Got a warning message from our 5250 appliance today, regarding the status of the MSDP. It suggests I "cleanup the partition and re-check status".
Is there a documented process for 'cleaning up the partition'? I feel like the MSDP is just a big blackhole that data goes into, but then I have little or no control over it.
In previous h/w, I was able to NFS mount the partition onto my RHEL master server, use unix tools to ID old images, and NBU commands to make the space available - nbstlutil inactive, nbstlutil cancel and bpexpdate, for example.
Yes MSDP maintenance can be tricky, as deleting a single (large) backup may have little to no affect in changing the available space.
There is possiblilty that there are "orphaned" images on the appliance - these are ones that have been deleted from the NetBackup catalog, but for unknown reasons were never deleted from MSDP. You would need to log a support case to get help with this one (there is a tool available from support which should only be used under guidance).
If you need to use NetBackup to delete backup images (e.g. bpexpdate), then this is difficult. If you know that some clients do not deduplicate well, then these would be good candidates to remove older backups which should show posititve results in reducing the pool utilisation.
Finally, if there is unused space in the appliance, it is very simple to grow the MSDP size using the appliance WebUI. Unused space may be shown as available or unallocated.