How to reclaim deduplication storage space manually (NetBackup appliances)
|Article:TECH180659|||||Created: 2012-02-03|||||Updated: 2014-09-18|||||Article URL http://www.symantec.com/docs/TECH180659|
How to reclaim deduplication storage space manually in NetBackup 5000 series and NetBackup 5200 series appliances.
For background and overview of the process, see http://www.symantec.com/docs/TECH124914.
To reclaim space manually for a NetBackup Media Server Deduplication Pool or for a PureDisk Storage Pool, see http://www.symantec.com/docs/TECH124914.
NOTE: If you want to reclaim as much space as possible and you are doing a backup, delete it right away and start another backup from the same policy while the deletion is going. Not all space will be reclaimed right away, but it will be reclaimed later.
- Expire all NetBackup images backed up to the PDDO storage pool.
- The PDDO data removal process needs to complete. The process runs automatically every 2 hours. If it needs to be forced sooner, the following command can be run on the metabase nodes:
- Once complete run the MB garbage collection policy from the PureDisk web GUI
- Once the garbage collection completes run the CR queue processing twice. They can both be started at the same time and the second one will start after the first completes.
- Expire all images backed up to the disk pool via NBU
- Check the Dedupe Storage queue folder. You should see lots of tlogs ready for processing
ls -l /disk/queue/
- Run CR queue processing
- Monitor processing progress by checking the storage log
tail -f /disk/log/spoold/storaged.log
- Run Garbage collection
/usr/openv/pdde/pdcr/bin/crcollect -v -m +1,+2
- Check the Dedupe Storage queue folder. If you see some tlogs for processing, repeat steps 3 through 5.
- Check the "used" and "Space available within containers" values in the following command
- If dsstat showed a high value for "Space needs compaction", run compaction processing
/usr/openv/pdde/pdcr/bin/crcontrol --compactstartNote: You will need to run steps 3-5 multiple times until there are 0 tlogs found when running
ls -l /disk/queue/
Article URL http://www.symantec.com/docs/TECH180659