Index Cache Cleanup

Hi ​@Aleksandra 

What is the CV version here?

Hi Aleksandra,

Best thing is to find out if the growth is real because of natural data growth or in fact there is a problem. For instance if you have secondary copies which are not able to catch up, then your index on primary will not age out. This is just an example. If you open a support case, they will be able to find out exactly what is going on. Also, bringing to the latest MR version is advisable as this may be a software issue as well.

Hi ​@CV_GK, 

the version is 11.32.125

Regards, 

Aleksandra

Hello ​@DoctorB ,

thank you for your explanation. 

Could you please advice what would be the best way to track the data growth in the last months, as well as to track the Index cache growth in the same time period?

And yes, there is a Primary (disk) and a secondary copy (tape) in the affected SP. The tape copy has 500 days retention. However, it exists since long ago, but the index began to grow noticeably only the last month. All the jobs for the tape copy are successful, there are no stuck or failed jobs.

Best regards, 

Aleksandra

Hi ​@Aleksandra 

We need to check a few things here, 

1. Need to identify whether the Index DB or index logs folder were consuming more space in the Index cache. Run the Index state report to check the status of index DBs. https://documentation.commvault.com/11.40/software/health_report_index_state.html
2. If it was Index DBs, then we can try adding the setting to reduce the frequency of index compaction. 
3. If it was Index logs, we would need to check if the Index backups are running properly for the Index server client. 
4. Need to check if Index DB/ Index logs of a particular client is consuming more space. Identify whether it was a backupset/subclient level index? 
5. If the number of objects backed up is high, then validate the hardware requirements of the Index cache in the MA server. 
   
   https://documentation.commvault.com/v11/commcell-console/system_and_hardware_requirements_indexing_version_2.html
   
   I would suggest to raise a case with Commvault support to investigate this further. 

Hello ​@Kalaivanan ,

thanks a lot for your reply. I’ve checked the Index Cache Location Report and the most of the space is consumed by the database (184Gb), whereas Logs consume just a little (2,7Gb). 

Index state report shows the status good for the Index on this MA.

Index Backups are all fine, run regularly according to the schedule and are all in status “Completed”.

It is backupset level Index. 

I’ve opened a case by Commvault, just as you suggested, and posted there all the information found so far. 

Best regards,

Aleksandra