2011-07-26 07:52 AM
Apparently I do not understand how snapvault works. I am trying to figure out why my secondary snapshots are taking up so much on my primary volume.
We currently use DFM to backup our primary qtrees to secondary volumes once a day. Some have different retention perios, but you get the idea. We have a volume which contains many qtrees (2 dozen?) which is running out of space. Looking at the volume I see the following:
/vol/myvolume is full (using or reserving 98% of space and 4% of inodes, using 98% of reserve).
Volume is 1.5 terabytes with 10% snapshot reserve.
The actual data on the volume is 800GB. That leaves almost 700GB of data unaccounted for. So, I looked at the "snapshots" for that directory and see:
1) Several snapshots called nighty.0, hourly.0 etc. -- These I expect based on the snapshot schedule. They SHOULD be taking up space up to the 10% snapshot reserve specified for that volume.
The next two, I am not sure why they are taking up space.
2) dfm_sv_backup(newsecondaryfiler)secondaryvolume.0 -- if these are "snapvaulted" to secondary, why is the space taken up on the primary? One qtree that is 30GB is taking up over 300GB in this snapshot.
3) oldsecondaryfiler(0118065368)_secondaryvolume_qtreename-src.3 -- We moved our secondary volumes to different storage last week. This required us to recreate the backup relationships.
Why are these last two taking up so much space on the primary? Isn't one of the points of these secondary backup location to not have to keep so much snapshot data on the primary?
Thanks in advanced for any help.