Duplication: continued attempts result in Status Code 96, for Policy that no longer exists.

[wellssh]

We recently moved from a Backup-to-Tape technology onto Backup-To-Disk technology.  We began testing different methods of Duplication and created various Lifecycle Storage Policies for this.  During our testing we removed some policies to start them fresh and found that the Lifecycle Policies associated with these removed/deleted policies continue to try and run.  They usually error out with a Status Code 96:  Unable to allocate new media for backup, storage unit has none available.

Why do these Duplication jobs continue to attempt to run when their associated policies have been removed?  I've rebooted the Master server in an attempt to refresh things, but that was unsuccessful.

Thanks,

Sven

[mph999]

Probably becuse they hadn't duplicated before the changes were made.

If you run :

nbstlutil stlilist -U

Do these images show as 'completed' ?

If not, you can cancel them :

nbstlutil cancel -backupid <backup id>

Regards,

Martin

[wellssh]

I can't cancel it as it has already stopped running with Status Code 96.  I will have to wait and see if it tries to run again.

[wellssh]

.....this until it runs again.

Still find it odd that it attempts to run duplication to a storage device other than what is actually configured in the Lifecycle Policy. ie.  initially the SLP was configured to Duplicate to Tape.  After the Policya was attempted a couple of times, I changed the configuration of the SLP by modifying the Duplication and "telling" it to use Disk instead of tape.

The SLP job still runs and attempts to use tape even though the Storage option in the Policy clearly lists the SLP configured for Disk.

It's like something is stuck in the system.

Thanks,

Sven

[Mark_Solutions · Accepted solution]

You should not edit a SLP whilst images are active on it.

When the images were written their lifecycle was determined and at that time they were destined to duplicate to tape.

To dispoise of these do the following:

1. Cancel all running duplications (if any are still running)
2. Wait a few seconds and then run, on the Master Server, \netbackup\bin\admincmd\nbstlutil inactive -lifecycle lifecyclename against the lifecycle to deactivate it (stops it duplicating but can still be backed up to)
3. Run \netbackup\bin\admincmd\nbstlutil stlilist -image_incomplete –l                    this gives you a list of all images currently in the lifecycle that are not complete (duplicated) – you probably need to pipe this out to a file.
4. You can then cancel the incomplete images for the lifecycle using nbstlutil cancel –lifecycle lifecyclename –destination tapestuname or just the specific images you want to cancel using nbstlutil cancel –lifecycle lifecyclename –destination tapestuname –backupid backupimageid

Obviously replace the parts in italics with what you have

Hope this helps

[Yasuhisa_Ishika]

Flow and residence of each copy become fixed, and any change made to SLP does not take effect to submitted backup images. Only you can do is cancel SLP operations using "nbstlutil cancel". If you gave already tried this, check if the argument and backup was correct.

[mph999]

The job has already been submitted ...

Certain settings of SLPs can be modified, AND these will be picked up ...  for example retention level.

The destination cannot be chaged for a SLP that has started, but not duplicated.

Any new SLPs that start, should however pick up this change.

Martin

[Marianne]

No need to wait until it runs again - just run the commands list in Martin's first reply:

[wellssh]

I ran this command:

nbstlutil -cancel -lifecycle <lifecycle name>

I received back:

"Operation not successful: unable to process request
Partial success, some images processed"

Not sure exactly what this means.

Thanks,

Sven

[Mark_Solutions]

did you want to cancel everything in your lifecycle - is it not currently actively processing your backups?

Look at my earlier post which produces a list of incomplete images, from which you could have just cancelled the ones you needed to

Worth running this now anyway to see what has not been processed

[wellssh]

....I rebooted all Media servers yesterday and I have not had this issue appear since.  Perhaps the old Policy was "stuck" in the system somewhere.

Thanks to everyone who provided valuable input!

Sven