We are on Networker 184.108.40.206. I have a issue with some savesets not being removed from EMC data domain. Savesets that have a retention time of 10 months back (xx/09/2017) are still present on the data domain device. The flag on these savets is set to "cb".
If I run the following command the "Clone retention time" does not change on these savesets:-
nsrmm -e "08/07/2018" -y -S xxxxxxxxxx
But if I run following command then the saveset is removed from the data domain volume:-
nsrmm -d -y -S xxxxxxxxxx
But from the command refference guide I understand that this only removes refferences to this saveset and trims the networker database.
But it seems like that it does not remove the data from data domain. Is my understanding correct?
And if so then how can I remove these savesets from data domain to free up space?
"nsrmm -e <date> -S #####" might not do anything. However it depends an the status the save set currently has.
If you want to set a save set to "recyclable" you better use
nsrmm -o recyclable -S #####
"nsrmm -d -S #####" will definitively delete the save set (all instances) from the media index and from a disk media.
Of course, it cannot delete the save set instance from tape.
To verify that just run
mminfo -q "ssid=#####" -r "ssid,volume"
Do not forget that to really free space the DD has to run its internal garbage collection which by default is just done once a week.
check for the "Server Protection" group which runs expiration job on NetWorker. What is the status of this job?
Are these savesets residing on a single volume which is no more in mounted state?
Actual data deletion happens on DD through its cleaning cycle till that time it just gets marked as eligible for deletion.
@mc_7, As you mentioned some save sets are having the issue, so I am assuming most of the save sets are getting expired, as the situation will change if no save sets are getting expired.
If no save sets are getting expired then check if the Expiry job is getting successful at all or not?
If the issue is with some of the save sets, then make sure you are considering all the clone copying of the save set, please check using the mminfo / index query so you make sure all clone retention of all the copies is over against the problematic save set is over.
The savesets not expiring automatically can be due to various reasons such as :
1. The chain of incremental and full backup being broken.
2. There is some dependencies on those savsets... You can check for dependencies using command : nsrim - X - D9
3. Nsrim runs automatically on the nw server, sometimes when the backup /activity is running and the server is not free. It will not allow networker server to run. So for that you can create a file of 0kb under /nsr/debug folder with name recover_space_anytime. This will allow nsrim to run even if the server is busy.
4. Its better to run nsrstage - C - V volume_name first.
But im your case the data still showing in cb is an issue... I think u should first bring that data to ce state and then run nsrim - X.
Also observe if daily nsrim is performing its task.
Did you upgrade from older version anytime soon?
Most likely there is some ssid which is depend on it and therefore status in not changed. Check the whole chain for the given saveset. If you really decided to remove it regardless, -d will remove it and nsrim -X (or nsrclone -P -W <volume name> should tag it as removed as well)... then this will be removed from DD upon next file system cleanup (depending when it is scheduled on your system).