Linux Remote/Local selections not working

[Michael_Willi6]

We're having a few problems now we've upgraded to BE10 from earlier versions.

When I do a backup of our Linux servers, if I have "follow remote mount points" unticked, and "follow local mount points" ticked, only the root filesystem and whatever is on the root hard-drive (i.e. other partitions on the same physical hard drive as the root filesystem) are backed up.

If I re-run the backup and tick on "follow remote mount points", now our directly attached storage drives are backed up, however all the NFS mounts are backed up too.

The behaviour I want is for all direct attached disks (i.e. local mount points) to be backed up and NFS mounts (i.e. remote mount points) to be completely ignored. Our mounts fluctuate a lot and we're unable to constantly change the selections list.

The hardware we're using is standard HP Proliant Servers using in this case the ida drivers (/dev/ida/c?d?p?) and standard cciss drivers (/dev/cciss/c?d?p?), both supplied in the kernel of SLES8.

Can you tell me how to get the agents to behave correctly?

[Rick_M_]

Can you provide an example of how these devices are mounted (ie the entries in your /etc/fstab?

The physical hardware type, scsi, cciss, ide or what have you, should not affect us in any way. RALUS recieves data by reading from the filesystem, rather than directly accessing the device.

It may be necessary for you to call Technical Support, so that they can obtain agent logs, and find out what messages the agent might be returning (in debug mode) when it attempts to traverse the mount points.

You are correct in your belief that this should work without enabling "follow remote mount points". In testing, we encountered no unusual behavior of the type you're describing with SLES8. The Agent does contain sufficient diagnostic capability for Technical Support to assist in determining what is happening, however.

[Michael_Willi6]

Hi Rick, thanks for your reply!

My /etc/fstab looks like this:

/dev/ida/c0d0p2 / reiserfs defaults 1 1
/dev/ida/c0d2p1 /alchroot reiserfs defaults 1 2
/dev/ida/c0d0p1 /boot ext2 defaults 1 2
/dev/cciss/c0d0p1 /linpo/d1 reiserfs defaults 1 2
/dev/cciss/c0d1p1 /linpo/d2 reiserfs defaults 1 2
/dev/cciss/c0d2p1 /linpo/d3 reiserfs defaults 1 2
/dev/cciss/c0d3p1 /linpo/d4 reiserfs defaults 1 2
/dev/ida/c0d1p1 /opt reiserfs defaults 1 2
/dev/ida/c0d0p4 swap swap pri=42 0 0
devpts /dev/pts devpts mode=0620,gid=5 0 0
proc /proc proc defaults 0 0
/dev/cdrom /media/cdrom auto ro,noauto,user,exec 0 0
/dev/fd0 /media/floppy auto noauto,user,sync 0 0
/dev/ida/c0d3p1 /linpo/d5 reiserfs defaults 1 2
nis1:/homes /homes nfs defaults 0 0
rndsuse:/ha1/d1 /ha1/d1 nfs defaults 0 0
alchdb1:/alchdb1/d1 /alchdb1/d1 nfs defaults 0 0
alchdevmstl1:/adml1/d1/sandbox /sandbox nfs defaults 0 0
alchdevmstl1:/adml1/d1/testdecks /testdecks nfs defaults 0 0
alchdevmstl1:/adml1/d1/alch/root /opt/alch/root nfs defaults 0 0

All devices were created using HP's web-based RAID Array Configuration Utility, and then mounted using SuSE's YaST2 module using the detault options for everything.

How does RALUS identify what is local or remote, perhaps I could run a couple of commands to see if other sub-systems have the same issue?

I'm using the standard kernel supplied with SLES, not patched, i.e. 2.4.19-64GB-SMP.

We have this same problem with the older agents, we do not have this problem with SCO using the older agents.

Regards,

Mike

[Renuka_-]

Hello,

Please check the debug logs for the linux agent also as suggested previously.
http://support.veritas.com/docs/272988

NOTE : If we do not receive your reply within two business days, this post would be marked assumed answeredand would be moved toanswered questions pool.