Date - 09-10
Affected Product & Version - NetVault Backup All
Affected Module & Version - NVDB
OS Version - All
Application Information - N/A
Netvault database backup on the server fails with a channel error.
Symptoms: The Netvault Database backup fails with a channel error even though it is backing up to a local device.
Behind the channel error is the message:
DATA :05020 96 0 91747 Other process has died unexpectedly
MEDWRITE:05020 190 1 91747 Device write data in: FAILURE
check the last element of the Netvault database that was being backed up. It may well be 'drivevents'
The solution to this is now the same as for solution 1778 where the NVDB backup is hanging rather than failing:
Occasionally these tables can become corrupted. The information contained in them is not critical for the day to day running of Netvault:Backup and the offending table can be removed using the following procedure:
1. Identify which table is corrupted. Its name will be clearly visible as the last word of the last line in the NVDB backup log. As an example we shall use "driveperformance".
2. Navigate your way to the Netvault installation directory then into reports\dbmgr\tables
3. Move your current "driveperformance" directory (or whichever table has been identified in step 1) out of the way to another directory completely separate from Netvault:Backup.
4. Stop and restart Netvault:Backup.
5. Try the NVDB backup again.
The table removed may well be rebuilt as and when Netvault:Backup needs to access it but if you prefer we can supply a blank table. The folder can also be replaced by restoring it from the last working NVDB backup