Forum Discussion

Vamsi1's avatar
Vamsi1
Level 5
9 years ago

NDMP Snapshots not Getting Deleted

Team,

NDMP backups are configured in our enivronment for Netapp Filers.

Backup are running for all the volume except one volume in a NetApp Filer. Issue is 

"Backup is getting intitated, Snapshot is getting created Backup goes hung state for few hours and failing with 99 error code"

Snapshot which was created on the filer was not cleared, Same thing repeated for 4 times & now on the filer we have 4 snapshots which were created and occupired almost 80 GB of disk space.

Tried to delete them manually from filer level but not luck we couldnt able to clear that. 

Hope some one can provide me a solution for this issue.

Lets me if you need any more ifnoirmation to understand the issue.

*****************************************

Netbackup Version 7.5.0.7

OS Solaris 10

Filer Net App

*********************************

Have a look at the attachment for the Backup log

  • Not something I've ever had to do, but on the filer you can disable ndmp, kill all active ndmp threads & then re-enable ndmp.

    This would obviously have to be done when no other (valid) ndmp backups were taking place.

  • I think you should contact Netapp support, netbackup is not involved in creating the snapshots, that is something happening inside the Netapp box.

  • As Nicolai states this is more than likely a NetApp issue, but just as a matter of interest what issue/error are you getting when trying to delete the snapshot from the filer?

    If it says it can't delete as the snapshot is in use (due to backup in progress) then you can check for running backups with the following command on the filer:

    backup status

    this should list all running backups on the filer. If there are 'running' or 'restartable' backups listed that aren't actually running (tied to the snapshot you can't delete) then

    backup terminate <ID>

    could be used to allow you to delete the 'offending' snapshot(s)

    Just for completeness, you can use the

    snap list <volume-name>

    to list all the snapshots on the volume.

     

    It may be that these (or other) 'stale' snapshots from previous backups are preventing the job from completing successfully I'm not sure, but it is something to look into or try.....

  • Thank you for the quick response Andy & Nicolai.

    When deleting the snapshot after terminating the ndmp it is getting the below message:

    Snapshot snapshot_for_backup.43653 is busy because of LUN clone, snapmirror, sync mirror, volume clone, snap restore, dump, CIFS share, volume copy, ndmp, WORM volume,

    When terminating the individual dumps showing:

    Cannot terminate ACTIVE dumps.

    Backup status is:

    ID     State     Type  Device   Start Date   Level  Path

    --  -----------  ----  ------  ------------  -----  ---------------

    0  ACTIVE       NDMP  ndmp    Apr 24 04:01    0    /vol/vol_rman_logs_st01/

    1  ACTIVE       NDMP  ndmp    Apr 24 12:05    0    /vol/vol_rman_logs_st01/

    2  ACTIVE       NDMP  ndmp    May 01 08:20    0    /vol/vol_rman_logs_st01/

    Currenly No Backups are running from Netbackup, But still the snapshots shows as ACTIVE.

     

  • As per Nicolai and Andy's post - this is a NetApp problem, not NetBackup....

  • Not something I've ever had to do, but on the filer you can disable ndmp, kill all active ndmp threads & then re-enable ndmp.

    This would obviously have to be done when no other (valid) ndmp backups were taking place.