cancel
Showing results for 
Search instead for 
Did you mean: 

Media not going back into scratch pool

h_m
Level 6
Media is not going back into scratch pool after the images have expired on the media.

The RETURN_UNASSIGNED_MEDIA_TO_SCRATCH_POOL = NO

is not set in vm.conf. The platform is Windows and changes have been made to bar code rules.

Has anyone come across this issue??
12 REPLIES 12

TempoVisitor
Level 4
A media is unassigned when :
- all its images have expired
- you have reached midnight

So, the media will go back to SCRATCH pool only after midnight.

Try to manualy expire images on a tape, with deassign :
- bpexpdate -m MEDIAID -d 0
The media should go back to SCRATCH

Or you can manualy lauch the commande NBU passes at midnight :
- bpexpdate -deassignempty
Don't worry with this command, it will never deassign media with data ... it just deassign media without anymore active images.

Kerkael

Stumpr2
Level 6
I haven't heard of problems but I suggest checking the configuration.
1. NB4.5_FP3 or greater, this doesn't work with the MP's
2. scratch pool definition in the vm.conf is exactly the same as the scratch pool
3. scratch pool attributes set to anyhost, check mark scratch pool

BTW
Expired tapes that are outside of the library (vaulted) are returned to the scratch pool and that is one reason why some administrators do not use this feature. They want only tapes that are actually available for use in the library to be in the scratch pool.

Jeffrey_Redingt
Level 5
Keep in mind that tapes will only return to the scratch pool if they started there. If a tape was originally in the NetBackup pool and then moved to scratch and then used in a new pool it will stay in the new pool that it was assigned too available to be used. It will not return to scratch unless it was created in the scratch pool.

Richard_Bannist
Level 5
busted !..

I'm one of those Admins Bob mentioned above. I don't use the 'scratch feature' as tapes external to the robot end up in scratch pool as well as physically present tapes. That to me is not a scratch pool, so i have always scripted the tape movements and switched off the scratch feature.

each to their own, but i have done it like this since 3.1.1
Rich

TempoVisitor
Level 4
> Keep in mind that tapes will only return to the
> scratch pool if they started there. If a tape was
> originally in the NetBackup pool and then moved to
> scratch and then used in a new pool it will stay in
> the new pool that it was assigned too available to be
> used. It will not return to scratch unless it was
> created in the scratch pool.

Uh ????
NBU doesnt' care where the Volume was created first.
When it's deassigned, it goes to Scratch, wherever it comes from.

Stumpr2
Level 6
kerkael,
Jeffery is correct.
Bob

zippy
Level 6
Can I get points for just replying?

JD

h_m
Level 6
;)

zippy
Level 6
I think Jeff is correct, it appears to work like he says.

JD

Stumpr2
Level 6
kerkael,
Jeffery is correct.
Bob


you doubt me??

h_m
Level 6
Jeff,

I apologise for only rewarding a helpful award, when it should have been correct.

Mathews
Level 4
Jeffrey,

Usually new blank tapes injected inside the LIB will go to Scratch pool ..then once drive pickup that media and write ..Daily,weekly or monthly backups and put those media to the corresponding Volume Pool , ..If in case if any media going to any volume pool after inventoring means we need to specify the which media ID should go to which volume pool from (Netbackup GUI>> Robot inventory >> Update Volume configuration >> Advanced option)...
In same case if any media from other Lib injected inside in another Lib....it dont have Catalog database...where it will go ?.....For this i saw in Advance option, that ....Media 5XXXX should go to Scratch, Media CXXXX should go to None pool , Media 4XXXX should go to Netbackup pool ..like this....Now i will be happy because in absence of me...onsite engineer injected that media..some other colligues do inventory ..i am sure it will go to Netbackup pool..instead of Scratch .when my shift start i will start the further import process..:
If i am wrong ....pls excuse and correct me