and assigned is left to the autonomic nodes.

              
 

              

                
To me, this breaks a fundamental rule of what is intent by requiring

                
the intent writer to understand too many details about the target of

                
the intent. The analogy (at least in the ONF and TMF) was specifying

                
to a taxi company that you want a taxi, but you need to take a certain

                
route, because you (the customer) know traffic patterns better than

                
the taxi driver.

                
 

                
Surely we could pick a more appropriate example?

              

              
 

              
By all means describe other use cases and what sort of intent they need.

              
 

              
    Brian

              

                
 

                
 

                
regards,

                
John

                
 

                
On Wed, Mar 16, 2016 at 5:41 AM, Duzongpeng

              

              
<duzongpeng@huawei.com> wrote:

              

                
 

                

                  
Hi, Michael,

                  
 

                  
 

                  
 

                  
         Thanks for the reply. I can understand that intent is not in

                  
the current milestones, and you are working on several milestones drafts

                

              

              
now.

              

                

                  
 

                  
 

                  
 

                  
         You are welcome to contribute to the intent draft at any

                  
time, and thanks in advance.

                  
 

                  
 

                  
 

                  
         According to your statements, in my understanding, we can

                  
keep regarding the “intent†example in “prefix-management ID†as an

                  
Autonomic intent.

                  
 

                  
         It’s semantics are just like “all nodes of type x do this;

                  
type y does thatâ€, so it is a high level policy.

                  
 

                  
 

                  
 

                  
         It really helps a lot. Thanks.

                  
 

                  
 

                  
 

                  
         I generally agree with your statement, while some other

                  
personal understandings exist (shown below). But perhaps they can be

                

              

              
discussed later.

              

                

                  
 

                  
 

                  
 

                  
Best

                  
Regards

                  
 

                  
Zongpeng Du

                  
 

                  
 

                  
 

                  
*From:* Michael Behringer (mbehring) [mailto:mbehring@cisco.com]

                  
*Sent:* Wednesday, March 16, 2016 4:17 PM

                  
*To:* Duzongpeng; anima@ietf.org

                  
*Cc:* Laurent Ciavaglia; Jéferson Campos Nobre; Sheng Jiang

                  
*Subject:* RE: [Anima] Does Autonomic network need some intervention

                  
beyond intent

                  
 

                  
 

                  
 

                  
Hi Zongpeng,

                  
 

                  
 

                  
 

                  
Yes, this is a recurring question and topic. Let me give my view on

                  
two

                  
levels:

                  
 

                  
 

                  
 

                  
 

                  
First, ANIMA isn’t the only thing “influencing†or “steering†a

                  
network, at least not initially. There is still traditional config,

                  
plus all the existing and emerging SDN models. My view has always

                  
been that ANIMA sets a “default†policy in a network, and all other

                  
means are more specific and can override intent. This is described in

                  
RFC7575. Main point here: ANIMA Intent isn’t the only tool in a network.

                  
 

                  
 

                  
 

                  
[duzongpeng] I agree that intent can coexist with traditional config.

                  
 

                  
I generally agree with the policy priorities, but I have some other

                  
concerns.

                  
 

                  
Some default policies/parameters set through “intent†perhaps are

                  
fundamental for the uplayer services.

                  
 

                  
It should be prevented that human operator uses some CLIs to change

                  
them by mistake, and the uplayer services are interrupted.

                  
 

                  
At least, some warning should be reported to the human operator.

                  
[duzongpeng]

                  
 

                  
 

                  
 

                  
Intent could perfectly well cover a high level policy such as “all

                  
nodes of type x do this; type y does thatâ€. (Which I think is what

                  
you mention

                  
below.) When we started with Intent, the vision was to keep it very

                  
high level, more along the line “do the right thing†J  It has become

                  
clear that in today’s networking, people will want to push more

                  
“parameters†than “policyâ€, and I think we’ll just have to acknowledge

                

              

              
that.

              

                

                  
 

                  
 

                  
 

                  
Personally I think it’ll pan out like this: Intent will have sections

                  
per autonomic function. An Intent parser will validate overall

                  
correctness (signature, etc), then segment the Intent into its bits:

                  
One part will be the autonomic fabric itself, then there will be bits

                  
for each autonomic function.

                  
 

                  
 

                  
 

                  
So the Intent parser on a node will grab intent, see there is some

                  
Intent for autonomic function A, will see whether this is running

                  
locally, and if so, it’ll just pass that piece of Intent to that function.

                  
 

                  
 

                  
 

                  
In this way, a function owner defines both the intent itself, and how

                  
it is consumed by his/her function.

                  
 

                  
 

                  
 

                  
[duzongpeng]Agree.[duzongpeng]

                  
 

                  
 

                  
 

                  
The downside is that there is no “control†an//d people can do what

                  
they want, in the worst case even push config, which is clearly NOT

                  
what Intent is meant to do. I’d hope that people don’t start writing

                  
Intent like “node

                  
17: here is your CLI; node 23: here is your CLIâ€; node 37: here is

                  
your CLIâ€. But what I describe above COULD be abused in this way.

                  
RFC7575 says this is NOT Intent.

                  
 

                  
 

                  
 

                  
[duzongpeng]Agree. It is hard to give an exact boundary.

                  
 

                  
However, I think we have two choices here.

                  
 

                  
The first one is to define a policy first in every ASA, but with some

                  
open parameters that can be configured. What we need to do is just to

                  
distribute the network-level parameters for that function. This way

                  
is suitable to some relatively static functions.

                  
 

                  
The second is more advanced, in the ideal case, the human operator

                  
can communicate with network using natural language or something

                  
similar. This way is more flexible, but harder to realize.

                  
 

                  
[duzongpeng]

                  
 

                  
 

                  
 

                  
In other words: Let’s not be too prescriptive, and hope that folks

                  
will do more or less reasonable things.

                  
 

                  
 

                  
 

                  
[duzongpeng]Agree.[duzongpeng]

                  
 

                  
 

                  
 

                  
I’ve been meaning to comment on the Intent drafts and possibly

                  
contribute, I’m just struggling with priorities at the moment (as you

                  
can see by my relative silence over the past weeks).

                  
 

                  
 

                  
 

                  
Michael

                  
 

                  
 

                  
 

                  
 

                  
 

                  
 

                  
 

                  
*From:* Anima [mailto:anima-bounces@ietf.org

                  
<anima-bounces@ietf.org>] *On Behalf Of *Duzongpeng

                  
*Sent:* 16 March 2016 02:38

                  
*To:* anima@ietf.org

                  
*Cc:* Laurent Ciavaglia <laurent.ciavaglia@nokia.com>; Jéferson

                  
Campos Nobre <jcnobre@inf.ufrgs.br>; Sheng Jiang

                  
<jiangsheng@huawei.com>

                  
*Subject:* [Anima] Does Autonomic network need some intervention

                  
beyond intent

                  
 

                  
 

                  
 

                  
Hello, everyone

                  
 

                  
 

                  
 

                  
         I understand that Autonomic network should be able to work

                  
with as less configuration as possible. In the ideal case, an

                  
Autonomic network should be able to work well while only intent is

                  
needed from human operators.

                  
 

                  
 

                  
 

                  
         However, I wonder whether Autonomic network may need some

                  
intervention from human operators through some low-level policies

                  
beyond intents.

                  
 

                  
 

                  
 

                  
         For example, in the

                  
https://tools.ietf.org/html/draft-ietf-anima-prefix-management-00, it

                  
is suggested that the prefix lengths for the CSG, ASG, RSG (different

                  
roles in IP RAN) can be assigned as an "intent". These configuration

                  
parameters need to be distributed in the autonomic domain to

                  
influence the detail configurations on each autonomic node.

                  
 

                  
 

                  
 

                  
         As intent is described as an abstract, declarative,

                  
high-level policy used to operate an autonomic domain, such as an

                

              

              
enterprise network.

              

                

                  
Perhaps we can rename the “intent†in the “prefix-management ID†to

                  
some other words.

                  
 

                  
 

                  
 

                  
Is “ASA parameters" ok here?

                  
 

                  
 

                  
 

                  
I think some characteristics for this kind of low-level policies are

                  
 

                  
1.       Network level parameters, need to be distributed by GRASP and

                

              

              
ACP

              

                

                  
 

                  
2.       Relatively static compared to intent, perhaps only needed to be

                  
configured once

                  
 

                  
3.       Unavoidable in current network environment, mostly for

                  
establishing the network infrastructure

                  
 

                  
4.       Rarely need coordinations with others parameters, usually is

                  
closed related to one specific autonomic function

                  
 

                  
 

                  
 

                  
Welcome for comment. Your suggestions will be appreciated.

                  
 

                  
 

                  
 

                  
 

                  
 

                  
Best Regards

                  
 

                  
Zongpeng Du

                  
 

                  
_______________________________________________

                  
Anima mailing list

                  
Anima@ietf.org

                  
https://www.ietf.org/mailman/listinfo/anima

                  
 

                  
 

                

                
 

                
 

                
 

                
 

                
_______________________________________________

                
Anima mailing list

                
Anima@ietf.org

                
https://www.ietf.org/mailman/listinfo/anima

                
 

              

            

            
 

          

          
 

          

            

              -- 

                  

                

            

              Laurent Ciavaglia

            

              Senior Research Manager

            

              Bell Labs, Nokia

            

               

            

              +33 160 402 636
                

            

              Route de Villejust | 91620
                  Nozay | France

            

              LinkedIn:
                laurentciavaglia

            
 

          

        

      

    

    

    
-- 

      
      
      
      
      
      
      
      
      
      
      

        
Laurent
            Ciavaglia

        
Senior
Research
            Manager

        
Bell
Labs,
            Nokia

        
 

        
+33 160 402 636 

        
Route de Villejust
            | 91620 Nozay
            | France

        
LinkedIn: laurentciavaglia

        
 

      

    

  


--------------020903040400010904040802--


From nobody Tue Mar 29 17:13:03 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 86D0E12DA45 for ; Tue, 29 Mar 2016 17:13:02 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id y3-PmYJZ8WUI for ; Tue, 29 Mar 2016 17:13:00 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 66E2412D120 for ; Tue, 29 Mar 2016 17:13:00 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id 7C41426C050C2; Wed, 30 Mar 2016 00:12:54 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2U0CvdE003552 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Wed, 30 Mar 2016 00:12:58 GMT
Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2U0CmEj002812 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 30 Mar 2016 02:12:55 +0200
Received: from [135.224.216.185] (135.239.27.38) by FR712WXCHHUB03.zeu.alcatel-lucent.com (135.239.2.74) with Microsoft SMTP Server (TLS) id 14.3.195.1; Wed, 30 Mar 2016 02:12:48 +0200
To: EXT Brian E Carpenter , "Michael Behringer (mbehring)" , John Strassner , Duzongpeng 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FB0468.9010404@gmail.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56FB19FF.2010602@alcatel-lucent.com>
Date: Wed, 30 Mar 2016 02:12:47 +0200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <56FB0468.9010404@gmail.com>
Content-Type: multipart/alternative; boundary="------------030307060105010104030703"
X-Originating-IP: [135.239.27.38]
Archived-At: 
Cc: Laurent Ciavaglia , =?UTF-8?Q?J=c3=a9ferson_Campos_Nobre?= , "anima@ietf.org" , Sheng Jiang 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 30 Mar 2016 00:13:02 -0000

--------------030307060105010104030703
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit

Brian,


On 30/03/2016 00:40, EXT Brian E Carpenter wrote:
> Without disrespect to the subsequent arguments, I think this is a key
> point. At the moment we're designing the ANI, not the Intent system,
> so what I'm looking for in this discussion is the features that the ANI
> has to support - which seem to be
> (a) ability to distribute the complete Intent to all nodes
> (b) ability to distribute partial Intent to all (interested) nodes later on

The perimeter of the ANI is arbitrary / matter of convention. We decide 
something for ANI in ANIMA, (many) others have defined their own scope 
for their autonomic architectures.
There are several surveys on that compare functionality of autonomic 
network architectures. The fact that some functionality are found in 
every approaches for more than a decade could be interpreted as a good hint.

In order to understand the features the ANI has to support, we need to 
understand what could/should be factorized in the ANI.
At least for me, this means to understand a bit more how the "intent" 
system would/could work end-to-end.

Also, RFC 7575, section 3.8 Common Autonomic Networking Infrastructure 
mentions:
"This Autonomic Networking Infrastructure may contain common control and 
management functions such as messaging, service discovery, negotiation, 
Intent distribution, self-monitoring, and diagnostics, etc.  A common 
approach to define and manage Intent is also required.
All the components around the "Autonomic Service Agents" should be 
common components, such that the Autonomic Service Agents do not have to 
replicate common tasks individually.

Thanks, Laurent.
>
> BTW I put "interested" in parenthesis because it seems dangerous to
> assume that a node's interests don't change. As ASAs start or stop
> in a given node, its interest in Intent will change. So unless we
> construct a quite elaborate pub/sub system, flooding is needed,
> even for partial intent. That's what GRASP supports, of course.

--------------030307060105010104030703
Content-Type: text/html; charset="windows-1252"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Brian,

    

    

    
On 30/03/2016 00:40, EXT Brian E
      Carpenter wrote:

    

    

      
Without disrespect to the subsequent arguments, I think this is a key
point. At the moment we're designing the ANI, not the Intent system,
so what I'm looking for in this discussion is the features that the ANI
has to support - which seem to be
(a) ability to distribute the complete Intent to all nodes
(b) ability to distribute partial Intent to all (interested) nodes later on

    

    

    The perimeter of the ANI is arbitrary / matter of convention. We
    decide something for ANI in ANIMA, (many) others have defined their
    own scope for their autonomic architectures.

    There are several surveys on that compare functionality of autonomic
    network architectures. The fact that some functionality are found in
    every approaches for more than a decade could be interpreted as a
    good hint.

    

    In order to understand the features the ANI has to support, we need
    to understand what could/should be factorized in the ANI.

    At least for me, this means to understand a bit more how the
    "intent" system would/could work end-to-end.

    

    Also, RFC 7575, section 3.8 Common Autonomic Networking
    Infrastructure mentions:

    "This Autonomic Networking Infrastructure may contain common control
    and management functions such as messaging, service discovery,
    negotiation, Intent distribution, self-monitoring, and diagnostics,
    etc.  A common approach to define and manage Intent is also
    required.

    All the components around the "Autonomic Service Agents" should be
    common components, such that the Autonomic Service Agents do not
    have to replicate common tasks individually.

    

    Thanks, Laurent.

    

      

BTW I put "interested" in parenthesis because it seems dangerous to
assume that a node's interests don't change. As ASAs start or stop
in a given node, its interest in Intent will change. So unless we
construct a quite elaborate pub/sub system, flooding is needed,
even for partial intent. That's what GRASP supports, of course.


    

  


--------------030307060105010104030703--


From nobody Tue Mar 29 17:51:36 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 07A6412D165 for ; Tue, 29 Mar 2016 17:51:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id sDXMJTpx_wIY for ; Tue, 29 Mar 2016 17:51:33 -0700 (PDT)
Received: from mail-pf0-x234.google.com (mail-pf0-x234.google.com [IPv6:2607:f8b0:400e:c00::234]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6E60312D120 for ; Tue, 29 Mar 2016 17:51:33 -0700 (PDT)
Received: by mail-pf0-x234.google.com with SMTP id 4so28048141pfd.0 for ; Tue, 29 Mar 2016 17:51:33 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=to:from:subject:organization:message-id:date:user-agent :mime-version:content-transfer-encoding; bh=DXA6XnPEgZRyQWsC+PqRGzSxYirl+7wf6iDF+IoNm3A=; b=IFSmCOHhN29eUp8T5KaxBYUz0RueJpCIjVz8wU/GhY6oVwTeIRG2xP8p3j0L+rI4X0 6M5rSkLi1nf/iIbtCvH8aq0hlVuWqT80+g7W/3QPds6UfUqWnXgbESpDyuHErZ1v/q2z 7g8lJKau38cLUaaizfz7oAUvk7wT4VG5DMIy0YFDFilI3UhegU/Yx34SqpJrE/BzZfGK MqbSm916dH0rU3nSl7GTcst3esv/8UoYKDBCdqbe7pm1rTTUpKEabDObJAwGydGyboya Zuk/Yb4sCvRDEuGcF1fa0pQ1xsSnmtV9IvilVHWd228nn0QzTgsPPcZLkXL93QFTPRfb Eh2w==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:to:from:subject:organization:message-id:date :user-agent:mime-version:content-transfer-encoding; bh=DXA6XnPEgZRyQWsC+PqRGzSxYirl+7wf6iDF+IoNm3A=; b=giv1JSMAKtR8h3C7nSUmEmVDToWGhR01kgNwhwZdy2GUYEeywYsPFZEQxvkHfV27vl PRcZjzpaNoby8t5Xs0tufLYUvXvkHO+jN8soCwOeLQaXUoUV9NJlawKsOFFwnuDrOibD Ts3lG01qyrHBUX1UcF6BPEacC8OQJSfZQmvN1Fx/cbHd9WO/J7NH0wua4bwj7reMmvwM 7bzjOEBf74+eDmRaxZ1Yid/lUQK9Qt/43GNWkPDnZNXq5UnwBcxqmWX5Lf51Z5W81vpt HIqR+BmSf6NApajNLpUGzQIydbk+eo/0FIPYlQGT3+LDSBQ5sI+luqQTSqEIfl1udLZq Ro/A==
X-Gm-Message-State: AD7BkJInjPGZa8NCj5GkqkNM46eeB9+BcJ/uKAuDwnfAvnMYZFHBpvQAhlKoL1yp2H3C/g==
X-Received: by 10.98.71.92 with SMTP id u89mr8246815pfa.100.1459299092954; Tue, 29 Mar 2016 17:51:32 -0700 (PDT)
Received: from ?IPv6:2001:df0:0:2006:c0da:ac17:5f6d:8e76? ([2001:df0:0:2006:c0da:ac17:5f6d:8e76]) by smtp.gmail.com with ESMTPSA id ez6sm1054865pab.12.2016.03.29.17.51.30 for  (version=TLSv1/SSLv3 cipher=OTHER); Tue, 29 Mar 2016 17:51:31 -0700 (PDT)
To: Anima WG 
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56FB2315.3030104@gmail.com>
Date: Wed, 30 Mar 2016 13:51:33 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: [Anima] Draft text on infrastructure ASAs
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 30 Mar 2016 00:51:35 -0000

Hi,

In various threads people have been asking about how components
of the ANI (autonomic networking infrastructure) could themselves
make use of GRASP. This has come up for the secure bootstrap,
formation of the ACP, and Intent distribution.

I have quickly scribbled a draft document on this. It's too late
for the IETF, and I won't be in BA anyway. Also, it's not in I-D
format for now, because it really benefits from a choice of fonts.

It's in the WG wiki:
https://trac.tools.ietf.org/wg/anima/trac/wiki/Infrastructure%20ASAs
so please have a look and more important, post your comments here.

Regards
   Brian


From nobody Tue Mar 29 23:52:20 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 41BE312DBF1 for ; Tue, 29 Mar 2016 23:52:19 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KQYf66wHbQmi for ; Tue, 29 Mar 2016 23:52:18 -0700 (PDT)
Received: from rcdn-iport-9.cisco.com (rcdn-iport-9.cisco.com [173.37.86.80]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D4D8012DD50 for ; Tue, 29 Mar 2016 23:52:17 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=2390; q=dns/txt; s=iport; t=1459320737; x=1460530337; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=HamAXmpTYaQYBsArYyzU4dfBrAGeWXq7f2M7Cj9Mjkw=; b=UZgpvpX3+oseiYLvw6viKi5K54UlLyfXzbOjOPZ+tK52MSxuCk44vSm1 n4b1ygeVpXKqvlde4JHLgcZpWCK9c/Suj0ggH/MvA+MudFWAQbfg67+3P op1BZZOg7MH/HzX3MyPpFsha0vDgHNQFxCzLcgXKdSljwK8nYqxmNlTXK Q=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0D1AQAmd/tW/5xdJa1dgzSBUAa6egENg?= =?us-ascii?q?XCGDQIcgSQ4FAEBAQEBAQFkJ4RBAQEBAwEjEUoHBAIBCBEEAQEBAgIREgMCAgI?= =?us-ascii?q?wFAEICAIEARIIiBcIsAWQeAEBAQEBAQEBAQEBAQEBAQEBAQEBFnyFIoRFhB8BA?= =?us-ascii?q?VyCQYJWBZdsAY1/jxSPDgEeAQFCg2Vshws2fgEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,415,1454976000"; d="scan'208";a="85859049"
Received: from rcdn-core-5.cisco.com ([173.37.93.156]) by rcdn-iport-9.cisco.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 30 Mar 2016 06:52:16 +0000
Received: from XCH-ALN-009.cisco.com (xch-aln-009.cisco.com [173.36.7.19]) by rcdn-core-5.cisco.com (8.14.5/8.14.5) with ESMTP id u2U6qHeA026201 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Wed, 30 Mar 2016 06:52:17 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-009.cisco.com (173.36.7.19) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Wed, 30 Mar 2016 01:52:16 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Wed, 30 Mar 2016 01:52:16 -0500
From: "Michael Behringer (mbehring)" 
To: Michael Richardson , "anima@ietf.org" 
Thread-Topic: [Anima] Another use case for Intent: "Freeze network enrolment"
Thread-Index: AQHRidMDO/a05lCyz0ONkyiIPytX+J9wk74wgACE1YCAAHPn8A==
Date: Wed, 30 Mar 2016 06:52:16 +0000
Message-ID: <24c1d1f90d2a4443bfc6c7eab1f491be@XCH-RCD-006.cisco.com>
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56FAA4B7.9010305@alcatel-lucent.com> <58704b9ce2cb43c0b07d9aac8b284187@XCH-RCD-006.cisco.com> <14863.1459277575@obiwan.sandelman.ca>
In-Reply-To: <14863.1459277575@obiwan.sandelman.ca>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.55.238.139]
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 30 Mar 2016 06:52:19 -0000
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From nobody Wed Mar 30 09:10:28 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AF73C12D7B4 for ; Wed, 30 Mar 2016 09:10:25 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -5.939
X-Spam-Level: 
X-Spam-Status: No, score=-5.939 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_FONT_FACE_BAD=0.981, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id AO3K8nuMpJt0 for ; Wed, 30 Mar 2016 09:10:23 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D5BF312D78B for ; Wed, 30 Mar 2016 09:10:22 -0700 (PDT)
Received: from fr712umx3.dmz.alcatel-lucent.com (unknown [135.245.210.42]) by Websense Email Security Gateway with ESMTPS id 952E8BFDF5D5C; Wed, 30 Mar 2016 16:10:17 +0000 (GMT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (fr711usmtp1.zeu.alcatel-lucent.com [135.239.2.122]) by fr712umx3.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2UGAK5o029307 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Wed, 30 Mar 2016 16:10:20 GMT
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id u2UGAJrm019537 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 30 Mar 2016 18:10:19 +0200
Received: from [135.224.216.185] (135.239.27.41) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Wed, 30 Mar 2016 18:10:19 +0200
From: Laurent Ciavaglia 
Organization: Bell Labs
To: anima 
Message-ID: <56FBFA6A.4010400@alcatel-lucent.com>
Date: Wed, 30 Mar 2016 18:10:18 +0200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="------------070203090008010604040808"
X-Originating-IP: [135.239.27.41]
Archived-At: 
Cc: Duzongpeng , Laurent Ciavaglia , Michael Behringer , Pierre Peloso 
Subject: [Anima] ANIMA intent discussion
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Wed, 30 Mar 2016 16:10:25 -0000

--------------070203090008010604040808
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Transfer-Encoding: 8bit

Dear all,

Michael, Zongpeng, Pierre and I had a discussion on ANIMA intent this 
morning, trying to clarify some points and progress toward a common 
understanding.

We used a small, partial set of questions and examples in support.

Below you may find a summary for your information. Comments and further 
interactions are most welcome.

Thanks,
Michael, Zongpeng, Pierre and Laurent.
---

--- Questions
1-Who writes intents?
2-How many intents?
3-How many domains?
4-What are the intent levels/hierarchy?
5-Where/by what is intent processed/compiled?
6-Flooding: what are the requirements? Propagation among who? How 
distributed? How frequent? -> if many “overlapping†intents, (partial) 
updates may become frequent.
7-How is intent understood by node/ASA? -> model, dictionary...
8-Can an ASA write an intent for another ASA? What about coordination 
or knowledge writing/receiving intents? Are these functions ASAs/other 
things..?


--- Intent examples
A-Do the right thing
B-Freeze network enrollment
C-Arrange VM guest distribution so that (CPU) utilization is < 70%
D-Assign prefixes to RAN nodes
E-Protect premium users traffic
F-Maximize energy savings


--- Discussion
1) Who writes intents?
    Laurent's view:
     Intents are issued by a management function. Can be human / via 
BSS/OSS.
    Michael's view:
     Origin of intent is "non-technical". i.e., registrar doesn't issue 
intent.
_Discussion:_ how to set the cursor on non-technical? In examples D or 
E, the intent writer has some rough knowledge/understanding of technical 
aspects of a network (wireless and address aspects for example D ; 
protection/restoration and traffic engineering for example E).
_Consensus:_ Intent is originated by human/at OSS level, NOT by a 
network device / function. (of course Intent can be ingested / inputted 
on a specific device)


2) How many intents?
    Laurent's view:
     There may be different ones, like "freeze network", "optimize 
energy savings", cf examples. All are valid and concurrent.
     Can we wrap that into one file?
     How do we share that?

_Discussion__:_
     We cannot define Intent too precisely today.
     But we can define where Intent is coming from and how to distribute 
it.

     Can Intent have targets?
         - Ex: an intent for traffic engineering and an intent for 
energy consumption optimization may need specific (same or different) 
targets. the functions could have different objectives per set of targets.
         - Ex: two network segments, core and metro. Then intent should 
be able to say "on core: optimize on availability"; on access: "optimize 
on energy saving."
         - Therefore we need scopes.

      How do you define "target"?
         - Is a role enough?
         - Pierre: Should be more than role. Example: Metro and core. 
Need to distinguish between different metro networks for example.
         - Michael: If we have sub-domains, and define intent per 
sub-domain, would that do?
         - Laurent: a combination of roles and (sub-)domain(s) should be 
able to cover most cases.

_Consensus:_ a combination of roles and (sub-)domain(s) should be able 
to cover most cases.


8) Can an ASA write an intent?
     Pierre: Example is the coordination function: that function could 
send out intent directly to modify the behavior of a group of ASAs.
      Multiple ASAs could do conflicting things. The coordination 
function must resolve that. Should it issue/modify intent itself?
     Laurent: q1: should it output something in the form of an intent? 
q2: should the coordination function update intent?
     Michael: Isn't it that the behavior of the coordination function 
must also be resolved by a human originated intent? i.e. the conflicting 
functions inform the operator via a feedback loop and the operator 
adjust the intents.
     Laurent: This is possible / complementary. In once case, the 
operator is the one closing the control loop / changing the behaviors 
via intent updates. With a coordination function, part/all of that 
process could be automated/driven by the machine (the coordination 
function closes the control loop, in the control range allowed by the 
operator). There may be intermediate steps where unforeseen conflicts 
will be resolved. This is run time conflict resolution.
     Michael: Do we foresee a case where networks resolve conflicts 
without human control?
     Laurent: this is possible (has been demonstrated in some projects). 
We should allow for both approaches to work, possibly in a phased approach.

     When unforeseen conflicts arise, ASAs can signal to each other, but 
they will not issue Intent themselves.

_Consensus:_ Cannot say today if other functions can/are allowed to 
issue intent / intent updates. Functions/ASAs may have to *signal* or 
*message* somehow, but that this isn’t considered     intent yet.
     For further study: try to design the intent system so that both 
approaches (feedback loop, coordination) are possible. More 
investigation/analysis needed.


3) How many domains?
Naming issue: "domain" by Laurent is a high level concept; could be 
understood as a DNS domain name. Needs to be clear what we're talking 
about.
     With domain-name based intent (i.e., access.att.net) and roles 
(e.g, access devices) we should be able to express what we need for now. 
Not 100% clear for now.
     Two types of domains: (administrative, technology) domains defined 
(a priori) by the operator e.g. RAN, MAN, WAN. and the domain of 
application of a policy i.e. resolving the policy defines a set of 
target entities to which teh policy applies, this is the domain of 
application of the given policy.

_Consensus:_ Need to clarify wording to avoid confusion.


4)What are the intent levels/hierarchy?
     Related to policy based manamegent model: business level, service 
level, network level, device level, etc. So far in ANIMA we consider 
only two levels: User and ASA.
     Do we need a service level? e.g, VPN.
     Can we express different levels with using roles? So a role could 
be very specific (route reflector), or service related (e.g., VPN 
head-end)?
     But then we may need to nest roles?!
     Maybe we need "business" "service" "network (ASA)" level?

_Consensus:_ to be further discussed.


Questions 5, 6, 7 were not discussed due to lack of time.
---


--------------070203090008010604040808
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: 8bit


  
    
  
  
    Dear all,

      

      Michael, Zongpeng, Pierre and I had a discussion on ANIMA intent
      this morning, trying to clarify some points and progress toward a
      common understanding.

      

      We used a small, partial set of questions and examples in support.
      

      

      Below you may find a summary for your information. Comments and
      further interactions are most welcome.

      

      Thanks, 

      Michael, Zongpeng, Pierre and Laurent.

      ---

      

      --- Questions

      1-Who writes intents? 

      2-How many intents? 

      3-How many domains? 

      4-What are the intent levels/hierarchy? 

      5-Where/by what is intent processed/compiled? 

      6-Flooding: what are the requirements? Propagation among who? How
      distributed? How frequent? -> if many “overlapping†intents,
      (partial) updates may become frequent. 

      7-How is intent understood by node/ASA? -> model, dictionary...
      

      8-Can an ASA write an intent for another ASA? What about
      coordination or knowledge writing/receiving intents? Are these
      functions ASAs/other things..?

      

      

      --- Intent examples

      A-Do the right thing

      B-Freeze network enrollment

      C-Arrange VM guest distribution so that (CPU) utilization is <
      70%

      D-Assign prefixes to RAN nodes

      E-Protect premium users traffic

      F-Maximize energy savings

      

      

      --- Discussion

      1) Who writes intents? 

         Laurent's view: 

          Intents are issued by a management function. Can be human /
      via BSS/OSS.

         Michael's view: 

          Origin of intent is "non-technical". i.e., registrar doesn't
      issue intent.

         Discussion: how to set the cursor on non-technical? In
      examples D or E, the intent writer has some rough
      knowledge/understanding of technical aspects of a network
      (wireless and address aspects for example D ;
      protection/restoration and traffic engineering for example E).

         Consensus: Intent is originated by human/at OSS level,
      NOT by a network device / function. (of course Intent can be
      ingested / inputted on a specific device)

      

              

      2) How many intents? 

         Laurent's view: 

          There may be different ones, like "freeze network", "optimize
      energy savings", cf examples. All are valid and concurrent. 

          Can we wrap that into one file? 

          How do we share that? 

      

          Discussion:

          We cannot define Intent too precisely today. 

          But we can define where Intent is coming from and how to
      distribute it. 

      

          Can Intent have targets? 

              - Ex: an intent for traffic engineering and an intent for
      energy consumption optimization may need specific (same or
      different) targets. the functions could have different objectives
      per set of targets. 

              - Ex: two network segments, core and metro. Then intent
      should be able to say "on core: optimize on availability"; on
      access: "optimize on energy saving."

              - Therefore we need scopes. 

      

           How do you define "target"? 

              - Is a role enough? 

              - Pierre: Should be more than role. Example: Metro and
      core. Need to distinguish between different metro networks for
      example. 

              - Michael: If we have sub-domains, and define intent per
      sub-domain, would that do? 

              - Laurent: a combination of roles and (sub-)domain(s)
      should be able to cover most cases. 

      

          Consensus: a combination of roles and (sub-)domain(s)
      should be able to cover most cases. 

      

      

      8) Can an ASA write an intent? 

          Pierre: Example is the coordination function: that function
      could send out intent directly to modify the behavior of a group
      of ASAs. 

           Multiple ASAs could do conflicting things. The coordination
      function must resolve that. Should it issue/modify intent itself?
      

          Laurent: q1: should it output something in the form of an
      intent? q2: should the coordination function update intent? 

          Michael: Isn't it that the behavior of the coordination
      function must also be resolved by a human originated intent? i.e.
      the conflicting functions inform the operator via a feedback loop
      and the operator adjust the intents.

          Laurent: This is possible / complementary. In once case, the
      operator is the one closing the control loop / changing the
      behaviors via intent updates. With a coordination function,
      part/all of that process could be automated/driven by the machine
      (the coordination function closes the control loop, in the control
      range allowed by the operator). There may be intermediate steps
      where unforeseen conflicts will be resolved. This is run time
      conflict resolution. 

          Michael: Do we foresee a case where networks resolve conflicts
      without human control?  

          Laurent: this is possible (has been demonstrated in some
      projects). We should allow for both approaches to work, possibly
      in a phased approach.

      

          When unforeseen conflicts arise, ASAs can signal to each
      other, but they will not issue Intent themselves. 

      

          Consensus: Cannot say today if other functions can/are
      allowed to issue intent / intent updates. Functions/ASAs may have
      to *signal* or *message* somehow, but that this isn’t considered
          intent yet.

          For further study: try to design the intent system so that
      both approaches (feedback loop, coordination) are possible. More
      investigation/analysis needed.

      

      

      3) How
        many domains? 

            Naming issue: "domain" by Laurent is a high level
      concept; could be understood as a DNS domain name. Needs to be
      clear what we're talking about. 

          With domain-name based intent (i.e., access.att.net) and roles
      (e.g, access devices) we should be able to express what we need
      for now. Not 100% clear for now. 

          Two types of domains: (administrative, technology) domains
      defined (a priori) by the operator e.g. RAN, MAN, WAN. and the
      domain of application of a policy i.e. resolving the policy
      defines a set of target entities to which teh policy applies, this
      is the domain of application of the given policy.

      

          Consensus: Need to clarify wording to avoid confusion.

      

      

      4) What
        are the intent levels/hierarchy? 

          Related to policy based manamegent model: business level,
      service level, network level, device level, etc. So far in ANIMA
      we consider only two levels: User and ASA. 

          Do we need a service level? e.g, VPN. 

          Can we express different levels with using roles? So a role
      could be very specific (route reflector), or service related
      (e.g., VPN head-end)? 

          But then we may need to nest roles?! 

          Maybe we need "business" "service" "network (ASA)" level?

      

          Consensus: to be further discussed.

      

      

      Questions 5, 6, 7 were not discussed due to lack of time.

      ---

    

  


--------------070203090008010604040808--


From nobody Wed Mar 30 17:49:21 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4F5FD12D0E4 for ; Wed, 30 Mar 2016 17:49:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.911
X-Spam-Level: 
X-Spam-Status: No, score=-1.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eAIKbSvCaptV for ; Wed, 30 Mar 2016 17:49:18 -0700 (PDT)
Received: from tuna.sandelman.ca (tuna.sandelman.ca [IPv6:2607:f0b0:f:3:216:3eff:fe7c:d1f3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 67DC712D0C7 for ; Wed, 30 Mar 2016 17:49:18 -0700 (PDT)
Received: from sandelman.ca (obiwan.sandelman.ca [IPv6:2607:f0b0:f:2::247]) by tuna.sandelman.ca (Postfix) with ESMTP id 5B181203C0 for ; Wed, 30 Mar 2016 20:52:26 -0400 (EDT)
Received: from obiwan.sandelman.ca (localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id 09D6663755 for ; Wed, 30 Mar 2016 20:49:17 -0400 (EDT)
From: Michael Richardson 
To: "anima\@ietf.org" 
In-Reply-To: <56FB0DC6.1030701@gmail.com>
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FAA188.2050901@joelhalpern.com> <56FB0DC6.1030701@gmail.com>
X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2
X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m
Archived-At: 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 00:49:20 -0000

--=-=-=
Content-Type: text/plain


Brian E Carpenter  wrote:
    >> This is also why I worry a bit about flooding to everything.

    > There's a technical issue there, which is that GRASP flooding uses
    > multicast UDP (what else could it reasonably use, anyway?). So if
    > the chunk to be flooded gets too big, we would have to chop it up.

This is why I really think that the URL for the Intent (maybe using the rfc6920
ni:// URLs) should be flooded, and let the nodes ask for the things they
need.

--
Michael Richardson , Sandelman Software Works
 -= IPv6 IoT consulting =-




--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBVvx0CoCLcPvd0N1lAQLDfAf9F00lfUUfSqrqbgG+hx+Mz9r0LNeF7kqd
/DBpYVczvvXg2lWoiChMTYZGB/5Wj766vfV5lu4Rc3WpUxSrhM1jYiCFufw35usU
hjpTqDOSdLQ/zpqDEnuU9xsFV7bV+J1Pvk/Yg3fixRSCEXDdwyefN2KKrg+U+t7L
TqDbvkjtOSh2EpdUbh49pFZm+SkZyqOylisfMffiNuW3eU8eNfgspARTjc3BnQ4K
fALSvvMOGDILHGZtKOEhkpAOqC+R1sIWD7FJ0QXYIXBwxx+rgkgd8wtCNGLQHXtq
VdNi0U9njIITB/KaS1boIa24OPmQrYwhKZSPH8BCIArxYh6G6/lXQg==
=xK0V
-----END PGP SIGNATURE-----
--=-=-=--


From nobody Wed Mar 30 18:23:28 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 638D012D521 for ; Wed, 30 Mar 2016 18:23:26 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.702
X-Spam-Level: 
X-Spam-Status: No, score=-2.702 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_LOW=-0.7, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=joelhalpern.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OSX5AMUGHfdA for ; Wed, 30 Mar 2016 18:23:24 -0700 (PDT)
Received: from maila2.tigertech.net (maila2.tigertech.net [208.80.4.152]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5739E12D558 for ; Wed, 30 Mar 2016 18:23:22 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by maila2.tigertech.net (Postfix) with ESMTP id A8B7D252050; Wed, 30 Mar 2016 18:23:21 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com; s=1.tigertech; t=1459387401; bh=aFod/cvgqMljcsnqdcztUv2bE/XDVQcwc2C/if8ZK5Q=; h=Subject:To:References:From:Date:In-Reply-To:From; b=CH9c/j65Gh0UIb75ohLCWtz0V38JcRpzhosKInZWKuJI9TY9YqNj8y1VL2fXzWmLb NtA2n03Zyv3y+e/JwZWp61rWvjzMgiYw9CJi/o3NTm5fRRXeGkowqvIeHK7nx+PiUd ZHM+8QdHqo/O8ITTy6e2Gk4tcitF2bnMWuAFh5dA=
X-Virus-Scanned: Debian amavisd-new at maila2.tigertech.net
Received: from Joels-MacBook-Pro.local (209-255-163-147.ip.mcleodusa.net [209.255.163.147]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by maila2.tigertech.net (Postfix) with ESMTPSA id 3ACCC240857; Wed, 30 Mar 2016 18:23:21 -0700 (PDT)
To: Michael Richardson , "anima@ietf.org" 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FAA188.2050901@joelhalpern.com> <56FB0DC6.1030701@gmail.com> <16238.1459385357@obiwan.sandelman.ca>
From: "Joel M. Halpern" 
Message-ID: <56FC7C06.9090508@joelhalpern.com>
Date: Wed, 30 Mar 2016 21:23:18 -0400
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:38.0) Gecko/20100101 Thunderbird/38.7.1
MIME-Version: 1.0
In-Reply-To: <16238.1459385357@obiwan.sandelman.ca>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 01:23:26 -0000

If we just passed around URLs (or URIs, or  UR?), how would the ASAs 
know which ones they needed to fetch?

Yours,
Joel

On 3/30/16 8:49 PM, Michael Richardson wrote:
>
> Brian E Carpenter  wrote:
>      >> This is also why I worry a bit about flooding to everything.
>
>      > There's a technical issue there, which is that GRASP flooding uses
>      > multicast UDP (what else could it reasonably use, anyway?). So if
>      > the chunk to be flooded gets too big, we would have to chop it up.
>
> This is why I really think that the URL for the Intent (maybe using the rfc6920
> ni:// URLs) should be flooded, and let the nodes ask for the things they
> need.
>
> --
> Michael Richardson , Sandelman Software Works
>   -= IPv6 IoT consulting =-
>
>
>
>
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>


From nobody Wed Mar 30 19:11:12 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EB76A12D567 for ; Wed, 30 Mar 2016 19:11:09 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.911
X-Spam-Level: 
X-Spam-Status: No, score=-1.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eZWr5jvK5xE7 for ; Wed, 30 Mar 2016 19:11:08 -0700 (PDT)
Received: from tuna.sandelman.ca (tuna.sandelman.ca [209.87.249.19]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 329AD12D549 for ; Wed, 30 Mar 2016 19:11:08 -0700 (PDT)
Received: from sandelman.ca (obiwan.sandelman.ca [IPv6:2607:f0b0:f:2::247]) by tuna.sandelman.ca (Postfix) with ESMTP id 386F2203C0 for ; Wed, 30 Mar 2016 22:14:16 -0400 (EDT)
Received: from obiwan.sandelman.ca (localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id B12E463755 for ; Wed, 30 Mar 2016 22:11:06 -0400 (EDT)
From: Michael Richardson 
To: anima 
In-Reply-To: <56FBFA6A.4010400@alcatel-lucent.com>
References: <56FBFA6A.4010400@alcatel-lucent.com>
X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2
X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m
Archived-At: 
Subject: Re: [Anima] ANIMA intent discussion
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 02:11:10 -0000

--=-=-=
Content-Type: text/plain


These are good questions, thank you for assembling them.

> 1) Who writes intents?

Let me answer with a question: can marketing/sales-people write intents?
(I suspect we all know the answer should be no, but probably we should be
explici that Intents are technical statements.
But consider if an Intent can be written like, "optimize network to provide a
minimum bandwidth of X between points Y and Z"...)

Based upon that devices do not write intent, I think that "freeze network" is
not a well specified Intent. Instead I suggest that the Intent should read
something like:
    "The network will start growing at INITIAL-DATE, and
    will continue until either condition-X or FINAL-DATE"



--
Michael Richardson , Sandelman Software Works
 -= IPv6 IoT consulting =-




--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBVvyHN4CLcPvd0N1lAQJxBgf/e54/4/1/a5SbBmcNsFjeqHQJ59dV10dz
LhrdeD4UHfifODRGMeWDEd028lpL+jh/3YfD7VgP1KAJbL1hZm/45GM7sqy4Lo3y
rP6VzNG/K+V9opraEiCCpSIiwwRhUsejvDr+KTyWchbszdc4xvW+B4c2/uA2rrRn
70gMOAslAhbDF+1FNJE0FaEPTR2kOU82dMdlxYHULOgQTDFJYLJ4VG01O57ADmBB
pep5987/aLPjphYqI9PdD1jU3XlxHxccVwxABL66FEvTKMC+HXs7m3VifFPbgWPA
IOQNCFCs5Ce2p3sej2QSR+DXGSCxePOkQRzhd7UFAst7LbrkpeZiRg==
=zEb5
-----END PGP SIGNATURE-----
--=-=-=--


From nobody Wed Mar 30 19:35:36 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D68AF12D6A5 for ; Wed, 30 Mar 2016 19:35:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id n4WI-fpazG3Q for ; Wed, 30 Mar 2016 19:35:33 -0700 (PDT)
Received: from mail-pf0-x236.google.com (mail-pf0-x236.google.com [IPv6:2607:f8b0:400e:c00::236]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3120512D563 for ; Wed, 30 Mar 2016 19:35:33 -0700 (PDT)
Received: by mail-pf0-x236.google.com with SMTP id e128so36371206pfe.3 for ; Wed, 30 Mar 2016 19:35:33 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=dTKM105kIzrchbABjqDHnglwf7XrbuPUuOz8W1EfISc=; b=PsMH8nmDohotSH7jzikPg7WxxmAk+aCNTtRnnGPS856t+ophJelD9McTpseAQwLew/ An9xFwBz3nwrJ7C8ekOXZNHXaxjo431TL6NZ4Vm67koX+azSbYtvbjiuz6Ay73zibSJz J2DTwWKRHZsApsnDP3RI1hB2QPWt/S1HgSJLS3tabcQ5LW36+444RwOHI+wnuS4Ez/BF tKp5JXN5TyYhmOuYsN/naV6kvvIFAFYBJ3nrp9S1egv/B0qpplkXxOWl8yz8o2IK4t42 qbh2QK4h47o1s5WK1YRUaSjlecsoFKGsrzQfhFxsKaiwRvNbuHEXyBMGuqziENzeIjVr UqIw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=dTKM105kIzrchbABjqDHnglwf7XrbuPUuOz8W1EfISc=; b=k6bhH5WQnlmA6yN9jwOZQ0oWXXPSqrNE9X07GRdYXX3bKvkBCZzEVUBBcbqaMFhEG1 qOLvNIjY3h4WbbF7h4gjnWsqKLO6V0XkFOdmn6NUaObBaQxJRs7Ru2kWncsYnJ8pCk2T nj9c2LCJNGTmDK2dgRRZte4aPnEgz04EfaxkRIpufXGannxiP+RrNorP4mvIMxtd0wtT 6igC4Z56sAU3fa7Us3ppwUyrZXqtNXG51Jkx45JJKZ53uVjgiYxapGXo7s1zpdI61LCM r0uH3CdaVlbq8lXhpNHjkWLhOmfbDbnOphFVmN22R6n0AVFkzr9D2ciDTfvFpdD3OVcG FSbQ==
X-Gm-Message-State: AD7BkJIAgUTvgm3lgC6byXjAOR3S0+G9+Cua1YAreus8tGxyZWU+5Rs9Pj9xfeEc71sGsw==
X-Received: by 10.98.89.22 with SMTP id n22mr18301635pfb.55.1459391732703; Wed, 30 Mar 2016 19:35:32 -0700 (PDT)
Received: from ?IPv6:2406:e007:705e:1:28cc:dc4c:9703:6781? ([2406:e007:705e:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id r5sm8933016pap.7.2016.03.30.19.35.29 (version=TLSv1/SSLv3 cipher=OTHER); Wed, 30 Mar 2016 19:35:31 -0700 (PDT)
To: "Joel M. Halpern" , Michael Richardson , "anima@ietf.org" 
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FAA188.2050901@joelhalpern.com> <56FB0DC6.1030701@gmail.com> <16238.1459385357@obiwan.sandelman.ca> <56FC7C06.9090508@joelhalpern.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56FC8CF8.70400@gmail.com>
Date: Thu, 31 Mar 2016 15:35:36 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <56FC7C06.9090508@joelhalpern.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 02:35:36 -0000

On 31/03/2016 14:23, Joel M. Halpern wrote:
> If we just passed around URLs (or URIs, or  UR?), how would the ASAs know which ones they needed to fetch?

And I'm not sure why we'd want to add a protocol.

What could be done is to flood a new version announcement (with version number
plus checksum, for example) that would prompt recipients to do an explicit
unicast synchronization of the whole thing. But of course that loses the
benefit of flooding and caching prior to the ASA needing the Intent.

   Brian

> 
> Yours,
> Joel
> 
> On 3/30/16 8:49 PM, Michael Richardson wrote:
>>
>> Brian E Carpenter  wrote:
>>      >> This is also why I worry a bit about flooding to everything.
>>
>>      > There's a technical issue there, which is that GRASP flooding uses
>>      > multicast UDP (what else could it reasonably use, anyway?). So if
>>      > the chunk to be flooded gets too big, we would have to chop it up.
>>
>> This is why I really think that the URL for the Intent (maybe using the rfc6920
>> ni:// URLs) should be flooded, and let the nodes ask for the things they
>> need.
>>
>> -- 
>> Michael Richardson , Sandelman Software Works
>>   -= IPv6 IoT consulting =-
>>
>>
>>
>>
>>
>> _______________________________________________
>> Anima mailing list
>> Anima@ietf.org
>> https://www.ietf.org/mailman/listinfo/anima
>>
> 
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
> 


From nobody Thu Mar 31 01:27:48 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6412A12D0BA for ; Thu, 31 Mar 2016 01:27:47 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kS9qjdAI53sV for ; Thu, 31 Mar 2016 01:27:46 -0700 (PDT)
Received: from alln-iport-2.cisco.com (alln-iport-2.cisco.com [173.37.142.89]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CB76112D098 for ; Thu, 31 Mar 2016 01:27:45 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1496; q=dns/txt; s=iport; t=1459412865; x=1460622465; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=TTLtK3mbhPETIn1UqNPa61OCnT6MfNj3fkPsvCau5qA=; b=V61Esw5XhfowvKhqsrM8P62FAv186stu0L87d6GDa7co6KpvDF22GcrP 12Xr7HPAaGFREg8fPRwqoh82mDhUAPkZ9sF6R4244mvBvPQv+cBevx1+H /X4udxGAc1SOMCcfnUzJGcl5LKuCkt4Qi6GEgc6EcM5dPP6gsoPHgIXfb I=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0B5BQB33vxW/4gNJK1dgzOBUAa4foIdg?= =?us-ascii?q?XCGDQKBPDoSAQEBAQEBAWUnhEEBAQEDATo1DwcEAgEIEQQBAR8JBzIUCQgCBAE?= =?us-ascii?q?SCIgXCMFaAQEBAQEBAQEBAQEBAQEBAQEBAQEBFYYehEaKFAWXcgGOAIFth3WFM?= =?us-ascii?q?o8UAScML4NnbIdvfgEBAQ?=
X-IronPort-AV: E=Sophos;i="5.24,421,1454976000"; d="scan'208";a="254110693"
Received: from alln-core-3.cisco.com ([173.36.13.136]) by alln-iport-2.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 31 Mar 2016 08:27:44 +0000
Received: from XCH-ALN-006.cisco.com (xch-aln-006.cisco.com [173.36.7.16]) by alln-core-3.cisco.com (8.14.5/8.14.5) with ESMTP id u2V8Ri2M026346 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 31 Mar 2016 08:27:44 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-ALN-006.cisco.com (173.36.7.16) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 31 Mar 2016 03:27:43 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Thu, 31 Mar 2016 03:27:44 -0500
From: "Michael Behringer (mbehring)" 
To: Michael Richardson , anima 
Thread-Topic: [Anima] ANIMA intent discussion
Thread-Index: AQHRip6r6CBKKtp6FEycGBHZ8l3kKJ9zI74AgAAS2xA=
Date: Thu, 31 Mar 2016 08:27:43 +0000
Message-ID: <9e9c9bd91fe94fafb7ec3280a6b494b6@XCH-RCD-006.cisco.com>
References: <56FBFA6A.4010400@alcatel-lucent.com> <1493.1459390266@obiwan.sandelman.ca>
In-Reply-To: <1493.1459390266@obiwan.sandelman.ca>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.49.80.32]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Subject: Re: [Anima] ANIMA intent discussion
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 08:27:47 -0000

> -----Original Message-----
> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael
> Richardson
> Sent: 31 March 2016 04:11
> To: anima 
> Subject: Re: [Anima] ANIMA intent discussion
>=20
>=20
> These are good questions, thank you for assembling them.
>=20
> > 1) Who writes intents?
>=20
> Let me answer with a question: can marketing/sales-people write intents?
> (I suspect we all know the answer should be no, but probably we should be
> explici that Intents are technical statements.
> But consider if an Intent can be written like, "optimize network to provi=
de a
> minimum bandwidth of X between points Y and Z"...)

Strongly disagree :-)  : If we do a good job, then a business person CAN an=
d SHOULD issue Intent. You should NOT require a technical person to do this=
.=20

That of course means in the end we need a suitable user interface.
=20
And I see as well that on the way to this goal, before functions are really=
 self-managing, we still need to give more technical guidance. But I'd see =
that as an intermediate step.=20

> Based upon that devices do not write intent, I think that "freeze network=
" is
> not a well specified Intent. Instead I suggest that the Intent should rea=
d
> something like:
>     "The network will start growing at INITIAL-DATE, and
>     will continue until either condition-X or FINAL-DATE"

That is definitely an option. Depends how flexible we want to make it.=20

Michael


From nobody Thu Mar 31 01:36:59 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4B81F12D143 for ; Thu, 31 Mar 2016 01:36:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.921
X-Spam-Level: 
X-Spam-Status: No, score=-6.921 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iA2jDqt2gM1q for ; Thu, 31 Mar 2016 01:36:56 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3614312D164 for ; Thu, 31 Mar 2016 01:36:51 -0700 (PDT)
Received: from fr712umx4.dmz.alcatel-lucent.com (unknown [135.245.210.45]) by Websense Email Security Gateway with ESMTPS id DE0F8D4BB4BD4 for ; Thu, 31 Mar 2016 08:36:47 +0000 (GMT)
Received: from fr712usmtp2.zeu.alcatel-lucent.com (fr712usmtp2.zeu.alcatel-lucent.com [135.239.2.42]) by fr712umx4.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2V8an0M014404 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Thu, 31 Mar 2016 08:36:49 GMT
Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id u2V8aJSJ016091 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for ; Thu, 31 Mar 2016 10:36:48 +0200
Received: from [135.224.210.108] (135.239.27.41) by FR711WXCHHUB02.zeu.alcatel-lucent.com (135.239.2.112) with Microsoft SMTP Server (TLS) id 14.3.195.1; Thu, 31 Mar 2016 10:36:34 +0200
Message-ID: <56FCE18B.6030600@alcatel-lucent.com>
Date: Thu, 31 Mar 2016 10:36:27 +0200
From: Martin Vigoureux 
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.7.0
MIME-Version: 1.0
To: 
References: <56FBFA6A.4010400@alcatel-lucent.com> <1493.1459390266@obiwan.sandelman.ca> <9e9c9bd91fe94fafb7ec3280a6b494b6@XCH-RCD-006.cisco.com>
In-Reply-To: <9e9c9bd91fe94fafb7ec3280a6b494b6@XCH-RCD-006.cisco.com>
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 8bit
X-Originating-IP: [135.239.27.41]
Archived-At: 
Subject: Re: [Anima] ANIMA intent discussion
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 08:36:58 -0000

I think that the person should at least be knowledgeable enough to 
understand (the implications of) what he/she is asking for.

-m

Le 31/03/2016 10:27, EXT Michael Behringer (mbehring) a écrit :
>> -----Original Message-----
>> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael
>> Richardson
>> Sent: 31 March 2016 04:11
>> To: anima 
>> Subject: Re: [Anima] ANIMA intent discussion
>>
>>
>> These are good questions, thank you for assembling them.
>>
>>> 1) Who writes intents?
>>
>> Let me answer with a question: can marketing/sales-people write intents?
>> (I suspect we all know the answer should be no, but probably we should be
>> explici that Intents are technical statements.
>> But consider if an Intent can be written like, "optimize network to provide a
>> minimum bandwidth of X between points Y and Z"...)
>
> Strongly disagree :-)  : If we do a good job, then a business person CAN and SHOULD issue Intent. You should NOT require a technical person to do this.
>
> That of course means in the end we need a suitable user interface.
>
> And I see as well that on the way to this goal, before functions are really self-managing, we still need to give more technical guidance. But I'd see that as an intermediate step.
>
>> Based upon that devices do not write intent, I think that "freeze network" is
>> not a well specified Intent. Instead I suggest that the Intent should read
>> something like:
>>      "The network will start growing at INITIAL-DATE, and
>>      will continue until either condition-X or FINAL-DATE"
>
> That is definitely an option. Depends how flexible we want to make it.
>
> Michael
>
> _______________________________________________
> Anima mailing list
> Anima@ietf.org
> https://www.ietf.org/mailman/listinfo/anima
>
>


From nobody Thu Mar 31 01:40:39 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7B33B12D143 for ; Thu, 31 Mar 2016 01:40:37 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.531
X-Spam-Level: 
X-Spam-Status: No, score=-14.531 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DGIy2Gff8eu3 for ; Thu, 31 Mar 2016 01:40:36 -0700 (PDT)
Received: from rcdn-iport-8.cisco.com (rcdn-iport-8.cisco.com [173.37.86.79]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2D16B12D17E for ; Thu, 31 Mar 2016 01:40:34 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=429; q=dns/txt; s=iport; t=1459413634; x=1460623234; h=from:to:subject:date:message-id:references:in-reply-to: content-transfer-encoding:mime-version; bh=4NgNtXVkRjRM1+4jAOZq3fWF5x4vzvVbZaDmCegbUts=; b=Cz7vvXdZqy2PMzl04Tw9VTsvux4/CGX0yvjICfh5Lf2mwMyel+EgIHOL CTYaZsu9d3T62ADj+3mUZIZ6rr+z8se8jBd0n2x4KCYC+AziMafDC2thN ZTOhEjys9JuD+RQNqN/ufcTGWbzVOn3cS1/OEH3CJ4PdiHt6sNWRUOpcV A=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0AfAgDk4fxW/4sNJK1dgzOBUAa4foIPA?= =?us-ascii?q?Q2BcIYNAoE8OBQBAQEBAQEBZSeEQQEBAQQ6SwQCAQgRBAEBHwkHMhQJCAIEARI?= =?us-ascii?q?IiB/BWwEBAQEBAQEBAQEBAQEBAQEBAQEBARWGHoRGihQFl3IBjgCBVwGNPI8UA?= =?us-ascii?q?R4BAUKDZ2yHb34BAQE?=
X-IronPort-AV: E=Sophos;i="5.24,421,1454976000"; d="scan'208";a="86623047"
Received: from alln-core-6.cisco.com ([173.36.13.139]) by rcdn-iport-8.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 31 Mar 2016 08:40:33 +0000
Received: from XCH-RCD-009.cisco.com (xch-rcd-009.cisco.com [173.37.102.19]) by alln-core-6.cisco.com (8.14.5/8.14.5) with ESMTP id u2V8eX3Z026700 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 31 Mar 2016 08:40:33 GMT
Received: from xch-rcd-006.cisco.com (173.37.102.16) by XCH-RCD-009.cisco.com (173.37.102.19) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 31 Mar 2016 03:40:32 -0500
Received: from xch-rcd-006.cisco.com ([173.37.102.16]) by XCH-RCD-006.cisco.com ([173.37.102.16]) with mapi id 15.00.1104.009; Thu, 31 Mar 2016 03:40:32 -0500
From: "Michael Behringer (mbehring)" 
To: Martin Vigoureux , "anima@ietf.org" 
Thread-Topic: [Anima] ANIMA intent discussion
Thread-Index: AQHRip6r6CBKKtp6FEycGBHZ8l3kKJ9zI74AgAAS2xCAAFjQgP//rSFA
Date: Thu, 31 Mar 2016 08:40:32 +0000
Message-ID: <97667ecb08b348d1937f88d00e597e01@XCH-RCD-006.cisco.com>
References: <56FBFA6A.4010400@alcatel-lucent.com> <1493.1459390266@obiwan.sandelman.ca> <9e9c9bd91fe94fafb7ec3280a6b494b6@XCH-RCD-006.cisco.com> <56FCE18B.6030600@alcatel-lucent.com>
In-Reply-To: <56FCE18B.6030600@alcatel-lucent.com>
Accept-Language: en-GB, en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
x-originating-ip: [10.49.80.32]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: 
Subject: Re: [Anima] ANIMA intent discussion
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 08:40:37 -0000

> -----Original Message-----
> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Martin
> Vigoureux
> Sent: 31 March 2016 10:36
> To: anima@ietf.org
> Subject: Re: [Anima] ANIMA intent discussion
>=20
> I think that the person should at least be knowledgeable enough to
> understand (the implications of) what he/she is asking for.

Agree. But a good user interface can solve that / help with it.=20

Michael


From nobody Thu Mar 31 02:13:36 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2662212D143 for ; Thu, 31 Mar 2016 02:13:35 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id J3pUshUubguM for ; Thu, 31 Mar 2016 02:13:30 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3F7B512D5E7 for ; Thu, 31 Mar 2016 02:13:29 -0700 (PDT)
Received: from fr712umx3.dmz.alcatel-lucent.com (unknown [135.245.210.42]) by Websense Email Security Gateway with ESMTPS id 0BE72358B4F8B; Thu, 31 Mar 2016 09:13:23 +0000 (GMT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (fr711usmtp1.zeu.alcatel-lucent.com [135.239.2.122]) by fr712umx3.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2V9DOMN001313 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 31 Mar 2016 09:13:24 GMT
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id u2V9DNXx004096 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 31 Mar 2016 11:13:23 +0200
Received: from [135.224.207.56] (135.239.27.38) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Thu, 31 Mar 2016 11:13:23 +0200
To: "EXT Michael Behringer (mbehring)" , Martin Vigoureux , "anima@ietf.org" 
References: <56FBFA6A.4010400@alcatel-lucent.com> <1493.1459390266@obiwan.sandelman.ca> <9e9c9bd91fe94fafb7ec3280a6b494b6@XCH-RCD-006.cisco.com> <56FCE18B.6030600@alcatel-lucent.com> <97667ecb08b348d1937f88d00e597e01@XCH-RCD-006.cisco.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56FCEA32.3090101@alcatel-lucent.com>
Date: Thu, 31 Mar 2016 11:13:22 +0200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <97667ecb08b348d1937f88d00e597e01@XCH-RCD-006.cisco.com>
Content-Type: multipart/alternative; boundary="------------000506090300080607070909"
X-Originating-IP: [135.239.27.38]
Archived-At: 
Subject: Re: [Anima] ANIMA intent discussion
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 09:13:35 -0000

--------------000506090300080607070909
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit

On 31/03/2016 10:40, EXT Michael Behringer (mbehring) wrote:
>> -----Original Message-----
>> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Martin
>> Vigoureux
>> Sent: 31 March 2016 10:36
>> To: anima@ietf.org
>> Subject: Re: [Anima] ANIMA intent discussion
>>
>> I think that the person should at least be knowledgeable enough to
>> understand (the implications of) what he/she is asking for.
> Agree. But a good user interface can solve that / help with it.
>
> Michael

I agree with both of you ;)
Using some sort of catalog (suggested by John S. few days ago on the 
list) is a first step: capabilities of the "network" (ASAs, other 
things...) are expressed, aggregated (maybe annotated) in a common 
format understandable to the user (technical and/or non-technical). This 
sets a first bound on what the user can or cannot do.
Other aspects to consider: (properties of) language (or other means) 
offered to the user to write the intents, ontologies / common lexicon, 
safeguards on values/syntax checking, etc.
a full project on its own! let's go for a BoF ;)

Laurent.


--------------000506090300080607070909
Content-Type: text/html; charset="windows-1252"
Content-Transfer-Encoding: 7bit


  
    
  
  
    On 31/03/2016 10:40, EXT Michael Behringer (mbehring) wrote:

    

      

        
-----Original Message-----
From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Martin
Vigoureux
Sent: 31 March 2016 10:36
To: anima@ietf.org
Subject: Re: [Anima] ANIMA intent discussion

I think that the person should at least be knowledgeable enough to
understand (the implications of) what he/she is asking for.


      

      
Agree. But a good user interface can solve that / help with it. 

Michael


    

    

    I agree with both of you ;)

    Using some sort of catalog (suggested by John S. few days ago on the
    list) is a first step: capabilities of the "network" (ASAs, other
    things...) are expressed, aggregated (maybe annotated) in a common
    format understandable to the user (technical and/or non-technical).
    This sets a first bound on what the user can or cannot do.

    Other aspects to consider: (properties of) language (or other means)
    offered to the user to write the intents, ontologies / common
    lexicon, safeguards on values/syntax checking, etc.

    a full project on its own! let's go for a BoF ;)

    

    Laurent.

    

  


--------------000506090300080607070909--


From nobody Thu Mar 31 02:22:52 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E74EC12D0FC for ; Thu, 31 Mar 2016 02:22:49 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.92
X-Spam-Level: 
X-Spam-Status: No, score=-6.92 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id l6wMSbpGZcoI for ; Thu, 31 Mar 2016 02:22:47 -0700 (PDT)
Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 875DB12D096 for ; Thu, 31 Mar 2016 02:22:47 -0700 (PDT)
Received: from fr712umx3.dmz.alcatel-lucent.com (unknown [135.245.210.42]) by Websense Email Security Gateway with ESMTPS id 313082D23C439; Thu, 31 Mar 2016 09:22:44 +0000 (GMT)
Received: from fr711usmtp1.zeu.alcatel-lucent.com (fr711usmtp1.zeu.alcatel-lucent.com [135.239.2.122]) by fr712umx3.dmz.alcatel-lucent.com (GMO-o) with ESMTP id u2V9Mj89014182 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 31 Mar 2016 09:22:45 GMT
Received: from FR711WXCHHUB01.zeu.alcatel-lucent.com (fr711wxchhub01.zeu.alcatel-lucent.com [135.239.2.111]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id u2V9MhAp013653 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Thu, 31 Mar 2016 11:22:44 +0200
Received: from [135.224.207.56] (135.239.27.38) by FR711WXCHHUB01.zeu.alcatel-lucent.com (135.239.2.111) with Microsoft SMTP Server (TLS) id 14.3.195.1; Thu, 31 Mar 2016 11:22:42 +0200
To: "EXT Michael Behringer (mbehring)" , Michael Richardson , anima 
References: <56FBFA6A.4010400@alcatel-lucent.com> <1493.1459390266@obiwan.sandelman.ca> <9e9c9bd91fe94fafb7ec3280a6b494b6@XCH-RCD-006.cisco.com>
From: Laurent Ciavaglia 
Organization: Bell Labs
Message-ID: <56FCEC61.8020302@alcatel-lucent.com>
Date: Thu, 31 Mar 2016 11:22:41 +0200
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0
MIME-Version: 1.0
In-Reply-To: <9e9c9bd91fe94fafb7ec3280a6b494b6@XCH-RCD-006.cisco.com>
Content-Type: multipart/alternative; boundary="------------040308030108090503030906"
X-Originating-IP: [135.239.27.38]
Archived-At: 
Subject: Re: [Anima] ANIMA intent discussion
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 09:22:50 -0000

--------------040308030108090503030906
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit

Please see inline with [lc][/lc]
On 31/03/2016 10:27, EXT Michael Behringer (mbehring) wrote:
>> -----Original Message-----
>> From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael
>> Richardson
>> Sent: 31 March 2016 04:11
>> To: anima 
>> Subject: Re: [Anima] ANIMA intent discussion
>>
>>
>> These are good questions, thank you for assembling them.
>>
>>> 1) Who writes intents?
>> Let me answer with a question: can marketing/sales-people write intents?
>> (I suspect we all know the answer should be no, but probably we should be
>> explici that Intents are technical statements.
>> But consider if an Intent can be written like, "optimize network to provide a
>> minimum bandwidth of X between points Y and Z"...)
> Strongly disagree :-)  : If we do a good job, then a business person CAN and SHOULD issue Intent. You should NOT require a technical person to do this.

[lc]: I think it should be possible for a business person to write 
intent. however, (maybe) this is out of scope of ANIMA, and even maybe 
of IETF (?).
Something we could try to achieve in ANIMA is to define an intent system 
that NOC engineers could use, and define interfaces or recommendations 
for upper layers (business).
[/lc]

>
> That of course means in the end we need a suitable user interface.
>   
> And I see as well that on the way to this goal, before functions are really self-managing, we still need to give more technical guidance. But I'd see that as an intermediate step.
>
>> Based upon that devices do not write intent, I think that "freeze network" is
>> not a well specified Intent. Instead I suggest that the Intent should read
>> something like:
>>      "The network will start growing at INITIAL-DATE, and
>>      will continue until either condition-X or FINAL-DATE"
> That is definitely an option. Depends how flexible we want to make it.
>
> Michael

--------------040308030108090503030906
Content-Type: text/html; charset="windows-1252"
Content-Transfer-Encoding: 7bit


  
    
  
  
    Please see inline with [lc][/lc]

    
On 31/03/2016 10:27, EXT Michael
      Behringer (mbehring) wrote:

    

    

      

        
-----Original Message-----
From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Michael
Richardson
Sent: 31 March 2016 04:11
To: anima <anima@ietf.org>
Subject: Re: [Anima] ANIMA intent discussion


These are good questions, thank you for assembling them.



        

          
1) Who writes intents?


        

        
Let me answer with a question: can marketing/sales-people write intents?
(I suspect we all know the answer should be no, but probably we should be
explici that Intents are technical statements.
But consider if an Intent can be written like, "optimize network to provide a
minimum bandwidth of X between points Y and Z"...)


      

      
Strongly disagree :-)  : If we do a good job, then a business person CAN and SHOULD issue Intent. You should NOT require a technical person to do this. 

    

    

    [lc]: I think it should be possible for a business person to write
    intent. however, (maybe) this is out of scope of ANIMA, and even
    maybe of IETF (?).

    Something we could try to achieve in ANIMA is to define an intent
    system that NOC engineers could use, and define interfaces or
    recommendations for upper layers (business).

    [/lc]

    

    

      

That of course means in the end we need a suitable user interface.
 
And I see as well that on the way to this goal, before functions are really self-managing, we still need to give more technical guidance. But I'd see that as an intermediate step. 



      

        
Based upon that devices do not write intent, I think that "freeze network" is
not a well specified Intent. Instead I suggest that the Intent should read
something like:
    "The network will start growing at INITIAL-DATE, and
    will continue until either condition-X or FINAL-DATE"


      

      
That is definitely an option. Depends how flexible we want to make it. 

Michael


    

  


--------------040308030108090503030906--


From nobody Thu Mar 31 06:22:01 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A818212D562 for ; Thu, 31 Mar 2016 06:21:58 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.911
X-Spam-Level: 
X-Spam-Status: No, score=-1.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EeILO0E4O3lv for ; Thu, 31 Mar 2016 06:21:56 -0700 (PDT)
Received: from tuna.sandelman.ca (tuna.sandelman.ca [209.87.249.19]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3F8DF12D13E for ; Thu, 31 Mar 2016 06:21:55 -0700 (PDT)
Received: from sandelman.ca (obiwan.sandelman.ca [IPv6:2607:f0b0:f:2::247]) by tuna.sandelman.ca (Postfix) with ESMTP id DA833203B8; Thu, 31 Mar 2016 09:25:05 -0400 (EDT)
Received: from obiwan.sandelman.ca (localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id BF85E6375A; Thu, 31 Mar 2016 09:21:54 -0400 (EDT)
From: Michael Richardson 
To: Brian E Carpenter 
In-Reply-To: <56FB2315.3030104@gmail.com>
References: <56FB2315.3030104@gmail.com>
X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2
X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m
Archived-At: 
Cc: Anima WG 
Subject: Re: [Anima] Draft text on infrastructure ASAs
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 13:21:59 -0000

--=-=-=
Content-Type: text/plain


Brian E Carpenter  wrote:
    > In various threads people have been asking about how components
    > of the ANI (autonomic networking infrastructure) could themselves
    > make use of GRASP. This has come up for the secure bootstrap,
    > formation of the ACP, and Intent distribution.

Thanks. My notion is that isn't not very difficult at all, it just needs to
be written down.

    > It's in the WG wiki:
    > https://trac.tools.ietf.org/wg/anima/trac/wiki/Infrastructure%20ASAs
    > so please have a look and more important, post your comments here.

As you say: "There is no magic in this"

Would it be worth having an ACP cold-start bootstrap document?

   > b) the "cold start" case in which the ACP is not yet operational, so
   > neither the proxy nor the registrar have ACP addresses. We can force the
   > pledge to wait in that situation, simply by ensuring that no proxy can
   > be discovered. But even so, each potential proxy needs to communicate
   > with the registrar, although there is no predefined secure channel.

It seems that the registrar, needs to run a proxy, and it needs to start by
giving itself an ACP address on a loopback interface that at the beginning,
goes nowhere.  The shell expands concentrically.

     "(even in insecure mode)"

I don't agree with your comment the proxies will ask the registrar where it
is using insecure GRASP: they aren't proxies yet, they are pledges.
Once the ACP is up, the proxies will ask where the registrar is using secure
GRASP over the ACP.

I agree with everything else.

--
Michael Richardson , Sandelman Software Works
 -= IPv6 IoT consulting =-




--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBVv0kb4CLcPvd0N1lAQIbwgf9HuR1H8vATQuBHCjSnjMzGV3dktQK3cg3
hI2RM1gPVh9dXWgVrVkRSvqg7Kk6iuTyWKreCdMPbODqMeNJ4u9Vg6Q7peHu0gsm
tr1NHMP8EcRnLkKH6bZZ16XaukQLuekgfKFQ5fT6eEupE7xu8gtmAF9XXkuXhBiT
uK6BAhUxZG/U+W/Cp67PmxLLXSMZJhChoBqYBP/2Sxlwf+ghFU9z2Q9nQxKYGtoV
6zsUCuYAZPNW+cWSkoHQLcOpFxufIUSrKpnAvYR6MELhXXVP/o0HU7lQH/1hMNDj
pdS1qDDquxYitKblv9gx5GAFsjKGLD4+YbKobv9RbfTqC39hW4a4lQ==
=ESCG
-----END PGP SIGNATURE-----
--=-=-=--


From nobody Thu Mar 31 06:37:15 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6FD5A12D606 for ; Thu, 31 Mar 2016 06:37:14 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.911
X-Spam-Level: 
X-Spam-Status: No, score=-1.911 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3pPACLwU1Gwa for ; Thu, 31 Mar 2016 06:37:13 -0700 (PDT)
Received: from tuna.sandelman.ca (tuna.sandelman.ca [IPv6:2607:f0b0:f:3:216:3eff:fe7c:d1f3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E3EB712D197 for ; Thu, 31 Mar 2016 06:37:12 -0700 (PDT)
Received: from sandelman.ca (obiwan.sandelman.ca [IPv6:2607:f0b0:f:2::247]) by tuna.sandelman.ca (Postfix) with ESMTP id 4481E204E3; Thu, 31 Mar 2016 09:40:23 -0400 (EDT)
Received: from obiwan.sandelman.ca (localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id 199686375A; Thu, 31 Mar 2016 09:37:12 -0400 (EDT)
From: Michael Richardson 
To: "Joel M. Halpern" 
In-Reply-To: <56FC7C06.9090508@joelhalpern.com>
References: <62b0d024950e4f73922b9fc6e63c05cf@XCH-RCD-006.cisco.com> <56F440D1.3060702@gmail.com>  <56F5DFD6.90000@gmail.com>  <56FAA188.2050901@joelhalpern.com> <56FB0DC6.1030701@gmail.com> <16238.1459385357@obiwan.sandelman.ca> <56FC7C06.9090508@joelhalpern.com>
X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2
X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m
Archived-At: 
Cc: "anima@ietf.org" 
Subject: Re: [Anima] Another use case for Intent: "Freeze network enrolment"
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 13:37:14 -0000

--=-=-=
Content-Type: text/plain


Joel M. Halpern  wrote:
    > If we just passed around URLs (or URIs, or  UR?), how would the ASAs know
    > which ones they needed to fetch?

They wouldn't know, so they'd have to retrieve all of the ones they didn't
have so that they would have a "complete" set of Intents.

But, we'd be flooding the URNs (~60 bytes), rather than the potentially
multiple packet-sized-each contents.  It's just flood by reference rather
than value.

And ASAs wouldn't need to get Intents they already had.

[If I had to design this from scratch, and we didn't have trickle flooding in
GRASP already, I'd use ICP (RFC2186)]

--
Michael Richardson , Sandelman Software Works
 -= IPv6 IoT consulting =-




--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQEVAwUBVv0oBYCLcPvd0N1lAQJ4AggAukgPVfH/8cj7UGt+Vjr6Tap0PV960Ts+
NDVerZttysRgglYvoODLo1KitW7oHbZ699d/ksgkq9HybhAIYQT+6VVO6rXpYMn/
vgsEIz2C83WipLoCG9yvjrYmmwD6EwjUAVvFg1jFIf961L/GMQUPCXZOVAkyT3Sq
Q0Mez/ZCoPJMWKFDR5EwlrsTklP3fX9rqS2tlrAhv+93fbrp+aeOaq7V2z8xD2ld
sdi0jZ0js7yHELd57KYMa4TSnObNeB5C/wdBZrJn4uFBPbGF/A2JfKgNPEaPNUmB
2rVfay0C0E5JW/U9HA79hqAAp2x2xnVuvbtsLwHptUOHaiGp+TeLww==
=XXrU
-----END PGP SIGNATURE-----
--=-=-=--


From nobody Thu Mar 31 12:15:11 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BD26812D786 for ; Thu, 31 Mar 2016 12:15:09 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id g1e1LSeWhncJ for ; Thu, 31 Mar 2016 12:15:07 -0700 (PDT)
Received: from mail-pf0-x233.google.com (mail-pf0-x233.google.com [IPv6:2607:f8b0:400e:c00::233]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D395A12D781 for ; Thu, 31 Mar 2016 12:15:07 -0700 (PDT)
Received: by mail-pf0-x233.google.com with SMTP id 4so75947360pfd.0 for ; Thu, 31 Mar 2016 12:15:07 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:cc:from:organization:message-id:date :user-agent:mime-version:in-reply-to:content-transfer-encoding; bh=pKge9wSfdpSipQuK37cDekrGM1rayPdUd4RCms9i62s=; b=gm/oBZlW0VOD0s7y5MrI9N5cZwd6cFQ0vU2byM6zADHZt/kaV6AJeEqI4WIRLumQuq s6sjzOKn9rrTY0KuZa6nOi9y+VhUu/NTXzuesmz5fa/MqlE0rGZkKMUEyubA3zdPnS0u OHmZwtJv92OeY4VMseC6lmQXrFYWwawbOHXDvhBymqvXyQ/VYWvIotmF/2pwj7tYknbG jLkn/+twSCnjqu0xFL5qNt9EZRAHbav/KMU7eo4UIbb9uMFIBZEssNI96SbFRIOSnDGD 5RdR3vVNb2ZY0zSgK9ynzjIE7j9XDSNIoG3I1hZywQ2LndYb/ir4/79Gy3HyTqawkmyW arBQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:cc:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=pKge9wSfdpSipQuK37cDekrGM1rayPdUd4RCms9i62s=; b=K6hQzuOIggKBeBoO02WJ6ZV4uRM9mvzJcxbrvPzzoVTLaFIkRV+NsUKzPizCmaPehC mWE+pidS6/yd+We4cVeoXFFUeczY2pZKfEMcNrGJ2TMiluPn+gmzpaqEKN/qoy7RrU7R uVx6C7pd2pDvYH6bXrCLpvFMSlSi/XIbHBhQEY5mSbYnJoLXwF26dwTrLAxJ02VS2dd/ 4hu4wAoCVmQaQq4r/M5d/VuVz1zp/jeHyLBsvVjI+P0Xkz7YYr55fG6kWres//bjqqNu cJ4NZEexYUBPFHWKTm8R6gmW1C8g9kw4922A2yFdHcjfUwrY4Hz6pko2sGV5hiQHfEWy frWg==
X-Gm-Message-State: AD7BkJKeGnps07l7c8ifDdvV737G+kUFLdHsf+19KJOdjyhf4Mitpm+IhuSbvY6B996Xcg==
X-Received: by 10.98.8.219 with SMTP id 88mr24506802pfi.51.1459451707463; Thu, 31 Mar 2016 12:15:07 -0700 (PDT)
Received: from ?IPv6:2406:e007:5e51:1:28cc:dc4c:9703:6781? ([2406:e007:5e51:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id tp9sm15236185pab.14.2016.03.31.12.15.05 (version=TLSv1/SSLv3 cipher=OTHER); Thu, 31 Mar 2016 12:15:06 -0700 (PDT)
To: Michael Richardson 
References: <56FB2315.3030104@gmail.com> <18029.1459430514@obiwan.sandelman.ca>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56FD773F.90909@gmail.com>
Date: Fri, 1 Apr 2016 08:15:11 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <18029.1459430514@obiwan.sandelman.ca>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Cc: Anima WG 
Subject: Re: [Anima] Draft text on infrastructure ASAs
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 19:15:09 -0000

On 01/04/2016 02:21, Michael Richardson wrote:
> 
> Brian E Carpenter  wrote:
>     > In various threads people have been asking about how components
>     > of the ANI (autonomic networking infrastructure) could themselves
>     > make use of GRASP. This has come up for the secure bootstrap,
>     > formation of the ACP, and Intent distribution.
> 
> Thanks. My notion is that isn't not very difficult at all, it just needs to
> be written down.
> 
>     > It's in the WG wiki:
>     > https://trac.tools.ietf.org/wg/anima/trac/wiki/Infrastructure%20ASAs
>     > so please have a look and more important, post your comments here.
> 
> As you say: "There is no magic in this"
> 
> Would it be worth having an ACP cold-start bootstrap document?

I think so, but I also think that Michael B is already on it.

>    > b) the "cold start" case in which the ACP is not yet operational, so
>    > neither the proxy nor the registrar have ACP addresses. We can force the
>    > pledge to wait in that situation, simply by ensuring that no proxy can
>    > be discovered. But even so, each potential proxy needs to communicate
>    > with the registrar, although there is no predefined secure channel.
> 
> It seems that the registrar, needs to run a proxy, and it needs to start by
> giving itself an ACP address on a loopback interface that at the beginning,
> goes nowhere.  The shell expands concentrically.

Yes, good thinking.

> 
>      "(even in insecure mode)"
> 
> I don't agree with your comment the proxies will ask the registrar where it
> is using insecure GRASP: they aren't proxies yet, they are pledges.
> Once the ACP is up, the proxies will ask where the registrar is using secure
> GRASP over the ACP.

You're quite right. A proxy has to start life as a pledge. And that's another
reason that there will be an expanding shell out from the registrar.

> I agree with everything else.

Excellent!

    Brian


From nobody Thu Mar 31 12:29:12 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7566512D776 for ; Thu, 31 Mar 2016 12:29:10 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -14.53
X-Spam-Level: 
X-Spam-Status: No, score=-14.53 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=cisco.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id R6dmpQ361F5i for ; Thu, 31 Mar 2016 12:29:08 -0700 (PDT)
Received: from rcdn-iport-1.cisco.com (rcdn-iport-1.cisco.com [173.37.86.72]) (using TLSv1.2 with cipher DHE-RSA-SEED-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 7749F12D60D for ; Thu, 31 Mar 2016 12:29:08 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=6694; q=dns/txt; s=iport; t=1459452548; x=1460662148; h=from:to:cc:subject:date:message-id:references: in-reply-to:reply-to:mime-version; bh=ALqe0p3l2+NtYgxbMjjlbyQIppHD30XSnExgqZpZIUA=; b=WXXdUIN9Rj/8/LkiADQDz9fxJtL6o234B50uqbqoNm2cDuK3CeVSH6it 2QNH0Yr4nACMa6+o90sjXh6gHs8oHZ/jEjSCCBs0FdlNjkaurt6sNwa25 /vWBShgjlT/fgdKNmETYha4vBIVExCy1i7f+dt/QZPqrGdh8ZPgbBdSLI U=;
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0ACAgAlef1W/4QNJK1dgmhMU32vQ4tSA?= =?us-ascii?q?Q2BcRcBCYVsAoFDOBQBAQEBAQEBZSeEQQEBAQMBAQEBKkELBQsCAQgYCSUPEgY?= =?us-ascii?q?LDhcCBAENBRmHeQMKCA64NoVaDYRtAQEBAQEBAQEBAQEBAQEBAQEBAQEBEQSKZ?= =?us-ascii?q?IJAgiuFKQWGKgmMXoQwMQGFcoYggXWBZoRNiFqHQYdTAR4BAUKDZ2wBiGwBAQE?=
X-IronPort-AV: E=Sophos; i="5.24,423,1454976000"; d="scan'208,217"; a="92373341"
Received: from alln-core-10.cisco.com ([173.36.13.132]) by rcdn-iport-1.cisco.com with ESMTP/TLS/DHE-RSA-AES256-SHA; 31 Mar 2016 19:29:07 +0000
Received: from XCH-ALN-003.cisco.com (xch-aln-003.cisco.com [173.36.7.13]) by alln-core-10.cisco.com (8.14.5/8.14.5) with ESMTP id u2VJT7H9019861 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=FAIL); Thu, 31 Mar 2016 19:29:07 GMT
Received: from xch-rcd-003.cisco.com (173.37.102.13) by XCH-ALN-003.cisco.com (173.36.7.13) with Microsoft SMTP Server (TLS) id 15.0.1104.5; Thu, 31 Mar 2016 14:29:06 -0500
Received: from xch-rcd-003.cisco.com ([173.37.102.13]) by XCH-RCD-003.cisco.com ([173.37.102.13]) with mapi id 15.00.1104.009; Thu, 31 Mar 2016 14:29:06 -0500
From: "Toerless Eckert (eckert)" 
To: Brian E Carpenter , Michael Richardson 
Thread-Topic: [Anima] Draft text on infrastructure ASAs
Thread-Index: AQHRih5UeMGiQd8rOkazM0xHmrimrp9z4CoAgABitYD//7AT1Q==
Date: Thu, 31 Mar 2016 19:29:06 +0000
Message-ID: 
References: <56FB2315.3030104@gmail.com> <18029.1459430514@obiwan.sandelman.ca>,<56FD773F.90909@gmail.com>
In-Reply-To: <56FD773F.90909@gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-ms-exchange-transport-fromentityheader: Hosted
Content-Type: multipart/alternative; boundary="_000_e8u6s2wxnfdis5cbcqqwqsuu1459452415772emailandroidcom_"
MIME-Version: 1.0
Archived-At: 
Cc: Anima WG 
Subject: Re: [Anima] Draft text on infrastructure ASAs
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
Reply-To: "Toerless Eckert \(eckert\)" 
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Thu, 31 Mar 2016 19:29:10 -0000

--_000_e8u6s2wxnfdis5cbcqqwqsuu1459452415772emailandroidcom_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

The sections i filled into the latest acp draft are trying to cover how acp=
 an bootstrap use grasp. Will also make some picture for monday to vusualiz=
e ;-)

Toerless


Sent from my Samsung Captivate Glide on AT&T

Brian E Carpenter  wrote:
On 01/04/2016 02:21, Michael Richardson wrote:
>
> Brian E Carpenter  wrote:
>     > In various threads people have been asking about how components
>     > of the ANI (autonomic networking infrastructure) could themselves
>     > make use of GRASP. This has come up for the secure bootstrap,
>     > formation of the ACP, and Intent distribution.
>
> Thanks. My notion is that isn't not very difficult at all, it just needs =
to
> be written down.
>
>     > It's in the WG wiki:
>     > https://trac.tools.ietf.org/wg/anima/trac/wiki/Infrastructure%20ASA=
s
>     > so please have a look and more important, post your comments here.
>
> As you say: "There is no magic in this"
>
> Would it be worth having an ACP cold-start bootstrap document?

I think so, but I also think that Michael B is already on it.

>    > b) the "cold start" case in which the ACP is not yet operational, so
>    > neither the proxy nor the registrar have ACP addresses. We can force=
 the
>    > pledge to wait in that situation, simply by ensuring that no proxy c=
an
>    > be discovered. But even so, each potential proxy needs to communicat=
e
>    > with the registrar, although there is no predefined secure channel.
>
> It seems that the registrar, needs to run a proxy, and it needs to start =
by
> giving itself an ACP address on a loopback interface that at the beginnin=
g,
> goes nowhere.  The shell expands concentrically.

Yes, good thinking.

>
>      "(even in insecure mode)"
>
> I don't agree with your comment the proxies will ask the registrar where =
it
> is using insecure GRASP: they aren't proxies yet, they are pledges.
> Once the ACP is up, the proxies will ask where the registrar is using sec=
ure
> GRASP over the ACP.

You're quite right. A proxy has to start life as a pledge. And that's anoth=
er
reason that there will be an expanding shell out from the registrar.

> I agree with everything else.

Excellent!

    Brian

_______________________________________________
Anima mailing list
Anima@ietf.org
https://www.ietf.org/mailman/listinfo/anima

--_000_e8u6s2wxnfdis5cbcqqwqsuu1459452415772emailandroidcom_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable











The sections i filled into the latest acp draft are trying to cover ho=
w acp an bootstrap use grasp. Will also make some picture for monday to vus=
ualize ;-)







Toerless














Sent from my Samsung Captivate =
Glide on AT&T





Brian E Carpenter <brian.e.carpenter@gmail.com> wrote:





On 01/04/2016 02:21, Michael Richardson wrote:

> 

> Brian E Carpenter <brian.e.carpenter@gmail.com> wrote:

>     > In various threads people have been askin=
g about how components

>     > of the ANI (autonomic networking infrastr=
ucture) could themselves

>     > make use of GRASP. This has come up for t=
he secure bootstrap,

>     > formation of the ACP, and Intent distribu=
tion.

> 

> Thanks. My notion is that isn't not very difficult at all, it just nee=
ds to

> be written down.

> 

>     > It's in the WG wiki:

>     > 
https://trac.tools.ietf.org/wg/anima/trac/wiki/Infrastructure%20ASAs
>     > so please have a look and more important,=
 post your comments here.

> 

> As you say: "There is no magic in this"

> 

> Would it be worth having an ACP cold-start bootstrap document?



I think so, but I also think that Michael B is already on it.



>    > b) the "cold start" case in which the=
 ACP is not yet operational, so

>    > neither the proxy nor the registrar have ACP ad=
dresses. We can force the

>    > pledge to wait in that situation, simply by ens=
uring that no proxy can

>    > be discovered. But even so, each potential prox=
y needs to communicate

>    > with the registrar, although there is no predef=
ined secure channel.

> 

> It seems that the registrar, needs to run a proxy, and it needs to sta=
rt by

> giving itself an ACP address on a loopback interface that at the begin=
ning,

> goes nowhere.  The shell expands concentrically.



Yes, good thinking.



> 

>      "(even in insecure mode)"

> 

> I don't agree with your comment the proxies will ask the registrar whe=
re it

> is using insecure GRASP: they aren't proxies yet, they are pledges.
> Once the ACP is up, the proxies will ask where the registrar is using =
secure

> GRASP over the ACP.



You're quite right. A proxy has to start life as a pledge. And that's anoth=
er

reason that there will be an expanding shell out from the registrar.



> I agree with everything else.



Excellent!



    Brian



_______________________________________________

Anima mailing list

Anima@ietf.org

https://www.ietf.or=
g/mailman/listinfo/anima







--_000_e8u6s2wxnfdis5cbcqqwqsuu1459452415772emailandroidcom_--


From nobody Thu Mar 31 18:03:00 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9F4CA12D505 for ; Thu, 31 Mar 2016 18:02:59 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.7
X-Spam-Level: 
X-Spam-Status: No, score=-2.7 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_LOW=-0.7, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7T5-2WIkJwqa for ; Thu, 31 Mar 2016 18:02:57 -0700 (PDT)
Received: from mail-pf0-x229.google.com (mail-pf0-x229.google.com [IPv6:2607:f8b0:400e:c00::229]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C7DF112D0AF for ; Thu, 31 Mar 2016 18:02:57 -0700 (PDT)
Received: by mail-pf0-x229.google.com with SMTP id e128so60309546pfe.3 for ; Thu, 31 Mar 2016 18:02:57 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113;  h=subject:to:references:from:organization:message-id:date:user-agent :mime-version:in-reply-to:content-transfer-encoding; bh=mG8s2g+V1+7Vlvqwa3PHWoPgOxrRbpa7xQFvoKD9woY=; b=WlNHF/KlR7je2rVI2l+h1WMncAxIu4xB3aXz/lo+Z8hNqWq68w2fjByrlnBSSgLb2S pu4cN9fpIKx8BIQW3QPwLGtkAiaJ5ljX3WyLeqhZGUfjMQk1v2ZlWuG+kjME8+o8O66f YgiDw/Ep769OU7EQUlKgpdVdJxI4YMynVClSWRXOxfJgCeG13ssqBIzIIFYSnao6tf0L k7nKOOV2LK9C4Wwci0rgIgMT9/MWL+MXgr2IBa4wIGyhOq7VMc+USmiasinaMKyalmxG IOT17XIoleMzL0rk4qHooBYE5/9H2oAB11xg+yhR/4d1+eJMPukZeoyxvjENPpL9Sl9M LEtw==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:subject:to:references:from:organization :message-id:date:user-agent:mime-version:in-reply-to :content-transfer-encoding; bh=mG8s2g+V1+7Vlvqwa3PHWoPgOxrRbpa7xQFvoKD9woY=; b=hDnBymLwHdU7zJZPdApnhzmbhfJaSu2dn36+Nu3MxlJsUoydLpV0Y+XB2VWo3hspqG T8KMABUmIlAvK97DU0s/eYPgmb/7iCnWh3k3d44RxbokYOG0tDqHDR7CSqpOBd0nEEq8 KMe9zwicE3z3pafYv3ieOpC0hXTcxadqqD5nogI4XCPevQ5CF0CGS2vqeASGlG+ACGFJ GZ/dtGB0bVNHv7h3nrB3ijGXybdbK9wW7KnUsHS5lkjloT5RUmsRaDmYHb9d7C7iWQeK WG8XDfAZfp51ULqm5vNmawEQQoJIqUtlZGc1DpRw0gIi26uhBlYzYQI1y0ErbLK0nEUz 3Jjg==
X-Gm-Message-State: AD7BkJLUnydr5qSbLpiy7kiZjdK7YoAR5hF/lrge7uKSLmwTe/RATJ/ZLcshJKpjexTU9g==
X-Received: by 10.98.32.23 with SMTP id g23mr26520498pfg.74.1459472577218; Thu, 31 Mar 2016 18:02:57 -0700 (PDT)
Received: from ?IPv6:2406:e007:5e51:1:28cc:dc4c:9703:6781? ([2406:e007:5e51:1:28cc:dc4c:9703:6781]) by smtp.gmail.com with ESMTPSA id yl1sm16057951pac.35.2016.03.31.18.02.54 for  (version=TLSv1/SSLv3 cipher=OTHER); Thu, 31 Mar 2016 18:02:55 -0700 (PDT)
To: Anima WG 
References: <20160321234459.10893.4388.idtracker@ietfa.amsl.com>
From: Brian E Carpenter 
Organization: University of Auckland
Message-ID: <56FDC8C6.4070406@gmail.com>
Date: Fri, 1 Apr 2016 14:03:02 +1300
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
MIME-Version: 1.0
In-Reply-To: <20160321234459.10893.4388.idtracker@ietfa.amsl.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Archived-At: 
Subject: Re: [Anima] I-D Action: draft-ietf-anima-autonomic-control-plane-02.txt
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 01 Apr 2016 01:02:59 -0000

Hi,

A few points I noticed:

> 3.  Requirements
...
>    The default mode of operation of the ACP is hop-by-hop, because this
>    interaction can be built on IPv6 link local addressing, which is
>    autonomic, and has no dependency on configuration (requirement 1).

This statement assumes that the link interfaces concerned have stable interface
identifiers that do not need to be configured. Somewhere it needs to be
specified how these identifiers are set, for example by citing RFC7217.

> 5.3.  Capability Negotiation
> 
>    Autonomic devices may have different capabilities based on the type
>    of device, OS version, etc.  To establish a trusted secure ACP
>    channel, devices must first negotiate their mutual capabilities in
>    the data plane.  This allows for the support of different channel
>    types in the future.

Unless I have misunderstood, we need to say "in the link-local data plane".
That is all we can assume at power-on.

> 5.4.  Channel Establishment
...
>    The channel establishment MUST be authenticated.  Whether or not, and
>    how, a channel is encrypted is part of the capability negotiation,
>    potentially controlled by Intent.
...
>    Since channels are by default established between adjacent neighbors,
>    the resulting overlay network does hop by hop encryption.  Each node
>    decrypts incoming traffic from the ACP, and encrypts outgoing traffic
>    to its neighbors in the ACP.  Routing is discussed in Section 5.7.

The first quoted paragraph says that encryption is optional, but the second
one just says that there is encryption. Which is it? Perhaps we should say
that encryption MUST be enabled by default, unless Intent says otherwise.

> 5.6.2.  The Base Addressing Scheme
...
>      8      40          3                     77
>    +--+--------------+------+------------------------------------------+
>    |FD| hash(domain) | Type |             (sub-scheme)                 |
>    +--+--------------+------+------------------------------------------+
...
>    o  "FD" identifies a locally defined ULA address.

According to RFC 5952, IPv6 addresses should use lower case, so s/FD/fd/.

> 5.6.3.1.  Sub-Scheme 1
...
>    o  Device ID: A unique value for each device.
> 
>    The device ID is derived as follows:

There is some commentary needed here. Firstly, this field in IPv6 is
called the Interface ID. It's OK to consider it as a device ID, but what
it really is is the Interface ID used for addresses assigned to the
ACP instance of loopback interface.

Secondly, the IPv6 addressing architecture requires it to be 64 bits;
that isn't a choice. (Long story, if you want the details see RFC 7136,
RFC 7421, RFC 7608, and draft-ietf-6man-rfc4291bis.)

> 5.6.4.  Usage of the Zone Field
...
>    If aggregation is required, the 13 bit value allows for up to 8191
>    zones.
...
>    ...In this case the ACP loopback is
>    configured with two ACP addresses: One for zone 0 and one for the
>    assigned zone.  This method allows for a smooth transition between a
>    flat addressing scheme and an hierarchical one.

I suggest citing RFC 7608 which makes it clear that the different "zone IDs"
are really just subnets within the overlay network, and routed as such. The
only trickery here is multihoming the node into two networks, which isn't
trickery at all. However, it is necessary to specify the address selection
method to be applied (vanilla RFC 6724 or something else?)

>  [We could divide the zone space
>    into manual and automatic space - to be discussed.]

I hate to see 'manual' in any Anima document, but especially here. Don't
we want the ACP to re-form itself after an earthquake, without any need for
humans?

> 7.4.  GRASP instance details

One implementation comment: I don't think we will really run multiple GRASP
instances (which would mean duplicating some data structures in a very
tricky way). I think we will actually tag each on-going Session ID within
GRASP to say whether it's running in (e.g.) insecure mode, secure-with-TLS mode,
or secure-via-ACP mode. The external effect will be the same, and per-session
granularity seems correct. We probably need to add some optional parameters to
the GRASP API to capture this.

...
>    o  GRASP packets received on an ACP (virtual) interfaces are assigned
>       to the ACP instance of GRASP
> 
>    o  GRASP/UDP packets received on L2 interfaces where the device is
>       willing to run ACP across are are assigned to a separate instance
>       of GRASP for that L2 interface.  We call those instances of GRASP
>       the "insecure L2 GRASP instances" and the ASA to perform the
>       discovery the "insecure L2 discovery ASA" (IL2D).

Here we hit a problem I've been concerned about for a while. Unless you
explicitly add support for link-local multicast to the ACP, GRASP/UDP
is insecure, because the only use of UDP is for LL multicast. But this
draft doesn't talk anywhere about how the ACP could support LL multicast.
Therefore, all GRASP Discovery and Flood Synchronization messages are
currently insecure. (Discovery Response, Synchronization and negotiation
are unicast and can be secured.)

Confirming some comments I made to Toerless a week or two ago:

>    For every L2-discovery instance of GRASP and its IL2D, the following
>    rules apply, amending and overriding the recommendations in
>    [I-D.ietf-anima-grasp]:
>
>    o  GRASP link-local multicast discovery messages MUST use GTSM
>       [RFC5082].  With GTSM, discovery packets are sent with a TTL of
>       255 and packets received with a TTL smaller than 255 are ignored
>       upon receipt.

I see no problem in adding that on any stack where the socket API permits.
Actually we could operate Discovery and Flood Synchronization in that
way all the time, but with link-local addresses we are probably safe
anyway.

>    o  The GRASP loop count of GRASP discovery packets must be set to 1
>       on sending.

That limits the scope of discovery to the LAN.

>    o  GRASP MUST send response messages for the discovery objected
>       defined here (overriding the MAY).

That's fine; it's only MAY in GRASP in case a node wants to hide.

>    o  GRASP MUST NOT respond to discovery objectives with the Divert
>       option - objectives learned and cached are solely for local
>       consumption.

That's not a problem. (Also, as insurance, the receiver can be coded to
ignore any discovered locators that were Diverts.)

>    o  GRASP MUST NOT relay discovery or any other messages across
>       different interfaces.

That is guaranteed by the loop count = 1.

>    TBD: The Details of the GRASP objective/packet formats still need to
>    be defined.  Eg: Need to define an allocation for the objective of
>    "Autonomic Node".

Sure, see my message yesterday. But let's get the conceptual level right first.

Regards
    Brian


From nobody Thu Mar 31 19:39:15 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D020912D517 for ; Thu, 31 Mar 2016 19:39:12 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.22
X-Spam-Level: 
X-Spam-Status: No, score=-4.22 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_KAM_HTML_FONT_INVALID=0.01, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4VvYGncshllz for ; Thu, 31 Mar 2016 19:39:10 -0700 (PDT)
Received: from dfwrgout.huawei.com (dfwrgout.huawei.com [206.16.17.72]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0BBE112D50E for ; Thu, 31 Mar 2016 19:39:10 -0700 (PDT)
Received: from 172.18.9.243 (EHLO lhreml707-cah.china.huawei.com) ([172.18.9.243]) by dfwrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BON59588; Thu, 31 Mar 2016 21:39:08 -0500 (CDT)
Received: from NKGEML414-HUB.china.huawei.com (10.98.56.75) by lhreml707-cah.china.huawei.com (10.201.5.199) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 1 Apr 2016 03:39:06 +0100
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by nkgeml414-hub.china.huawei.com ([10.98.56.75]) with mapi id 14.03.0235.001; Fri, 1 Apr 2016 10:39:02 +0800
From: Duzongpeng 
To: Laurent Ciavaglia , "EXT Michael Behringer (mbehring)" , Martin Vigoureux 
Thread-Topic: [Anima] ANIMA intent discussion
Thread-Index: AQHRip6yZEilVhzOO0eSNltNsN15iJ9ySdAAgABpOoCAAAJxgIAAASQAgAAJLACAAaMSoA==
Date: Fri, 1 Apr 2016 02:39:02 +0000
Message-ID: 
References: <56FBFA6A.4010400@alcatel-lucent.com> <1493.1459390266@obiwan.sandelman.ca> <9e9c9bd91fe94fafb7ec3280a6b494b6@XCH-RCD-006.cisco.com> <56FCE18B.6030600@alcatel-lucent.com> <97667ecb08b348d1937f88d00e597e01@XCH-RCD-006.cisco.com> <56FCEA32.3090101@alcatel-lucent.com>
In-Reply-To: <56FCEA32.3090101@alcatel-lucent.com>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.149.226]
Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFE18Fnkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A0B0205.56FDDF4D.001D, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: bbc47d1d5374185ebddb7519a6a14342
Archived-At: 
Cc: "anima@ietf.org" , John Strassner 
Subject: Re: [Anima] ANIMA intent discussion
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 01 Apr 2016 02:39:13 -0000

--_000_BAFEC9523F57BC48A51C20226A5589575FDFE18Fnkgeml514mbxchi_
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

Hi Laurent, Michael, Martin

         I agree that different kinds of users can issue intent, but their =
intents can be different, and with different purposes.

The network operator will be able to issue intent very detailed, and perhap=
s non-technical people will have more general intents.

Non-technical people will mainly care about how to use the network, for the=
ir specific services.

The network operator will care about the network infrastructure, to make su=
re the network can provide proper services to uplayer users.

I think we can firstly working on some Administrative intent just like we h=
ave firstly worked on some infrastructure functions in ANIMA.


I support that it needs to be done about way to writing intent, especially =
for the uplayer users, and it is very important.

John perhaps can help because he has already done some work on this.

In conclusion, it turns to be very flexible in the definition of intent (to=
o many ways exist), and needs a huge amount of work.

My suggestion is that we need to figure out some steps on this, not to star=
t too many things in parallel in ANIMA.


Best Regards
Zongpeng Du

From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Laurent Ciavaglia
Sent: Thursday, March 31, 2016 5:13 PM
To: EXT Michael Behringer (mbehring); Martin Vigoureux; anima@ietf.org
Subject: Re: [Anima] ANIMA intent discussion

On 31/03/2016 10:40, EXT Michael Behringer (mbehring) wrote:


-----Original Message-----

From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Martin

Vigoureux

Sent: 31 March 2016 10:36

To: anima@ietf.org

Subject: Re: [Anima] ANIMA intent discussion



I think that the person should at least be knowledgeable enough to

understand (the implications of) what he/she is asking for.



Agree. But a good user interface can solve that / help with it.



Michael

I agree with both of you ;)
Using some sort of catalog (suggested by John S. few days ago on the list) =
is a first step: capabilities of the "network" (ASAs, other things...) are =
expressed, aggregated (maybe annotated) in a common format understandable t=
o the user (technical and/or non-technical). This sets a first bound on wha=
t the user can or cannot do.
Other aspects to consider: (properties of) language (or other means) offere=
d to the user to write the intents, ontologies / common lexicon, safeguards=
 on values/syntax checking, etc.
a full project on its own! let's go for a BoF ;)

Laurent.

--_000_BAFEC9523F57BC48A51C20226A5589575FDFE18Fnkgeml514mbxchi_
Content-Type: text/html; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable











Hi Laurent=
, Michael, Martin


 =
;


 &nbs=
p;       I agree that different kinds of user=
s can issue intent, but their intents can be different, and with different =
purposes.



 =
;


The network operator will be able to issue intent very d=
etailed, and perhaps non-technical people will have more general
 intents.


 


Non-technical people will mainly care about how to use t=
he network, for their specific services.


 


The network operator will care about the network infrast=
ructure, to make sure the network can provide proper services
 to uplayer users.


 


I think we can firstly working on some Administrative in=
tent just like we have firstly worked on some infrastructure
 functions in ANIMA.


 =
;


 &nbs=
p;      



I support that it needs to be done about way to writing =
intent, especially for the uplayer users, and it is very important.


 =
;


John perhaps can help because he has already done some w=
ork on this.


 


In conclusion, it turns to be very flexible in the defin=
ition of intent (too many ways exist), and needs a huge amount
 of work. 


 


My suggestion is that we need to figure out some steps o=
n this, not to start too many things in parallel in ANIMA.


 


 


Best Regar=
ds


Zongpeng D=
u


 =
;






From:=
 Anima [mailto:anima-bounces@ietf.org]
On Behalf Of Laurent Ciavaglia

Sent: Thursday, March 31, 2016 5:13 PM

To: EXT Michael Behringer (mbehring); Martin Vigoureux; 
anima@ietf.org

Subject: Re: [Anima] ANIMA intent discussion






 


On 31/03/2016 10:40, EXT Michae=
l Behringer (mbehring) wrote:








-----Original Message-----

From: Anima [mailto:anima-bounces@ietf.org] On Behalf Of Martin


Vigoureux


Sent: 31 March 2016 10:36


To: anima@ietf.=
org


Subject: Re: [Anima] ANIMA intent discussion=



 


I think that the person should at least be knowle=
dgeable enough to


understand (the implications of) what he/she is a=
sking for.




 


Agree. But a good user interface can solve that /=
 help with it. 


 


Michael


=


I agree with both of you ;)

Using some sort of catalog (suggested by John S. few days ago on the list) =
is a first step: capabilities of the "network" (ASAs, other thing=
s...) are expressed, aggregated (maybe annotated) in a common format unders=
tandable to the user (technical and/or non-technical).
 This sets a first bound on what the user can or cannot do.

Other aspects to consider: (properties of) language (or other means) offere=
d to the user to write the intents, ontologies / common lexicon, safeguards=
 on values/syntax checking, etc.

a full project on its own! let's go for a BoF ;)



Laurent.






--_000_BAFEC9523F57BC48A51C20226A5589575FDFE18Fnkgeml514mbxchi_--


From nobody Thu Mar 31 20:56:30 2016
Return-Path: 
X-Original-To: anima@ietfa.amsl.com
Delivered-To: anima@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BA3A312D1AC for ; Thu, 31 Mar 2016 20:56:28 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -4.22
X-Spam-Level: 
X-Spam-Status: No, score=-4.22 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RCVD_IN_MSPIKE_H3=-0.01, RCVD_IN_MSPIKE_WL=-0.01, SPF_PASS=-0.001, T_KAM_HTML_FONT_INVALID=0.01, T_RP_MATCHES_RCVD=-0.01] autolearn=ham autolearn_force=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jekPYOAOz6qT for ; Thu, 31 Mar 2016 20:56:26 -0700 (PDT)
Received: from dfwrgout.huawei.com (dfwrgout.huawei.com [206.16.17.72]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 335E012D1B8 for ; Thu, 31 Mar 2016 20:56:26 -0700 (PDT)
Received: from 172.18.9.243 (EHLO lhreml702-cah.china.huawei.com) ([172.18.9.243]) by dfwrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id CXP68744; Thu, 31 Mar 2016 22:56:25 -0500 (CDT)
Received: from NKGEML413-HUB.china.huawei.com (10.98.56.74) by lhreml702-cah.china.huawei.com (10.201.5.99) with Microsoft SMTP Server (TLS) id 14.3.235.1; Fri, 1 Apr 2016 04:56:12 +0100
Received: from NKGEML514-MBX.china.huawei.com ([fe80::40a8:f0d:c0f3:2ca5]) by NKGEML413-HUB.china.huawei.com ([10.98.56.74]) with mapi id 14.03.0235.001; Fri, 1 Apr 2016 11:56:08 +0800
From: Duzongpeng 
To: Laurent Ciavaglia , anima 
Thread-Topic: ANIMA intent discussion
Thread-Index: AQHRip6yZEilVhzOO0eSNltNsN15iJ90cLjg
Date: Fri, 1 Apr 2016 03:56:08 +0000
Message-ID: 
References: <56FBFA6A.4010400@alcatel-lucent.com>
In-Reply-To: <56FBFA6A.4010400@alcatel-lucent.com>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.111.149.226]
Content-Type: multipart/alternative; boundary="_000_BAFEC9523F57BC48A51C20226A5589575FDFE1D4nkgeml514mbxchi_"
MIME-Version: 1.0
X-CFilter-Loop: Reflected
X-Mirapoint-Virus-RAPID-Raw: score=unknown(0), refid=str=0001.0A0B0205.56FDF169.00D6, ss=1, re=0.000, recu=0.000, reip=0.000,  cl=1, cld=1, fgs=0, ip=0.0.0.0, so=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32
X-Mirapoint-Loop-Id: 327667af41f9739d1a785783f2735e5f
Archived-At: 
Cc: Michael Behringer , Pierre Peloso 
Subject: Re: [Anima] ANIMA intent discussion
X-BeenThere: anima@ietf.org
X-Mailman-Version: 2.1.17
Precedence: list
List-Id: Autonomic Networking Integrated Model and Approach 
List-Unsubscribe: , 
List-Archive: 
List-Post: 
List-Help: 
List-Subscribe: , 
X-List-Received-Date: Fri, 01 Apr 2016 03:56:29 -0000

--_000_BAFEC9523F57BC48A51C20226A5589575FDFE1D4nkgeml514mbxchi_
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFE1D4nkgeml514mbxchi_
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: base64
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--_000_BAFEC9523F57BC48A51C20226A5589575FDFE1D4nkgeml514mbxchi_--