Cluster resource Comatose with CIFS

[alexpeeters]

Hi all,

I've got the exact same errors, and I'm in the process of debugging right now. The first thing I've found is that on the faulty volume object in eDir there is an attribute called: linuxNCPMountPoint. In my case I filled it with: NSS /media/nss/BEHEER. I've copied that from a working OES2SP1 cluster.

That resolved the 626 errors. But I'm still not able to set the CIFS Virtual Server name correctly and connect to that name. However, connection to the resourcevolume through clusternode CIFS Virtual Name using CIFS is now functional.

Maybe I will post some more information later on as I'm still searching for answers.

Kind regards,

Alex Peeters

Hi,

We have a OES 2 SP1 Linux 2x node cluster. Every thing has been working
fine.

However we want to add CIFS access to a couple of the volumes. When
adding the CIFS protocol, we now have a problem that the CLUSTER
resource goes COMATOSE

Each server has a full copy of both partitions in the tree and Timeync
seems fine.

We did have a SLP probably when we set the servers up, however this has
since been resolved and running slptools shows up ndap resources.



ERROR: ENTRY: CIFSNDSGetCIFSServerInfo: Read attribute failed: -626

May 3 10:22:44 servername CIFS[2046]: CRITICAL: CLI:
CIFSNDSGetCIFSServerInfo() CIFSNDSGetCIFSServerInfo

May 3 10:22:44 servername CIFS[2046]: ERROR: ENTRY: Resolve Name failed
with error code -626

May 3 10:22:44 servername CIFS[2046]: ERROR: ENTRY: CIFSNDSReadAttr:
Error in ReadAttr -626

May 3 10:22:44 servername CIFS[2046]: ERROR: ENTRY:
CIFSNDSGetCIFSServerName: Read attribute failed: -626

May 3 10:22:44 servername CIFS[2046]: WARNING: CLI: CIFS: Unable to get
server name from the FDN, error = -626

May 3 10:22:46 servername adminus daemon: umounting volume NEWUSERS
lazy=1

Here is an output of Time Sync

---------------------------+---------+---------+-------
DS Replica Time Time is Time
Server name Version Depth Source in sync +/-
---------------------------+---------+---------+-------
..srvA.SERVERS..20216.89 0 Non-NetWare Yes 0
..srvB.SERVERS..20217.06 0 Non-NetWare Yes 0
..srvC(NW).org 20216.55 0 Primary Yes 0
..srvD.SERVERS..20217.06 0 Non-NetWare Yes 0
---------------------------+---------+---------+------

The only other thing is the Cluster Resource sit in

ou=cluster.ou=res

and the servers sit in (and the Cifs Proxy Users)

ou=server.ou=res

I have applied the same trustee permissions to the ou=cluster resource
as set on ou=servers.

Any thoughts??