To: ips@ece.cmu.edu
Subject: 
MIME-Version: 1.0
From: "Julian Satran" <Julian_Satran@il.ibm.com>
Message-ID: <OF2D6A05C8.784D5D2F-ONC2256D60.006FE5A2-C2256D61.001BF581@telaviv.ibm.com>
Date: Sat, 12 Jul 2003 08:05:47 +0300
Content-Type: text/plain; charset="US-ASCII"
Sender: owner-ips@ece.cmu.edu
Precedence: bulk

Yes as Phaneedra pointed out it may MaxRecvData... can change. During 
recovery one should pay attention to those changes and this is the reason 
for defining what to expect on retransmission.

Regards,
Julo
----- Forwarded by Julian Satran/Haifa/IBM on 11-07-03 23:22 -----

Eddy Quicksall <eddy_quicksall@ivivity.com> 
10-07-03 14:50

To
Julian Satran/Haifa/IBM@IBMIL, "Ips@Ece.Cmu.Edu (ips@ece.cmu.edu)" 
<ips@ece.cmu.edu>
cc

Subject
iSCSI: MaxRecvDataSegmentLength changing in FFP






MaxRecvDataSegmentLength can change during Full Feature Phase (USE is 
ALL).
 
Was it the intent that it can change at any point (even in the middle of 
transmitting a sequence)?
 
Eddy
 
----- Forwarded by Julian Satran/Haifa/IBM on 11-07-03 23:22 -----

Andre Hedrick <andre@linux-ide.org> 
10-07-03 14:57

To
Eddy Quicksall <eddy_quicksall@ivivity.com>
cc
Julian Satran/Haifa/IBM@IBMIL, "Ips@Ece.Cmu.Edu (ips@ece.cmu.edu)" 
<ips@ece.cmu.edu>
Subject
Re: iSCSI: MaxRecvDataSegmentLength changing in FFP







Hi Eddy,

From what it is read about variable pdu lengths changing on the fly, one
has to assume the boundaries imposed can be in flux.  This may be more for
flexablitiy when dealing with target whose effective hba is a messaging
monster (FC).

Well on a side note, is there anyone testing or shipping erl2 who is
interested in compatiblity testing?

One additional side note, well question.  I assume any initiator which
issues notunderstood for standard keys exchanged is deemed a bad
initiator?  This is regardless if the ERL supported but either side.

Cheers,

Andre Hedrick
LAD Storage Consulting Group

On Thu, 10 Jul 2003, Eddy Quicksall wrote:

> MaxRecvDataSegmentLength can change during Full Feature Phase (USE is 
ALL).
> 
> Was it the intent that it can change at any point (even in the middle of
> transmitting a sequence)?
> 
> Eddy
> 
> 


----- Forwarded by Julian Satran/Haifa/IBM on 11-07-03 23:22 -----

"mphaneen" <mphaneen@npd.hcltech.com> 
10-07-03 15:49
Please respond to
"mphaneen" <mphaneen@npd.hcltech.com>


To
"Eddy Quicksall" <eddy_quicksall@ivivity.com>, Julian 
Satran/Haifa/IBM@IBMIL, <ips@ece.cmu.edu>
cc

Subject
Re: iSCSI: MaxRecvDataSegmentLength changing in FFP






Hi Eddy,
    Yes, the MaxRecvDataSegmentLength can be changed even in
the middle of the sequence transmission. And an initiator (ERL > 0)
would need to issue R-Data SNACK to request retransmission
(possibly, from the time of change in MaxRecvDataSegmentLength)
of the unacknowledged data from target, expecting resegmenation.
 
    HTH
    Phani

Phaneendra. M
Member Technical Staff
HCL Technologies Ltd.
Networking Technologies Lab
Chennai, India.
Phone: +91 044 23728366 extn: 1135
 
Visit us at:
http://www.hcltech.com/san
 
Success is sweet and sweeter if long delayed and gotten through many 
struggles and defeats.
----- Original Message ----- 
From: Eddy Quicksall 
To: Julian Satran ; Ips@Ece.Cmu.Edu (ips@ece.cmu.edu) 
Sent: Thursday, July 10, 2003 5:20 PM
Subject: iSCSI: MaxRecvDataSegmentLength changing in FFP

MaxRecvDataSegmentLength can change during Full Feature Phase (USE is 
ALL).
 
Was it the intent that it can change at any point (even in the middle of 
transmitting a sequence)?
 
Eddy