FC trouble

[To-M]

Send the logs in to support I want to take a look at them to see if there is anything in the dmesg, error, critical error and test logs to help out.

[symm]

What Lun Numbers are you using ?
I had problems with VMware where I was using the same lun numbers, Vmware did not like that.
Check the knowledgebase there was a post about the lun number problem.

[jmo]

What Lun Numbers are you using ?
I had problems with VMware where I was using the same lun numbers, Vmware did not like that.
Check the knowledgebase there was a post about the lun number problem.

As all the (first) disks are boot disks, they're on LUN 0 of the respective group. I even receive a warning (or was that an error?) when there's no LUN 0 defined

I've checked the knowledge base, couldn't find the arcticle you were refering to, but stumbled over article 162... not the one you meant, but possibly related to this case. Although I've seen absolutely no errors in the logs, especially no kernel errors.

I'll get back when I have more information.

With regards
Jens

[symm]

Hi jmo

I know your using xen server.
and not sure if this will apply to you, but i've seen this in both vmware and virtual iron.
The lun number must be different across all luns, even in different targets.

http://communities.vmware.com/message/701181#701181

[To-M]

Yes, this is even true for iSCSI with VMware, example below:

Target 0 :

Lun 1
Lun 2

Target 1:

Lun 3
Lun 4

and so on.

[Robotbeat]

Yes, this is even true for iSCSI with VMware, example below:

Target 0 :

Lun 1
Lun 2

Target 1:

Lun 3
Lun 4

and so on.

Is this true even when using Fibre Channel?

[Robotbeat]

Is it true that we can't use lun0 in VMWARE under fibre channel? I have a customer right now doing that and so far (since yesterday) this hasn't been an issue. Is it an issue?

[jmo]

Send the logs in to support I want to take a look at them to see if there is anything in the dmesg, error, critical error and test logs to help out.

Logs will be coming up on Saturday or Monday, I currently have no access to them.

I have had a close look at those files and couldn't find anything obvious nor related to the affected LVs (although I'm no experienced *DSS* engineer, I feel knowledgable in the Linux area - I've seen it break it's egg shell and was on the mailing list when Andrew S T. was arguing with Linus T. ).

Thanks for looking into this

Jens