A month ago I upgraded our Netbackup server from 6.0MP5 to 6.5.2a. We were told this was necessary to support new hardware (VTL), and addition of some new clients (MAC and Unix). My Netbackup environment has been in place working flawlessly (more or less) for over two years, backing up Windows and SQL servers.
Since the upgrade I have experienced "Status Code 239 - the specified client does not exist in the specified policy" three or four times on Full SQL backup policies. These policies were in place and working long before the upgrade, and have survived previous maintenance pack upgrades with no problem. Nothing has changed but the NBU upgrade.
Configuration Background: My environment consists of an MS SQL 2000 cluster running 7 SQL instances, along with several standalone SQL 2000 and 2005 servers with their own databases. Each of these SQL instances contain dozens of databases, so each day I am backing up hundreds of databases ranging from a few MB in size to several hundred GB.
My Netbackup SQL Policies are separated by instance, and each instance has a Full backup policy and a Transaction Log backup policy. The Full's run every night, and the Trans run every one or two hours during the day starting between 4 and 6:00 am.
Since the upgrade to 6.5.2a, though, I have had random failures of one or another Full Backup policy. Each time has been a different instance on the SQL cluster - I don't think I have had any failures on the standalone SQL servers. By the next morning, though, the Trans backups for that instance are running fine, but the previous nights' Full failed to backup every database. This morning I came in and had a status 2 (none of the files were backed up) alert in my email for this SQL instance, and 45 database failures in the job monitor in Netbackup.
Each time this has happened, I can either rerun the policy or wait until the next scheduled time, and it is successful.
Has anyone seen this behavior after upgrading to 6.5.2a?
