From owner-ipfc@standards.gadzoox.com Thu Feb 1 22:24:13 2001 Received: from standards.gadzoox.com (standards.gadzoox.com [216.52.31.24]) by ietf.org (8.9.1a/8.9.1a) with SMTP id WAA18593 for ; Thu, 1 Feb 2001 22:24:12 -0500 (EST) Received: (from majordom@localhost) by standards.gadzoox.com (8.8.7/8.8.7) id RAA14387 for ipfc-list; Thu, 1 Feb 2001 17:36:50 -0800 X-Authentication-Warning: standards.gadzoox.com: majordom set sender to owner-ipfc@standards.gadzoox.com using -f Received: from lightsand.com ([208.50.99.84]) by standards.gadzoox.com (8.8.7/8.8.7) with SMTP id RAA14384 for ; Thu, 1 Feb 2001 17:36:25 -0800 Received: from muralipc (dhcp094.lightsand.com [192.168.1.94]) by lightsand.com (8.9.3+Sun/8.9.1) with SMTP id SAA26103; Thu, 1 Feb 2001 18:43:04 -0800 (PST) From: "Murali Rajagopal" To: "Mark A. Carlson" Cc: , , , , , "Erik Nordmark" Subject: RE: draft-ietf-ipfc-mib-framework-03.txt (Fwd) Date: Thu, 1 Feb 2001 18:44:26 -0800 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) In-Reply-To: Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Sender: owner-ipfc@standards.gadzoox.com Precedence: bulk Content-Transfer-Encoding: 7bit Mark: Please provide your plan of action regarding this. -Murali -----Original Message----- From: owner-ipfc@standards.gadzoox.com [mailto:owner-ipfc@standards.gadzoox.com]On Behalf Of Erik Nordmark Sent: Tuesday, December 12, 2000 4:00 PM To: muralir@lightsand.com Cc: Mark A. Carlson; gavin@gadzoox.com; lhu3@yahoo.com; nordmark@jurassic.Eng.Sun.COM; narten@raleigh.ibm.com; ipfc@standards.gadzoox.com Subject: RE: draft-ietf-ipfc-mib-framework-03.txt (Fwd) The framework document was discussed in the IESG. The overarching comment was that for the document to serve the WG it should be more exact and specific when it comes to the set of requirements that the WG are imposing on the fibre channel mibs. See the detailed comments. Erik - Section 3. text like: The Fibre Channel MIBs may consist of an arbitrary collection of groups,and may contain objects of type: READ-ONLY, READ-WRITE, or WRITE-ONLY. The WRITE-ONLY was a SMIv1 concept which no longer exists. The following principles should be observed in developing MIBs. This then goes on to list a set of motherhood and apple pie items that are common to all MIBs... so not ipfc specific. Then follow a few ipfc specific items. I can live with it, but the motherhood and apple pie might be omited. The one statement for that is that they should be SMIv2 compliant. - section 4. 4.1 General These MIBs should provide for proxy SNMP management. In other words, at a relatively high level of the MIB, table(s) should exist that allowmultiple entities to be referenced. The indexing of the table shouldallow multiple entities to be referenced through a single SNMP agent,at a single IP address.In general, an agent will proxy for a group of devices from a singlevendor. The method which an NMS uses to determine a set of addressableagents is beyond the scope of this document. This was needed for SNMPv1 and SNMPv2c. With SNMPv3 and the new SNMP Architecture (RFC2571), this is best handled by using contextNames. If you do that, then there are no more special considerations for proxy w.r.t. the MIB. That then all becomes an SNMP agent configuration issue, for which we have the SNMP PROXY MIB (RFC2573). That then at the same time specifies a std for an NMS to learn about the various contexts (logical MIBs if you like). Another possibility is that maybe they want to use the entity MIB (RFC2737) for this purpose too. Did they evaluate that one? I guess what I would like to see here is that maybe they evaluate some of these alternatives and then justyify why they choose a particular approach. In addition, RFC 2837 doesn't have any proxy indexing (which is good) so the framework document doesn't seem to capture what is currently done in the ipfc MIBs. - Section 4.2 Security I don't think that the MIB needs to have anything specific to allow for only read access. That is an implementation issue. For an SNMP agent with only SNMPv1, it would be something like specifying only community names with READ access. In any event, it is not something that needs/should be specified in a MIB itself (or maybe they mean that just english text in the security section of such a mib should explain it... that is fine... but not clear from what they write here. Also... I am not sure that we would encourage read-only access via SNMP and write access via HTTP??? If you do that, I would rather encourage to use SNMPv3. Section 5.1 Not sure I am happy with the terms "Proxy Management Domain" and "Proxy Master". We do not have such terms in SNMP, and also our use of the term "proxy" is not inline with what they describe (See RFC2571 and RFC2573). Their wording seems to be more the description of a master agent and subagent, somthing that we do know about as in RFC2741 Other nits: - the reference [1] is not referenced anywhere and seems not needed to me. - reference [2] should be to RFC2578 instead of 1902 which they already have as [7] - but worse, references [2] through [7] are not referenced anywhere. So they just seem gratuitous references. - I see no security section for this document itself. Sect 4.2 talks about what nbeeds to be done in MIBs in the ipfc environment I believe. - no standard copy-right section. --- From owner-ipfc@standards.gadzoox.com Fri Feb 2 09:25:28 2001 Received: from standards.gadzoox.com (standards.gadzoox.com [216.52.31.24]) by ietf.org (8.9.1a/8.9.1a) with SMTP id JAA09972 for ; Fri, 2 Feb 2001 09:25:27 -0500 (EST) Received: (from majordom@localhost) by standards.gadzoox.com (8.8.7/8.8.7) id EAA14840 for ipfc-list; Fri, 2 Feb 2001 04:47:29 -0800 X-Authentication-Warning: standards.gadzoox.com: majordom set sender to owner-ipfc@standards.gadzoox.com using -f Received: from s3.smtp.oleane.net (s3.smtp.oleane.net [195.25.12.10]) by standards.gadzoox.com (8.8.7/8.8.7) with SMTP id EAA14837 for ; Fri, 2 Feb 2001 04:47:23 -0800 Received: from smtp2.cluster.oleane.net (smtp2.cluster.oleane.net [195.25.12.17]) by s3.smtp.oleane.net (Postfix) with ESMTP id D013221239 for ; Fri, 2 Feb 2001 14:56:56 +0100 (CET) Received: from oleane (dyn-1-2-204.Vin.dialup.oleane.fr [194.2.4.204]) by smtp2.cluster.oleane.net with SMTP id f12DgtO69255 for ; Fri, 2 Feb 2001 14:42:55 +0100 (CET) Message-ID: <004901c08d1e$29c3b180$8001a8c0@oleane.com> From: "Peter Lewis" To: Subject: Beyond Sonet/SDH Conference Date: Fri, 2 Feb 2001 14:43:44 +0100 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0046_01C08D26.8B2C8C00" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2314.1300 X-MIMEOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Sender: owner-ipfc@standards.gadzoox.com Precedence: bulk This is a multi-part message in MIME format. ------=_NextPart_000_0046_01C08D26.8B2C8C00 Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable Everything on Gigabit Ethernet and 10 Gigabit Ethernet promises in the = MAN & WAN. Digital Wrapper, Resilient Packet Ring, Spatial Reuse Protocol, PONs, = ASON technical descriptions : Beyond Sonet/SDH Conference, Paris, 10-13 April. Close to the Gigabit Ethernet Exhibition , 11-12 April =20 http://www.upperside.fr/babeyondsdh.htm ------=_NextPart_000_0046_01C08D26.8B2C8C00 Content-Type: text/html; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable
Everything on Gigabit Ethernet and = 10 Gigabit=20 Ethernet promises in the MAN & WAN.
Digital = Wrapper, Resilient=20 Packet Ring, Spatial Reuse Protocol, PONs, ASON technical descriptions=20 :
Beyond Sonet/SDH = Conference,=20 Paris, 10-13 April.
Close to the Gigabit = Ethernet Exhibition=20 , 11-12 April
 
http://www.upperside.fr/= babeyondsdh.htm
------=_NextPart_000_0046_01C08D26.8B2C8C00-- From owner-ipfc@standards.gadzoox.com Mon Feb 12 16:40:19 2001 Received: from standards.gadzoox.com (standards.gadzoox.com [216.52.31.24]) by ietf.org (8.9.1a/8.9.1a) with SMTP id QAA22121 for ; Mon, 12 Feb 2001 16:40:18 -0500 (EST) Received: (from majordom@localhost) by standards.gadzoox.com (8.8.7/8.8.7) id LAA20543 for ipfc-list; Mon, 12 Feb 2001 11:49:45 -0800 X-Authentication-Warning: standards.gadzoox.com: majordom set sender to owner-ipfc@standards.gadzoox.com using -f Received: from lightsand.com ([208.50.99.84]) by standards.gadzoox.com (8.8.7/8.8.7) with SMTP id LAA20540 for ; Mon, 12 Feb 2001 11:49:40 -0800 Received: from muralipc ([192.168.2.52]) by lightsand.com (8.9.3+Sun/8.9.1) with SMTP id MAA23272 for ; Mon, 12 Feb 2001 12:58:46 -0800 (PST) From: "Murali Rajagopal" To: Subject: FW: IP over Fibre throughput numbers Date: Mon, 12 Feb 2001 13:00:12 -0800 Message-ID: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0002_01C094F3.BC285EE0" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Sender: owner-ipfc@standards.gadzoox.com Precedence: bulk This is a multi-part message in MIME format. ------=_NextPart_000_0002_01C094F3.BC285EE0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Please respond to the IPFC reflector if you have any comments. -----Original Message----- From: Dennis Solari [mailto:dys70@amdahl.com] Sent: Wednesday, January 17, 2001 10:37 AM To: muralir@lightsand.com Subject: Fw: IP over Fibre throughput numbers Just curious if you received this email from me? Dennis (Amdahl Corp) 678-319-4226 ----- Original Message ----- From: Dennis Solari To: muralir@lightsand.com Sent: Tuesday, January 16, 2001 9:41 PM Subject: IP over Fibre throughput numbers Murali, I am currently testing the speed I can get with IP over fibre channel using: * two Sun Ultra 5's (360Mhz) * both are have Emulex LP8000 SCSI/IP combo fibre channel PCI cards * connected through a Brocade silkworm 2400 switch. I can only drive IP over fibre to as high as ~200 Mbs. using iperf (variation of ttcp utility) What are some numbers you have seen doing bandwith testing while tweaking the Unix TCP/IP stack? Do you know were I can get other performance numbers of people doing the same type performance tests? Thanks in Advance, Dennis ------=_NextPart_000_0002_01C094F3.BC285EE0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Please=20 respond to the IPFC reflector if you have any = comments.
 
-----Original Message-----
From: Dennis Solari=20 [mailto:dys70@amdahl.com]
Sent: Wednesday, January 17, 2001 = 10:37=20 AM
To: muralir@lightsand.com
Subject: Fw: IP over = Fibre=20 throughput numbers

Just curious if you received this email from = me?
 
Dennis (Amdahl Corp)
678-319-4226
 
----- Original Message -----=20
From: Dennis = Solari
Sent: Tuesday, January 16, 2001 9:41 PM
Subject: IP over Fibre throughput numbers

Murali,
I am currently testing the speed I can get with IP = over fibre=20 channel using:
* two Sun Ultra 5's (360Mhz)
* both are have Emulex LP8000 SCSI/IP = combo fibre=20 channel PCI cards
* connected through a Brocade silkworm 2400=20 switch.
 
I can only drive IP over fibre to as high = as ~200=20 Mbs. using iperf (variation of ttcp utility)
 
What are some numbers you have seen doing bandwith = testing=20 while tweaking the Unix TCP/IP stack?
Do you know were I can get other performance numbers = of people=20 doing the same type performance tests?
 
Thanks in Advance,
Dennis 
------=_NextPart_000_0002_01C094F3.BC285EE0-- From owner-ipfc@standards.gadzoox.com Mon Feb 12 16:55:23 2001 Received: from standards.gadzoox.com (standards.gadzoox.com [216.52.31.24]) by ietf.org (8.9.1a/8.9.1a) with SMTP id QAA22417 for ; Mon, 12 Feb 2001 16:55:22 -0500 (EST) Received: (from majordom@localhost) by standards.gadzoox.com (8.8.7/8.8.7) id MAA20561 for ipfc-list; Mon, 12 Feb 2001 12:20:57 -0800 X-Authentication-Warning: standards.gadzoox.com: majordom set sender to owner-ipfc@standards.gadzoox.com using -f Received: from lightsand.com ([208.50.99.84]) by standards.gadzoox.com (8.8.7/8.8.7) with SMTP id MAA20558 for ; Mon, 12 Feb 2001 12:20:52 -0800 Received: from muralipc ([192.168.2.52]) by lightsand.com (8.9.3+Sun/8.9.1) with SMTP id NAA24706; Mon, 12 Feb 2001 13:29:21 -0800 (PST) From: "Murali Rajagopal" To: "Infante, Jon" Cc: "Weber, William L" , "IPFC Mailing List \(E-mail\)" Subject: RE: Question about WW_NN and RFC 2625 Date: Mon, 12 Feb 2001 13:30:47 -0800 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 In-Reply-To: <3A5E4F86.21BE36B2@emulex.com> Sender: owner-ipfc@standards.gadzoox.com Precedence: bulk Content-Transfer-Encoding: 7bit Jon: I have to agree that the document does not spell it out. I also agree that there no reason not to allow other NAA types for WW_NN. But when you consider FARP, there is an implication that the WW_PN and WW_NN use the same NAA Type. (Nobody said anything about a different NAA type during FARP discussions!) Hope this clarifes what I said. -Murali -----Original Message----- From: owner-ipfc@standards.gadzoox.com [mailto:owner-ipfc@standards.gadzoox.com]On Behalf Of Infante, Jon Sent: Thursday, January 11, 2001 4:28 PM To: Murali Rajagopal Cc: Weber, William L; IPFC Mailing List (E-mail) Subject: Re: Question about WW_NN and RFC 2625 Murali, I disagree. The Fibre Channel Network header ONLY contains the WWPN. That is the main reason the NAA for the WWPN must be 1 (mapping MAC addresses to WWPNs). Correct me if I'm wrong. however, I don't see any technical reason for the WWNN NAA to be 1 nor did I infer that was the intent of the many discussions the group had on this topic??? Jon Murali Rajagopal wrote: > The intent was NAA=1 for both. > If it did not get reflected, then it was an oversight. > > -Murali > > -----Original Message----- > From: owner-ipfc@standards.gadzoox.com > [mailto:owner-ipfc@standards.gadzoox.com]On Behalf Of Infante, Jon > Sent: Thursday, January 11, 2001 8:13 AM > To: Weber, William L > Cc: IPFC Mailing List (E-mail) > Subject: Re: Question about WW_NN and RFC 2625 > > Bill, > > My understanding is that only the World Wide Port Name requires NAA = 1. > It is not a requirement for the World Wide Node Name to have NAA = 1; > > Jon Infante > Emulex > > "Weber, William L" wrote: > > > I have a question concerning RFC 2625, IP and ARP over Fibre Channel, and > > Network_Address_Authority (NAA) of World Wide Names (WWN). I thought this > > would be the proper forum to present the question. > > The question is: > > Can the Node WWN (WW_NN) use NAA=5 for IEEE registered format and still > > be > > compliant/compatible with RFC 2625? > > > > I understand that the RFC 2625 requires the Port WWN (WW_PN) to use NAA = > 1 > > for IEEE 48-bit addresses. RFC 2625 does not explicitly state the NAA for > > the WW_NN, however the examples in Appendix E (i.e. E.4.2) show the WW_NN > > with NAA=1. > > > > Thanks, > > // Bill Weber Phone: +1-610-648-7022 > > // Unisys Corporation Fax: +1-610-648-4010 > > // 2476 Swedesford Rd. mailto:William.Weber@unisys.com > > // Malvern, PA 19355-9700 From owner-ipfc@standards.gadzoox.com Mon Feb 12 18:27:00 2001 Received: from standards.gadzoox.com (standards.gadzoox.com [216.52.31.24]) by ietf.org (8.9.1a/8.9.1a) with SMTP id SAA23744 for ; Mon, 12 Feb 2001 18:26:58 -0500 (EST) Received: (from majordom@localhost) by standards.gadzoox.com (8.8.7/8.8.7) id NAA20707 for ipfc-list; Mon, 12 Feb 2001 13:48:03 -0800 X-Authentication-Warning: standards.gadzoox.com: majordom set sender to owner-ipfc@standards.gadzoox.com using -f Received: from emulex.emulex.com (emulex.emulex.com [138.239.112.1]) by standards.gadzoox.com (8.8.7/8.8.7) with SMTP id NAA20704 for ; Mon, 12 Feb 2001 13:47:58 -0800 Received: from emulex.com (jinfantepc.emulex.com [138.239.96.81]) by emulex.emulex.com (8.9.1a/8.9.1) with ESMTP id OAA02650; Mon, 12 Feb 2001 14:52:29 -0800 (PST) Message-ID: <3A886A0D.3890B5DC@emulex.com> Date: Mon, 12 Feb 2001 14:56:13 -0800 From: "Infante, Jon" Organization: Emulex Corporation X-Mailer: Mozilla 4.74 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Murali Rajagopal CC: "Weber, William L" , "IPFC Mailing List (E-mail)" Subject: Re: Question about WW_NN and RFC 2625 References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit Sender: owner-ipfc@standards.gadzoox.com Precedence: bulk Content-Transfer-Encoding: 7bit Murali, I agree that the document does not spell it out and that there no reason not to allow other NAA types for WWNN. With regards to FARP, I thought special consideration was given to make FARP a generic ELS command and not specifically tie it to our IP Profile. Thus the WWPN and WWNN in the FARP payload specifies the full 8 byte WWN and not just the last 6 bytes as in a IEEE address. Considering you can specify the full 8 byte WWNN, the full 8 byte WWPN and the Match Address Code Points specify match on just the WWNN, just the WWPN, or both, I don't feel there is any implication that the WWPN and WWNN use the same NAA Type. In fact, I don't think the FARP chapter mentions NAA at all. Jon Murali Rajagopal wrote: > Jon: > > I have to agree that the document does not spell it out. > > I also agree that there no reason not to allow other NAA types for WW_NN. > > But when you consider FARP, there is an implication that the WW_PN and WW_NN > use the same NAA Type. (Nobody said anything about a different NAA type > during FARP discussions!) > > Hope this clarifes what I said. > > -Murali > -----Original Message----- > From: owner-ipfc@standards.gadzoox.com > [mailto:owner-ipfc@standards.gadzoox.com]On Behalf Of Infante, Jon > Sent: Thursday, January 11, 2001 4:28 PM > To: Murali Rajagopal > Cc: Weber, William L; IPFC Mailing List (E-mail) > Subject: Re: Question about WW_NN and RFC 2625 > > Murali, > > I disagree. The Fibre Channel Network header ONLY contains the > WWPN. That is the main reason the NAA for the WWPN must > be 1 (mapping MAC addresses to WWPNs). > > Correct me if I'm wrong. however, I don't see any technical reason > for the WWNN NAA to be 1 nor did I infer that was the intent of > the many discussions the group had on this topic??? > > Jon > > Murali Rajagopal wrote: > > > The intent was NAA=1 for both. > > If it did not get reflected, then it was an oversight. > > > > -Murali > > > > -----Original Message----- > > From: owner-ipfc@standards.gadzoox.com > > [mailto:owner-ipfc@standards.gadzoox.com]On Behalf Of Infante, Jon > > Sent: Thursday, January 11, 2001 8:13 AM > > To: Weber, William L > > Cc: IPFC Mailing List (E-mail) > > Subject: Re: Question about WW_NN and RFC 2625 > > > > Bill, > > > > My understanding is that only the World Wide Port Name requires NAA = 1. > > It is not a requirement for the World Wide Node Name to have NAA = 1; > > > > Jon Infante > > Emulex > > > > "Weber, William L" wrote: > > > > > I have a question concerning RFC 2625, IP and ARP over Fibre Channel, > and > > > Network_Address_Authority (NAA) of World Wide Names (WWN). I thought > this > > > would be the proper forum to present the question. > > > The question is: > > > Can the Node WWN (WW_NN) use NAA=5 for IEEE registered format and > still > > > be > > > compliant/compatible with RFC 2625? > > > > > > I understand that the RFC 2625 requires the Port WWN (WW_PN) to use NAA > = > > 1 > > > for IEEE 48-bit addresses. RFC 2625 does not explicitly state the NAA > for > > > the WW_NN, however the examples in Appendix E (i.e. E.4.2) show the > WW_NN > > > with NAA=1. > > > > > > Thanks, > > > // Bill Weber Phone: +1-610-648-7022 > > > // Unisys Corporation Fax: +1-610-648-4010 > > > // 2476 Swedesford Rd. mailto:William.Weber@unisys.com > > > // Malvern, PA 19355-9700 From owner-ipfc@standards.gadzoox.com Mon Feb 12 19:07:20 2001 Received: from standards.gadzoox.com (standards.gadzoox.com [216.52.31.24]) by ietf.org (8.9.1a/8.9.1a) with SMTP id TAA24183 for ; Mon, 12 Feb 2001 19:07:19 -0500 (EST) Received: (from majordom@localhost) by standards.gadzoox.com (8.8.7/8.8.7) id OAA20787 for ipfc-list; Mon, 12 Feb 2001 14:32:42 -0800 X-Authentication-Warning: standards.gadzoox.com: majordom set sender to owner-ipfc@standards.gadzoox.com using -f Received: from lightsand.com ([208.50.99.84]) by standards.gadzoox.com (8.8.7/8.8.7) with SMTP id OAA20784 for ; Mon, 12 Feb 2001 14:32:38 -0800 Received: from muralipc ([192.168.2.52]) by lightsand.com (8.9.3+Sun/8.9.1) with SMTP id PAA29109; Mon, 12 Feb 2001 15:41:11 -0800 (PST) From: "Murali Rajagopal" To: "Infante, Jon" Cc: "Weber, William L" , "IPFC Mailing List \(E-mail\)" Subject: RE: Question about WW_NN and RFC 2625 Date: Mon, 12 Feb 2001 15:42:36 -0800 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 In-Reply-To: <3A886A0D.3890B5DC@emulex.com> Sender: owner-ipfc@standards.gadzoox.com Precedence: bulk Content-Transfer-Encoding: 7bit Jon: I fully agree with your statements. They are technically accurate. A guess it helps to assert that a different NAA type for the WW_NN should be allowed. This should put to rest any subjective interpretations such as mine and others (we would not be having a email discussion on this at this time). Regards, Murali -----Original Message----- From: owner-ipfc@standards.gadzoox.com [mailto:owner-ipfc@standards.gadzoox.com]On Behalf Of Infante, Jon Sent: Monday, February 12, 2001 2:56 PM To: Murali Rajagopal Cc: Weber, William L; IPFC Mailing List (E-mail) Subject: Re: Question about WW_NN and RFC 2625 Murali, I agree that the document does not spell it out and that there no reason not to allow other NAA types for WWNN. With regards to FARP, I thought special consideration was given to make FARP a generic ELS command and not specifically tie it to our IP Profile. Thus the WWPN and WWNN in the FARP payload specifies the full 8 byte WWN and not just the last 6 bytes as in a IEEE address. Considering you can specify the full 8 byte WWNN, the full 8 byte WWPN and the Match Address Code Points specify match on just the WWNN, just the WWPN, or both, I don't feel there is any implication that the WWPN and WWNN use the same NAA Type. In fact, I don't think the FARP chapter mentions NAA at all. Jon Murali Rajagopal wrote: > Jon: > > I have to agree that the document does not spell it out. > > I also agree that there no reason not to allow other NAA types for WW_NN. > > But when you consider FARP, there is an implication that the WW_PN and WW_NN > use the same NAA Type. (Nobody said anything about a different NAA type > during FARP discussions!) > > Hope this clarifes what I said. > > -Murali > -----Original Message----- > From: owner-ipfc@standards.gadzoox.com > [mailto:owner-ipfc@standards.gadzoox.com]On Behalf Of Infante, Jon > Sent: Thursday, January 11, 2001 4:28 PM > To: Murali Rajagopal > Cc: Weber, William L; IPFC Mailing List (E-mail) > Subject: Re: Question about WW_NN and RFC 2625 > > Murali, > > I disagree. The Fibre Channel Network header ONLY contains the > WWPN. That is the main reason the NAA for the WWPN must > be 1 (mapping MAC addresses to WWPNs). > > Correct me if I'm wrong. however, I don't see any technical reason > for the WWNN NAA to be 1 nor did I infer that was the intent of > the many discussions the group had on this topic??? > > Jon > > Murali Rajagopal wrote: > > > The intent was NAA=1 for both. > > If it did not get reflected, then it was an oversight. > > > > -Murali > > > > -----Original Message----- > > From: owner-ipfc@standards.gadzoox.com > > [mailto:owner-ipfc@standards.gadzoox.com]On Behalf Of Infante, Jon > > Sent: Thursday, January 11, 2001 8:13 AM > > To: Weber, William L > > Cc: IPFC Mailing List (E-mail) > > Subject: Re: Question about WW_NN and RFC 2625 > > > > Bill, > > > > My understanding is that only the World Wide Port Name requires NAA = 1. > > It is not a requirement for the World Wide Node Name to have NAA = 1; > > > > Jon Infante > > Emulex > > > > "Weber, William L" wrote: > > > > > I have a question concerning RFC 2625, IP and ARP over Fibre Channel, > and > > > Network_Address_Authority (NAA) of World Wide Names (WWN). I thought > this > > > would be the proper forum to present the question. > > > The question is: > > > Can the Node WWN (WW_NN) use NAA=5 for IEEE registered format and > still > > > be > > > compliant/compatible with RFC 2625? > > > > > > I understand that the RFC 2625 requires the Port WWN (WW_PN) to use NAA > = > > 1 > > > for IEEE 48-bit addresses. RFC 2625 does not explicitly state the NAA > for > > > the WW_NN, however the examples in Appendix E (i.e. E.4.2) show the > WW_NN > > > with NAA=1. > > > > > > Thanks, > > > // Bill Weber Phone: +1-610-648-7022 > > > // Unisys Corporation Fax: +1-610-648-4010 > > > // 2476 Swedesford Rd. mailto:William.Weber@unisys.com > > > // Malvern, PA 19355-9700 From owner-ipfc@standards.gadzoox.com Tue Feb 13 17:47:15 2001 Received: from standards.gadzoox.com (standards.gadzoox.com [216.52.31.24]) by ietf.org (8.9.1a/8.9.1a) with SMTP id RAA01990 for ; Tue, 13 Feb 2001 17:47:14 -0500 (EST) Received: (from majordom@localhost) by standards.gadzoox.com (8.8.7/8.8.7) id NAA21586 for ipfc-list; Tue, 13 Feb 2001 13:05:03 -0800 X-Authentication-Warning: standards.gadzoox.com: majordom set sender to owner-ipfc@standards.gadzoox.com using -f Received: from ludwig.troikanetworks.com (ludwig.troikanetworks.com [12.31.172.3]) by standards.gadzoox.com (8.8.7/8.8.7) with SMTP id NAB21583 for ; Tue, 13 Feb 2001 13:04:58 -0800 Received: by host03.troikanetworks.com with Internet Mail Service (5.5.2653.19) id ; Tue, 13 Feb 2001 14:15:16 -0800 Message-ID: From: Bill Moshier To: ipfc@standards.gadzoox.com Subject: RE: IP over Fibre throughput numbers Date: Tue, 13 Feb 2001 14:15:15 -0800 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C0960A.70D46160" Sender: owner-ipfc@standards.gadzoox.com Precedence: bulk This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C0960A.70D46160 Content-Type: text/plain; charset="iso-8859-1" Dennis - I've been able to get more than 850Mbs (more than 85 MBytes/sec), on Windows 2k using Troika fibre-channel cards and Brocade switches, using iometer to measure the throughput. You might check to see if you can use large packets (size greater than one fibre-channel frame). My numbers were with 16Kbyte ip packets. Bill Moshier -----Original Message----- From: Murali Rajagopal [mailto:muralir@lightsand.com] Sent: Monday, February 12, 2001 1:00 PM To: ipfc@standards.gadzoox.com Subject: FW: IP over Fibre throughput numbers Please respond to the IPFC reflector if you have any comments. -----Original Message----- From: Dennis Solari [mailto:dys70@amdahl.com] Sent: Wednesday, January 17, 2001 10:37 AM To: muralir@lightsand.com Subject: Fw: IP over Fibre throughput numbers Just curious if you received this email from me? Dennis (Amdahl Corp) 678-319-4226 ----- Original Message ----- From: Dennis Solari To: muralir@lightsand.com Sent: Tuesday, January 16, 2001 9:41 PM Subject: IP over Fibre throughput numbers Murali, I am currently testing the speed I can get with IP over fibre channel using: * two Sun Ultra 5's (360Mhz) * both are have Emulex LP8000 SCSI/IP combo fibre channel PCI cards * connected through a Brocade silkworm 2400 switch. I can only drive IP over fibre to as high as ~200 Mbs. using iperf (variation of ttcp utility) What are some numbers you have seen doing bandwith testing while tweaking the Unix TCP/IP stack? Do you know were I can get other performance numbers of people doing the same type performance tests? Thanks in Advance, Dennis ------_=_NextPart_001_01C0960A.70D46160 Content-Type: text/html; charset="iso-8859-1"
Dennis - I've been able to get more than 850Mbs (more than 85 MBytes/sec), on
Windows 2k using Troika fibre-channel cards and Brocade switches, using
iometer to measure the throughput. You might check to see if you can use large
packets (size greater than one fibre-channel frame).  My numbers were with 16Kbyte
ip packets. 
 
Bill Moshier
-----Original Message-----
From: Murali Rajagopal [mailto:muralir@lightsand.com]
Sent: Monday, February 12, 2001 1:00 PM
To: ipfc@standards.gadzoox.com
Subject: FW: IP over Fibre throughput numbers

Please respond to the IPFC reflector if you have any comments.
 
-----Original Message-----
From: Dennis Solari [mailto:dys70@amdahl.com]
Sent: Wednesday, January 17, 2001 10:37 AM
To: muralir@lightsand.com
Subject: Fw: IP over Fibre throughput numbers

Just curious if you received this email from me?
 
Dennis (Amdahl Corp)
678-319-4226
 
----- Original Message -----
Sent: Tuesday, January 16, 2001 9:41 PM
Subject: IP over Fibre throughput numbers

Murali,
I am currently testing the speed I can get with IP over fibre channel using:
* two Sun Ultra 5's (360Mhz)
* both are have Emulex LP8000 SCSI/IP combo fibre channel PCI cards
* connected through a Brocade silkworm 2400 switch.
 
I can only drive IP over fibre to as high as ~200 Mbs. using iperf (variation of ttcp utility)
 
What are some numbers you have seen doing bandwith testing while tweaking the Unix TCP/IP stack?
Do you know were I can get other performance numbers of people doing the same type performance tests?
 
Thanks in Advance,
Dennis 
------_=_NextPart_001_01C0960A.70D46160--