Teradata backup with W2003 & NBU 5.1mp4

[Rick_Brown]

Hello Everyone,

We are backing up 8 Teradata nodes between 3 W2003 servers running NBU 5.1 using CA-7 as the scheduler to kick off NBU jobs.

When the first job is kicked off for a teradata backup, the job is completed in 6 seconds typically. It them kicks off a second job that gathers all the data, like one would expect. There's even a third jobs that is run after that -- all for the same backup job.

Question:
Is there a way to keep the first job open during the entire backup job as it does for an Oracle, DB2, or SQL database? Maybe a script that needs to be added to the .bch file or something.
We're looking for a way to limit the lenght of time a job can run (in case it gets stuck), stop the process and then send an unlock on the Teradata database.

Any help is very much appreciated.
Thank you.

[zippy]

Rick,

It sounds to me that this job is working as it should.

I'm not sure what you really want to do.

You say the jobs have been getting stuck, I think you should try to find out why these type of jobs are getting stuck, is there some problem with the hardware on the netbackup side that causing this stuck state?

There are ways to script a timer of some sorts but I think its just a patch to your current problem.

JD

[Rick_Brown]

HI James!
Thanks for the quick response.
Most of the time NBU works with Teradata fine. We have several DB updates that need to run at certain times and if the backup has a lock on the DB they cannot run.
This is why we need to specify a time "window" the backup should have to complete. If it goes past that time, it needs to unlock the DB and move on.
The environment is fairly complex with many different systems involved 24/7.

We have several more jobs we want to migrate over to the W2003 / NBU5.1 systems. Until we can have safegards in place to make sure that if the backups are having a problem, they won't be interupting service levels to the users or worse yet - not getting updated information at all, we cannot move forward. One of our Teradata tables is about 1TB in size & there are many.
If there was a way to keep the first job in the Teradat backup active as in a SQL backup we could trak that time and take the result to if/then/else the status code returned to unlock the DB if needed.

[Yasuhisa_Ishika]

Well, there are no way to realize that let first job keep active.
But with some notify scripts an techniques, we can track stuck jobs.
For detail, see System Administrator's Guide vol II, page 167.

NetBackup 5.1 System Administrator's Guide Volume II
http://support.veritas.com/docs/268091

And, just for canceling stack jobs, get active job IDs with "bpdbjobs -report", and cancel them with "bpdbjobs -cancel job_id1,job_id2" at a time.