Fellow toasters:
We're trying to use Veritas NetBackup 3.2GA (patched to the ...443 patch for Solaris) to backup a couple of f760s. The NDMP module is installed and patched. There's a tape library shared between the two filers, each with 2 AIT-2 drives. The connection between the filers and the library is SCSI-diff, from the main board SCSI connection.
NDMPd is running on the filer and we can exercise the robotics with no problems from the NetBackup server (a SPARC Ultra 10 running Solaris 2.7). Everything seems to be configured properly, as far as we can tell. The class we created seems to run "manual backups" without too much problem. But the scheduled backups for large filer volume typically do not complete successfully. Such backups get underway and write some number of gigabytes to tape (like 45GB or more or the 450GB volume) and then the backup ends prematurely with a status 196 (suggesting the backup was trying to run outside its backup window).
We tried adding the /etc/system changes suggested in the release notes for Solaris, but this did not seem to help much.
Also, does anybody fully understand the interaction between "duration" and the schedules' start times? It looks like backup may be failing when they start BEFORE the duration ends, but when they are still running AFTER the duration ends.
--rick _________________________________________________________________________ Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
Share information about yourself, create your own public profile at http://profiles.msn.com.