Hi all,

I have a 7 node cluster installed with SLES10SP4 / OES2SP3 (upgraded from SLES10SP3 / OES2SP2a), including the latest patches / maintenance releases. The cluster is hosting volumes for our students, 20 volumes in total.

The problem we have is that sometimes the volumes are unreachable, while they are still online in the cluster. When you go to the folder through a linux console it works fine (/media/nss/<VOLUME>) but when you try the volumes through windows explorer you see no content or you get an error message that the volume is unreachable. I'v been looking at my /var/log/messages to check for any errors during the time it occurs and I see the following messages:

Oct 4 10:16:29 ROC-DN-NODE5 kernel: Couldn't get the security equivalence vector for uid 4294865098
Oct 4 10:16:46 ROC-DN-NODE5 kernel: Couldn't get the security equivalence vector for uid 4294865097
Oct 4 10:16:59 ROC-DN-NODE5 kernel: Couldn't get the security equivalence vector for uid 4294865096
Oct 4 10:17:05 ROC-DN-NODE5 kernel: Couldn't get the security equivalence vector for uid 4294865095
Oct 4 10:17:12 ROC-DN-NODE5 kernel: Couldn't get the security equivalence vector for uid 4294865094

Another message I see is:

Oct 4 10:45:30 ROC-DN-NODE5 kernel: OID_AddEntryIfNotThere - Unable to map GUID to DN using NCP eDir

I cannot find anything useful for these messages, does anybody have an idea? And is it related to our problem (I guess so ;-))?

The ncpserv.log is filled with messages like this one:

[! 2011-10-04 10:17:06] znssDeleteFile: zDelete failed with rc=20859

I really hope someone can help me because now I have to migrate the resource to another node and reboot the node with the problem. And you can just wait for the next node to get the same problem.
Any and all help will be greatly appreciated.

Regards,

Frank