i'm not using multipath, and it seems that i have the exact same problem.
(identical ietd version, debian squeeze on amd64, open-iscsi as initiator)

when i restart ietd during i/o, the session on the initiator side goes unsuable
readonly, with no way to get it working other than logout/login again.
this of course fails all pending i/o and is bad.
longer timeouts on initiator side just make it a longer wait for failure.

have you found a solution yet ?

maybe the trick described here will help ?
http://blog.wpkg.org/2007/09/27/solving-reliability-and-scalability-problems-with-iscsi-part-2/

regards


-- 
You received this message because you are subscribed to the Google Groups 
"open-iscsi" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/open-iscsi?hl=en.

Reply via email to