cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted

Not all resources are utilized in my SUG

Hi all,

I have a storage unit group that has two storage units (one is LTO5 tape library with two drives, and the other is LTO6 library with two drives).

the duplication to tape jobs write to the SUG, so it should utilize the 4 drives. and this is what was happening before. But now it only writes to one LTO, and the other duplication job will say "

awaiting resource B04-SUG. No drives are available.

 Any ideas of what could be the problem ?

Tags (2)
1 Solution

Accepted Solutions
Highlighted
Accepted Solution!

Re: Not all resources are utilized in my SUG

Hi Marianne and Mph999

We solved the problem, so i thought its worthy to post it here.
It was windows driver problem. we removed all the drives and re-added them again after updating the windows driver and all is working fine. Now jobs are utilizing the 8 drives at the same time.

Thanks a lot for your support.

View solution in original post

8 Replies
Highlighted

Re: Not all resources are utilized in my SUG

Have you checked if all drives are UP?

Or for orphaned/stuck drive or media allocations in nbrbutil -dump output? 

Highlighted

Re: Not all resources are utilized in my SUG

Hi Marianne

All drives are up.
Attached is the dump output

the thing is that its writing to the LTO5 machine for example, then you will find it writing to the LTO6 machine (but not on both at the same time).

Highlighted

Re: Not all resources are utilized in my SUG

How many drives in total?

There are 8 MDS allocations in EMM:

4 drives seem to be in use - or some orphaned allocations?
allocationKey=3285686 mediaId=O02164 driveName=HP.ULTRIUM6-SCSI.002 stuName=ae-eng-bmd-hp2-hcart3-robot-tld-2
allocationKey=3286126 mediaId=O02165 driveName=HP.ULTRIUM6-SCSI.001 stuName=ae-eng-bmd-hp2-hcart3-robot-tld-2
allocationKey=3288493 mediaId=M01830 driveName=HP.ULTRIUM6-SCSI.003 stuName=ae-eng-bmd-hp2-hcart3-robot-tld-1
allocationKey=3288546 mediaId=M01840 driveName=HP.ULTRIUM5-SCSI.000 stuName=ae-eng-bmd-hp2-hcart2-robot-tld-0

and then 4 allocations seemingly doing 'nothing' , probably waiting for resources: 
allocationKey=3288491  mediaId=   driveName=  stuName= 
allocationKey=3288492  mediaId=   driveName=  stuName= 
allocationKey=3288544  mediaId=   driveName=  stuName= 
allocationKey=3288545  mediaId=   driveName=  stuName= 

You can try to release orphaned allocations with:
nbrbutil -releaseMDS <allocationKey> 

If this does not work, you can try to complely clear all resource allocations. This will however kill all active jobs.
nbrbutil -resetAll

Highlighted

Re: Not all resources are utilized in my SUG

Following on from Mariannes expert advice ....

We do, occassionallyt get stuck allocations, sometimes these don't even appear in nbrbutil -dump - which is when nbrbutil -resetall is useful.  At that time, I would also stop service on the master and ensure that there are no files in /usr/openv/netbackup/db/media/drives (on media server) - if there are, delete them as files in here (which will be called <drivename> will stop drives being used.

The files are created by Netackup when a drive is used, but should be deleted when the tape is unmounted.  Occassionally they can be left behind (for example if a machine suddenly restarts) and then cause problems.  So anything not deleted with services stopped can be manually deleted in :

/usr/openv/netbackup/db/media/drives

and whilst you are there, check

/usr/openv/netbackup/db/media/tpreq

Highlighted

Re: Not all resources are utilized in my SUG

Thanks Marianne and Mph999

Unfortunatley the monthly job which does this write is done.
I will try to test it when i have an empty slot this month, and will come back on this.
I did the releaseMDS and resetAll commands, and they were successful.

We have 4 tape libraries, each one has two drives. so we have 8 drives in total.
The job in problem is the monthly one which should write to one STU (Storage Unit Group) that contains 4 drives, 2 from an LTO5 tape library, and 2 from an LTO6 tape library. But as described earlier, it only writes to one of the tape libraries at a time, so it utilizes two drives only at a time. The confusing thing is that while its running, it jumps between the LTO's. Lets say it takes 4 days to finish, today you will find it writing to the LTO6 machine, and tomorrow you will find it writing to the LTO5 machine, but not on both !!!

Anyway, after executing the commands, i will give it a try and come back here.

Thanks,

Highlighted

Re: Not all resources are utilized in my SUG

You may also want to review the individual STU settings and especially the selection criteria in the STUG. 

Explanation over here: https://www.veritas.com/support/en_US/doc/18716246-126559472-0/v41275355-126559472

Highlighted

Re: Not all resources are utilized in my SUG

i did that already.

they are all tape libraries in the SUG, and i choosed "Round Robin"

Highlighted
Accepted Solution!

Re: Not all resources are utilized in my SUG

Hi Marianne and Mph999

We solved the problem, so i thought its worthy to post it here.
It was windows driver problem. we removed all the drives and re-added them again after updating the windows driver and all is working fine. Now jobs are utilizing the 8 drives at the same time.

Thanks a lot for your support.

View solution in original post