Skip to main content
Solved

DDB Compaction On Media Agent with multiple DDB Stores

  • November 7, 2025
  • 2 replies
  • 45 views

dude
Community All Star
Forum|alt.badge.img+20
  • Community All Star

I have a media agent (MA1)  that hosts DDB Cloud Disk 1 and DDB On Prem Disk 2. DDB Cloud is running out of space and I need to run DDB Compaction to reduce space until we get more disks.

The outlined process says to

1- Stop all jobs

2 - Ensure that there is no SIDB2 process running on the MediaAgent.

I can stop all jobs that are using the DDB Cloud Disk 1, no issues here, however the media agent will continue to run jobs for our On-Prem copies using DDB On Orem Disk 2 which will continue to have the SIDB2 running.

 

Can I run the DDB Compaction for the Cloud DDB  even though the SIDB2 is running for the On Prem jobs?

What would be the recommend action here?

On a different note/request, is there a way Commvault Team could take a look at streamlining the compaction process via workflow or a different method that would not require a command line for each DDB store (file/DB/VM etc)?

 

Thank you

 

Best answer by kkarthikk

Hi ​@dude ,

 

The compaction process requires that no SIDB2 process is running on the Media Agent before you start compaction.


The SIDB2 process handles all DDB partitions on the Media Agent, and running compaction while SIDB2 is active (even for a different partition) can cause conflicts or corruption.

It is recommended to place the Media Agent in Maintenance Mode. This will suspend all jobs to all DDBs hosted on the Media Agent (both Cloud and On-Prem).

Wait for all SIDB2 processes to stop and run the compaction for the Cloud DDB.

Once compaction is complete, take the Media Agent out of maintenance mode and resume jobs.

But if the DDB disk is getting full and running out of space, may i know whether the DDB disk size is as per the hardware specifications mentioned below.

https://documentation.commvault.com/2023e/commcell-console/hardware_specifications_for_deduplication_mode.html

And ensure data aging and pruning are working. If there are many pending prunable or delete records, space may not be reclaimed efficiently.


Run a Data Retention Forecast Compliance (DRFC) report to identify jobs that should be aged off.


And currently there is no workflow or automated method available in Commvault to streamline DDB compaction across multiple DDBs or partitions.

Regards,

Karthik

2 replies

Forum|alt.badge.img+6
  • Vaulter
  • Answer
  • November 10, 2025

Hi ​@dude ,

 

The compaction process requires that no SIDB2 process is running on the Media Agent before you start compaction.


The SIDB2 process handles all DDB partitions on the Media Agent, and running compaction while SIDB2 is active (even for a different partition) can cause conflicts or corruption.

It is recommended to place the Media Agent in Maintenance Mode. This will suspend all jobs to all DDBs hosted on the Media Agent (both Cloud and On-Prem).

Wait for all SIDB2 processes to stop and run the compaction for the Cloud DDB.

Once compaction is complete, take the Media Agent out of maintenance mode and resume jobs.

But if the DDB disk is getting full and running out of space, may i know whether the DDB disk size is as per the hardware specifications mentioned below.

https://documentation.commvault.com/2023e/commcell-console/hardware_specifications_for_deduplication_mode.html

And ensure data aging and pruning are working. If there are many pending prunable or delete records, space may not be reclaimed efficiently.


Run a Data Retention Forecast Compliance (DRFC) report to identify jobs that should be aged off.


And currently there is no workflow or automated method available in Commvault to streamline DDB compaction across multiple DDBs or partitions.

Regards,

Karthik


dude
Community All Star
Forum|alt.badge.img+20
  • Author
  • Community All Star
  • November 10, 2025

@kkarthikk Thanks.