Unsolved
This post is more than 5 years old
9 Posts
0
1334
March 25th, 2014 10:00
Unmapping a LUN
We've been having some trouble with getting multi-path errors on our Citrix 6.1 environment when attached to our VNX5300. (Mostly during PHD backups). They always get back to full connectivity within a few seconds, but the Citrix rep said all we needed to really do was unmap the LUN, and then remap it. (On the VNX side, which struck me as odd, you'd think you'd do that on the Xenserver side.)
Anybody had any experience with this? I'm guessing that removing the LUN from the storage group, and then re-adding it should accomplish that. Anybody had any experience with doing that, and any things to look out for?
Thanks
Jim
dynamox
9 Legend
•
20.4K Posts
0
March 25th, 2014 11:00
that suggestion does not make any sense to me. What failover mode are you using ?
jericson1
9 Posts
0
March 27th, 2014 06:00
It seemed kind of odd to me, as well. OTOH, I'm still pretty new to this.
We are running with Mode 4 (ALUA) enabled.
dynamox
9 Legend
•
20.4K Posts
0
March 27th, 2014 06:00
are all hosts experience this issue at the same time ? If yes, how are they zoned ?
jericson1
9 Posts
0
March 31st, 2014 09:00
The Citrix farm gets the error. (Predominantly when there is are disc backups being run via PHD.) The error is a multipath alert, though all the paths are active again almost immediately.
There is no error on the VNX 5300 at all.
We had one of the Citrix tech support looking at the error, and he said there was an error popping up on the first LUN, and his saying taking it offline and then on again would fix it. (I wasn't here that day, so my reports of his are second hand.)
jericson1
9 Posts
0
March 31st, 2014 12:00
Also, it's all iSCSI, so I don't believe we have any zones setup. (Just some different subnets.)
dynamox
9 Legend
•
20.4K Posts
0
March 31st, 2014 20:00
have you looked at SP events during the time when you experience these "disconnects" ? Do you see anything interesting around that time frame?