|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: iSCSI: Question about CIDI guess I missed that in the July 10 draft. I had assumed that the reply might come back on one of the other connections, but on reflection that was rather silly of me :-) Robert.. > On Wed, 23 Aug 2000 14:35:30 -0400 you said: > Since the login response is returned on the same connection as the login > command came in on, the initiator knows that its the login on that > connection that is being accepted ( or rejected ). > > Passing the CID back in the login reply at offset 8 is not what the RFC > defines ( although it doesn't seem an unreasonable idea). Since that place > in the login response is reserved, a very picky initiator implementation > might object that the field contained a non zero value. > > Steve > > > -----Original Message----- > From: rbg@ipperformance.com [mailto:rbg@ipperformance.com] > Sent: Wednesday, August 23, 2000 1:28 PM > To: Ayman.M.Ghanem@seagate.com > Cc: ips@ece.cmu.edu > Subject: Re: iSCSI: Question about CID > > > > I'm passing it back at the same offset (8) in the login reply. > > Robert.. > > > On Wed, 23 Aug 2000 10:19:31 -0500 you said: > > I have a question about the CID field sent by the initiator in a login > > message. > > When the target sends back login accepted, how does the initiator know > > which > > login request was accepted (in case it wants to establish multiple > > connections > > per session with each connection requiring a login)?. The login response > > does > > not return the CID to the initiator. > > > > -Ayman > > > _____________________________________________________________ > Robert Gordon rbg@ipperf.com > IP Performance, Inc > Austin, Texas. <http://www.ipperf.com> _____________________________________________________________ Robert Gordon rbg@ipperf.com IP Performance, Inc Austin, Texas. <http://www.ipperf.com>
Home Last updated: Tue Sep 04 01:07:45 2001 6315 messages in chronological order |