Peter, thanks for your comments.

I checked latter in the snoop capture file and 
there is further short TCP session.

But it consists of a single iscsi Login Command.
(It's packet #396 in the file.)
There is no response to this login packet from the target. 
After 2.68 seconds the target FIN's the connection.
Maybe this is where the core dump occurs?

But for this TCP session, the Initiator is 192.168.1.102
and the target is 192.168.1.100
This is different to the session I decoded in my
previous post.  There the target was also 192.168.1.100
but the initiator had an Internet IP like 76.XX.XXX.XXX

Here are the key-value pairs from the iscsi Login packet:
  SessionType=Normal
  AuthMethod=CHAP
  TargetName=iqn.1986-03.com.sun:02:25515cb7-069d-e4a2-c701-82ef4809bffb
  InitiatorName=iqn.2005-03.com.sns:mac-manul-pordss-macboo

Note that the 'InitiatorName' is specified here,
so this does not seem to tie in with Jim's
current theory on what's happening.

Maybe this 'Normal' session and the previous 'Discovery'
sessions are from different initiator version??
But the 'InitiatorName' matches, so I don't know. Strange.

I think we need to get Frank to redo the capture.
Best Regards
Nigel Smith
-- 
This message posted from opensolaris.org
_______________________________________________
storage-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/storage-discuss

Reply via email to