Hi all,

We had a volume dismount unexpectedly today, with what appears to be a
volume corruption. The server is OES Linux SP2, and is a VMware guest
running on ESX Server 3.5, clustered using NCS with another VMware
guest. The data volume is on a raw LUN (RDM) in our fibre channel disk
array, and the disk array itself reports no media errors or anything
like that.

Here are some extracts from /var/log/messages:
The volume went down at around 10:00, we then did a full pool verify,
remounting it to test at 12:36, and then a full rebuild, remounting the
volume at 16:51.

Any suggestions as to what the problem is? We've tested and everything
*seems* OK (including trustees on relevant directories), but we're
reluctant to remount the volume if there is some sort of trustee corruption.

I did a search on Novell and on Google for these errors:

CIFS[9950]: ERROR: CODIR: Unknown volume name received as parameter
CIFS[9950]: ERROR: RPC: Update of trustee failed for path /Snapshots

There were ZERO hits, and that scares me, because it usually means i'll
get the "no one else in the world has ever seen this problem" response,
which is never fun. ;-)

Thanks in advance,