Question

Detecting data inside of a mount point/disk library not associated to active backup data?

  • 31 May 2023
  • 1 reply
  • 56 views

Badge +2

Hello there CV community,

We are in a situation where we expect there may be some data in one of our disk libraries that is not associated to retained backup jobs. The jobs data stored on this disk library utilizes deduplication, so primarily the folder data contains deduplication chunks. We would like to validate that the content on the disk library is “current” i.e. associated with retained & deduplicated backup data. In this case, the storage is Azure object storage.

Based on Damian’s post here: Clean Orphan Data , what is it? | Community (commvault.com), our hope was that we could use the DDB space reclamation feature alongside the ‘Clean orphaned data’ option to do this automatically. To test this, I ran the operation in a test environment after having stored some arbitrary files alongside legitimate backup data within the storage. This wasn’t successful, and I expect I have either misunderstood his description of the functionality or perhaps Commvault is specifically looking for data that matches the characteristics of Commvault chunk data and thus my test is flawed.

In any event, the files I added alongside the legitimate chunk data was not removed and I’m left looking for another strategy to accomplish this. The container contains upwards of 100 million blobs, making manual remediation a chore, but it might be possible if it were possible to get an output from Commvault of actively chunks stored on disk. 

Anyone have any thoughts or suggestions? Thanks in advance for your time.

Cheers,


1 reply

Userlevel 6
Badge +15

Do the mount paths that the DDBs are on support sparse files?

 

Reply