Hi ,
Recently I found issue on my Netbackup cataloge backup policy , it takes too long to run and finally makes no progress .
here are the related settings if you want to know :
max job per client is : 10
1 tape library with 15 drives ( so I have 15 STUs )
all "policies storage" option are set to " any available " . ( I have 20 policies )
Please look at the suggestions in Anshu_Pathak's post.
Other than that, time to log a support call with Veritas.
Firstly - a catalog backup cannot be multiplexed or multi-streamed.
How big is the images folder on your master server?
Are you backing up to tape directly attached to the master?
How many jobs do you see for the catalog backup?
There should be 4 -
Parent job plus 3 child jobs.
1st child job is the staging of the NBDB to a staging area (should go relative quick)
2nd child job backs up the staging area (also relatively quick)
3rd job backs up the images folder and other config files. (This job could take a long time, depending on size of images folder, read speed from disk and path-to-tape.)
How many jobs do you see?
Can you please locate all of them?
Copy the text in Details tab and post here (A lot easier if you post text, rather than screenshots).
Oh! One more thing:
1 tape library with 15 drives ( so I have 15 STUs )
If all drives are in the same robot, same type/density, attached to same media server, then you should have 1 (ONE) storage unit with number of drives set to 15.
But 15 tape drives attached to a single server are way too many.
I have not seen any media server that was capable of writing to more than 4 tape drives simultaneously at a constant speed of +- 100Mbytes/sec.
With 15 tape drives, you should ideally have 4 or 5 media servers.
How many jobs do you see? 4 as you said one parent and 3 child jobs
Can you please locate all of them? yes
Feb 6, 2019 9:37:07 AM - Info nbjm (pid=4256) starting backup job (jobid=3456) for client co-itsrv10.int, policy CatalogBK, schedule Full
Feb 6, 2019 9:37:07 AM - Info nbjm (pid=4256) requesting CATALOG_BACKUP_RESOURCE resources from RB for backup job (jobid=3456, request id:{00A951DB-56AE-45C4-802A-DD17600CB09A})
Feb 6, 2019 9:37:07 AM - requesting resource co-itsrv10.int.NBU_CATALOG.MAXJOBS
Feb 6, 2019 9:37:07 AM - granted resource co-itsrv10.int.NBU_CATALOG.MAXJOBS
Feb 6, 2019 9:37:07 AM - estimated 0 kbytes needed
Feb 6, 2019 9:37:07 AM - begin Parent Job
Feb 6, 2019 9:37:07 AM - begin Catalog Backup: Start Notify Script
Feb 6, 2019 9:37:07 AM - Info RUNCMD (pid=6548) started
Feb 6, 2019 9:37:07 AM - Info RUNCMD (pid=6548) exiting with status: 0
Operation Status: 0
Feb 6, 2019 9:37:07 AM - end Catalog Backup: Start Notify Script; elapsed time 0:00:00
Feb 6, 2019 9:37:07 AM - begin Catalog Backup: Database Manager Query
Operation Status: 0
Feb 6, 2019 9:44:06 AM - end Catalog Backup: Database Manager Query; elapsed time 0:06:59
Feb 6, 2019 9:44:06 AM - begin Catalog Backup: Validate Image
Operation Status: 0
Feb 6, 2019 9:44:06 AM - end Catalog Backup: Validate Image; elapsed time 0:00:00
Feb 6, 2019 9:44:06 AM - begin Catalog Backup: End Notify Script
Feb 6, 2019 9:44:06 AM - Info RUNCMD (pid=7812) started
Feb 6, 2019 9:44:07 AM - Info RUNCMD (pid=7812) exiting with status: 0
-----------------
I have one HP EML with one robot and 15 drives .
please look at the two attached pictures.
Okay - that looks like the parent job.
Can you please post the child jobs as well?
(Will deal later with STU config.)
Additional things that happen in catalog backup is DR file + drpackage creation -> writting it to NFS share or locally on another partition and optionally sending an email with these files as attachment. This can also cause delay or hung catalog backup jobs if there is an issue with filesystem, network filesystem access or email client.
