LMS 3.0 running on Solaris 10
I get messages stating /var and /opt are near full. In the past, I've done some file (syslog, job purge, etc) cleanup/deletions to fix the problem. However, we're getting these message more and more lately.
I talked to our UNIX admin and we have other drives/partitions on the same server that are a lot larger than the ones being used.
We backed up everything and tried to move /var from /dev/md/dsk/d3 to /dev/md/dsk/d91. At the same time, we moved /opt from /dev/md/dsk/d5 to /dev/md/dsk/d92.
The system was halted during this move.
The server was rebooted and we started LMS. The inventory in Common Services and RME (these are the sections we primarily use) look fine. I ran into a problem when I tested the Sync Archive jobs. It errored out after just 2 seconds stating CM0051 Unable to connect to ConfigMgmtSever process Cause: URN_NOT_FOUND : urn "Archive Mgmt" : Not found !! Action: Please restart ConfigMgmtSever process.
I checked the processes and ConfigMgmtServer had a state of Program started - No mgt msgs received
Does anyone have any experience with this?
Does LMS use the disk device address or the raw device address? Maybe there is a file with these paths stored and it needs updated?
Paths cannot be easily changed, but physical drives/partitions should not cause problems. That is, if /var/adm/CSCOpx and /opt/CSCOpx still exist, but just happen to be on different physical disks, everything should still work. Post the output of the pdshow command, the dcmaservice.log, and the daemons.log.