Clusternode 2 cant find quorumdisk

Hi!

Specs: We have a 2003-sp2-cluster, which has been running fine for more
than 2 years now. Setup is Active/Active, with fileshares on each node
(3 and 4 disks each, 3 and 4 shares).

Today I did the usual MS-Update thingie, moving the ressources to node
1, updating node 2, reboot node 2. When I reverted node's 2 ressource to
itself, everything worked fine.
I just recognized by a glance, that I didn't see all clusterdisks in
Explorer on node 2 (which have been there usually, but node 1's disks
were not accessible).

Update for node 1 has been scheduled for tomorrow morning.

After 2 hours working fine, I got connection probs to the shares on node
2, first only temporarily, after some minutes I couldn't connect to it
anymore.

I got Errors 1055 and 1069 in eventlog, Errorcode 64.
I found
http://support.microsoft.com/?scid=k...2478&x=14&y=15
but that didn't help.

Checking cluster.log on node 2 I found
"[DiskArb] Failed to read (sector 12), error 170" for the quorumdisk and
"FMGetQuroumResource failed, error 170".

I googled a bit, but a reboot or manually "net stop/start clussvc"
didn't help anything (http://support.microsoft.com/kb/249194) - still
the node 2 is not able to mount the quorumdisk.

Free space on the Quorum is 1GB of 8GB.

Any ideas, how to fix this?

thx and greetings
michi

This is why you have a cluster, patch management !

First of all, move all your cluster groups back to Node 1, which is
unpatched.
Now you have working shares and resources again, so you are minimizing
downtime !

Second, uninstall the patches you added to Node 2, and reboot.
Test your cluster groups, move them into Node 2 and ensure all is working.

At this point you have a working cluster.

You can now test your updates/patches on a test cluster, and figure out
which of the patches is causing your problems.

Rgds,
Edwin.




"Michael Hirczy" <> wrote in message
news:493e503a$0$11868$ t...
> Hi!
>
> Specs: We have a 2003-sp2-cluster, which has been running fine for more
> than 2 years now. Setup is Active/Active, with fileshares on each node
> (3 and 4 disks each, 3 and 4 shares).
>
> Today I did the usual MS-Update thingie, moving the ressources to node
> 1, updating node 2, reboot node 2. When I reverted node's 2 ressource to
> itself, everything worked fine.
> I just recognized by a glance, that I didn't see all clusterdisks in
> Explorer on node 2 (which have been there usually, but node 1's disks
> were not accessible).
>
> Update for node 1 has been scheduled for tomorrow morning.
>
> After 2 hours working fine, I got connection probs to the shares on node
> 2, first only temporarily, after some minutes I couldn't connect to it
> anymore.
>
> I got Errors 1055 and 1069 in eventlog, Errorcode 64.
> I found
> http://support.microsoft.com/?scid=k...2478&x=14&y=15
> but that didn't help.
>
> Checking cluster.log on node 2 I found
> "[DiskArb] Failed to read (sector 12), error 170" for the quorumdisk and
> "FMGetQuroumResource failed, error 170".
>
> I googled a bit, but a reboot or manually "net stop/start clussvc"
> didn't help anything (http://support.microsoft.com/kb/249194) - still
> the node 2 is not able to mount the quorumdisk.
>
> Free space on the Quorum is 1GB of 8GB.
>
> Any ideas, how to fix this?
>
> thx and greetings
> michi