RE: iSCSI - Question about StatSN and NOP-In PDU

[Eddy Quicksall <eddy_quicksall@ivivity.com>]

Section 10.19 says:

 

Otherwise, when a target sends a NOP-In that is not a response to a Nop-Out received from the initiator, the Initiator Task Tag MUST be set to 0xffffffff and the Data Segment MUST NOT contain any data (DataSegmentLength MUST be 0).

Section 10.19.2 says:

 

The StatSN field will always contain the next StatSN. However, when the Initiator Task Tag is set to 0xffffffff StatSN for the connection is not advanced after this PDU is sent.

 

So, this means you should not advance StatSN as you have suggested.

 

Eddy

-----Original Message-----
From: Julian Satran [mailto:Julian_Satran@il.ibm.com]
Sent: Monday, December 02, 2002 8:18 PM
To: Mark Freedkin
Cc: ips@ece.cmu.edu; owner-ips@ece.cmu.edu
Subject: Re: iSCSI - Question about StatSN and NOP-In PDU

Mark,

I am not sure that I understand what is exceptional in this situation.
I assume that your Nop-In has TTT (it is ping) and requests a Nop-Out.


Julo

Mark Freedkin <Mark.Freedkin@quantum.com>
Sent by: owner-ips@ece.cmu.edu

03/12/02 02:04

To	ips@ece.cmu.edu
cc
Subject	iSCSI - Question about StatSN and NOP-In PDU

Our iSCSI target implementation generates an unsolicited NOP-In PDU as a
"ping" mechanism after 5 minutes of inactivity within an established
connection.  We also increment the StatSN that is carried within that PDU.
If the initiator happens to send a PDU at the exact same time as our NOP-In
PDU, the initiator and target StatSN values will get out-of-sync.

Is there anything in the Draft 19-90 specification that describes how to
handle this unique condition?

Mark Freedkin
Quantum Corporation