Backup exec 12.5 Jobs recovering on CAS server to up to 13 Managed Media servers

[Justin_Brooks]

Ok

Backup Exec 12.5

1 CAS server, 14 MMS. when a policy kicks off differential jobs in the morning, to disk (B2D) the first few to each backup folder run ok ( Backup To Disk Folder allowing 1 concurrent job only), then the rest QUEUED up, and go into recover mode. 

To repair, i must reboot the Managened mediaserver that i s "recovering jobs" which can be any of the 14, sometimes between 5-7 at a time, while attempting to cancel the jobs, from CAS. If that doesnt work, rebooting the CAS server ( which the job engine service takes between 10-20 minutes to fully start) gets things back online, then i must restart all the jobs that i cancelled or recovered....and it works

Every night before backups, i reboot all my MMS servers, after deleting the on-disk backups. 

I believe the issue has to do with the fact when a job completes, its not fully "releasing" the b2d folder, to allow another job to run to it, hence the "recovery" The "recovered" jobs never appear locally on the mediaserver either...


This issue appears to have started randomly last Tuesday. I had a Media server go offline, and i couldnt get communication (under the Mediaserver tab) on the CAS to come back online. I ended up, uninstalling Backup Exec, and re-installing. However, i dont think i made any chances to MediaMaster00 whatsoever....

Need help!

[CraigV]

Hi Justin,

First off I'd suggest loading up SP2 as a minimum. It's pretty stable in a CASO environment, and we run it quite successfully.
Secondly...what is you configuration? Where do you store catalogs and media set/device information? Locally on each server, or distributed for example.
What I've done from my side is to name the media sets according to their site...for example: Cape Town Daily Media Set; Port Elizabeth Daily media Set etc.
CASO would pump every server's media information into the standard media sets. If you can give a bit more information on how you have set it up, we can go from there.

Laters!

[Justin_Brooks]

 
We are on SP2, with some hotfixes

Catalogs and Device Media information is on CASO, (centralized)

CASO runs and holds everything, the MMS servers are basically Slaves, directed by Policys, media sets guided by the CASO.

I recreated all the policies that had issues. im 99.9% something got corrupted, in regards to the communication to the devices....just dont know what

[CraigV]

OK...then I'd suggest running a database repair on the CASO server first.

[CraigV]

Hi Justin,

Did you run the DB repair and has it helped at all?

[CraigV]

Hi Justin,

Did you run the DB repair and has it helped at all?

[Justin_Brooks]

no dice. I run a script on each mediaserver to synce the time with the master...it was working great, till Saturday...so that means there is a comm issue. Here is another issue. All my automatic Duplicate to tape jobs...50% arent even running at all. THe B2D job will complete, but the B2tape job wont kick off...it just sits in scheduled..and if i do RUN NOW, it fails, and doesnt even reschedule itself for the following week ( i do duplicates for Full system backups on Sundays) I think the DB is all messed up 

[CraigV]

Mmm...1 of 2 solutions left really:

1. See if you can run the whole host of DB utilities via BEutility.exe...
2. Trash the DB, lose all your jobs, and recreate. MOre drastic that the above, but I have personally seen that resolve some isues.