Jobs retry delay & Schedule backup attempts settings

[owen1234]

Hi I am new to NetBackup, and just take over the role from previous admin.

Just reliase that our NetBackup server do not retry failed job.

Looking at the settings at our Master Server properties:

Job retry delay: 60 minutes

Schedule backup attempts: 3 tries in 1 hours

Looks like the "Schedule backup attempts" setting is not correct, am I right? The job only retry after 1 hour, but the attempts only allow within 1 hour.

What if I change the setting to

Job retry delay: 60 minutes (or 30 minutes)

Schedule backup attempts: 3 tries in 1 hours (3 tries in 6 hours/3 tries in 12 hours)

Please help.Thanks a lot.

[Nicolai]

Welcome to the new job as NBU admin Owen.

There is likely a reason for the previous admin setting job re-try delay to 60.  But since the settings can be changed without restart and is for error recovery why not try it out and observe the result. After all some settings are to be set after personal opinion.

That said I think 3 tries in 1 hour is a little too aggressive. Lower values should do it.

Description of global attributes in Netbackup.

http://www.symantec.com/docs/HOWTO34254

[owen1234]

Thanks Nicolai for your comment.

Actually I would like to confirm is the previous settings, which has cause the failed backup to attempt retry? It never retry after the jobs failed.Thanks.

Job retry delay: 60 minutes

Schedule backup attempts: 3 tries in 1 hours

[Marianne · Accepted solution]

The previous settings are not making sense - 3 tries in one hour, with 60 minutes between the retries!
There is no way that this can work!

The default is 2 tries every 12 hours, with 10 minutes between retries.

I feel that every 12 hours is too far apart as most week-night backup windows are less than 12 hours for most customers.

So, I like 2 tries every 3 hours with 10-15 min delay between retries.

Decide on something that will work for you. The delay between tries should be long enough for someone to troubleshoot and fix the problem.
A retry without doing anything about it makes no sense.
A status 59 will just fail again if nobody adds a Server entry on the client for the media server trying to back it up. 
A status 96 will fail again if nobody adds more tapes. Etc... etc... etc...

The interval (every # hours) should be big enough so that not too much load is put on the scheduler (nbpem), the resource broker (nbrb) and the job manager (nbjm), but small enough to ensure that failed jobs are retried within the backup window.

Hope this is making sense!

[owen1234]

The previous settings are not making sense - 3 tries in one hour, with 60 minutes between the retries!
There is no way that this can work!

- Exactly what I think. So now I can confirm the job never retry is due to this wrong settings.

Thank you for your suggestion of the settings, I will made a decision after consider all the policies.

Thank you very much!