Why suddenly increased /opt/openv/db/image frowth is increasing after 7.5 upgradation
Created: 06 Jan 2013 | Updated: 19 Jan 2013 | 21 comments
This issue has been solved. See solution.
Hi Folks,
After we have upgraded 7.5.0.1 suddly database image growth is increaed and we have increase file system size 100GB last month but within one month again it reached 95%...Please let us know any other modifcation required on 7.5.0.1 datagrowth reduction..
Thanks,
Susi.S
Discussion Filed Under:
Comments 21 Comments • Jump to latest comment
The upgrade should not cause the space to be filled up in the images folder.
The db/images directory has a list of directories for clients and if you have recently added new clients or changed the retention of your policies this might be causing the space to be filled up. Have you changed the policy to be full backups every day with long retention periods?
You should have considered the space requirement while doing the upgrade.If your environment is very big you might need to do some analysis. You can either increase the space of your disk again after doing a good analysis or check the client in the images folder and find out which client is taking the space.
Catalog Compression
http://www.symantec.com/business/support/index?pag...
Catalog Archiving
http://www.symantec.com/business/support/index?pag...
Also you can read NBU Admin Guide for Windows, Volume I catalog sizing
If you can't increase the space then below is the link to move the images to different drive
https://www-secure.symantec.com/connect/forums/tec...
NBU 7.5 actually stores a bit less in images folder because the header files are moved into EMM database.
The biggest space consumers remain the '.f'' files - same as previous versions.
Factors that play a role (extract from NBU 7.5 Admin Guide I):
lots of files, lots of backups (especially full) with long retention periods.
Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links
Hi Marianne,
We found some of client incremental backup size is more than 300GB for daily basis.Those client images utilizing much more size...Is the any way to compress image size of those clients alone ???
Adding to Mariannes excellent post, you could have some files left in the tmp directories eg:
/usr/openv/netbackup/db/images/<client>/<ctime>/tmp
Files go in there during backup, and get moved up a level if it completes successfully, and should be deleted if it fails. If the system has an unclean shoutdown, files could be left behind, but, I wouldn't expect this to be a regular occurance, so it not likely to be the issue - however, worth checking.
As Marianne said, NBU 7.5.x does noy use more catalog space, in fact it uses slightly less - I'm not aware of any known issues with it, so sussect the most likely cause is one of those listed by Marianne.
Martin
Hi Martin,
under /tmp is not utlizing much more size..I could see /usr/openv/netbackup/db/images/<client>/<ctime>/catstore utlizing much more size...Is there any way to compress it?
Can you tell us the Operating System you are using and if you have identified which area is suing up the disk space?
Is it the /images/ directory or somewhere else (tmp / temp etc.)?
Thanks
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Its solaris and windows client...under /image/client only utilizing much more size....
We found some of client incremental backup size is more than 300GB for daily basis.Those client images utilizing much more size...Is the any way to compress image size of those clients alone ???
That is normal the more the data the image would be bigger, we have incrementals more than 500 GB in our environment. You can do the archiving or compression on complete catalog as per my first post but not for some clients. Please consider again your retention If you have longer retention then this will happen. You have to plan accordingly as per you environment
Actually we have expired most of the incremental backup of the particular client but still its not removing from /opt/openv/netbackup/db/images/client/time-stamp..Please let us know how we will remove those files under /opt/openv/netbackup/db/images/client/time-stamp..
If they are on tape then use bpexpdate -deassignempty -force
If on disk use nbdelete -allvolumes
Followup using bpimage -cleanup -allclients
If you have already expired them then the last command above should do the trick for you
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Hi Mark,
We have catalog image on the disk only...I have did below steps to get free space..
1)Expired entire november month on one client(this client is utilized much space)
2)Run bpimage cleanup to cleanup images..
Please let us know nbdelete -allvolumes will delete only expired images???
Yes - it is perfectly safe to run - from the guide:
Queries the image list in the EMMdatabase to obtain the list of volumes with
deleted fragments. It removes the fragments from those volumes and deletes
eligible imported snaps, unimported snaps, and tar images in that order.
-allvolumes calls bpdm on master server to delete imported snaps. It also
queries the storage servers with unimported snapshots but does not direct
bpdm to delete them.
This clears down the disk itself - the bpimage command cleans the catalog
Does it run with 100% success?
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Yes..bpimage cleanup completed 100% success...Please find the below for details..
It does seem extreme to be deleting backup images due to lack of space on the NBU server
Would it not be better to add disk space rather than deleting your backups?
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Actually problem is we have taken cold data base backup for particular server and removed exclude list for all database file systems,then we forgot to add in the exclude list after cold backup completes...Because of that reason its huge file system having that database server and image size increased due this issue.,,
Incrementals running as Full? Please provide backup policy details
bppllist -L <policy_name>
will restore -- where there is a Will there is a way
We have schedueld correct one but its database file system we forgot to add on exclude_list after cold backup completes...
Partial output? Incremental schedule has to be in same policy as the Full schedule.
will restore -- where there is a Will there is a way
Problem with expiring entire images is that you expire valid, necessary data along with unecessary backup data just because you don't have enough space on the master server??
You can compress images for this client only.
Extract from NBU Commands manual:
bpimage -[de]compress [-allclients | -client <name>]
Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links
Yoh could also move and link the data for that client elsewhere - havent checked that out in 7.5 yet but worth checking as you could just move that clients images folder to a drive with more space
Authorised Symantec Consultant
Don't forget to give a "Thumbs Up" or mark as "Solution" if someones advice has helped you.
Would you like to reply?
Login or Register to post your comment.