Several hourly backups missing after manual submission

[X2]

I have a policy (Standard) to backup some files (exported MySQL logs) hourly. The Full schedule runs at midnight (window 00:00-02:00) and DIFF (hourly) runs every hours from 0200-midnight.

I have noticed that if the DIFF backup fails and is resubmitted by our operations team manually, it does not run hourly for next several hours.Last time it took 10 hours to start running the hourly backups. Last time we had a few failures (58 or 25) was in December and I had noticed the same thing.

Can someone shed some light on this "hiccup" of the scheduler?

Master/Media are RHEL 6 x64 with NetBackup 7.7.3. Client is RHEL 5 x64.

[Marianne · Accepted solution]

It is the initial failure and the global x tries every xx hours (default is 2 tries every 12 hours) causing the issue. When a backup fails, the scheduler all of a sudden reckons that it does not need to submit another backup for 12 hours.

I have seen this happen since NBU 5.x days but nobody seems to log a Support call or else the issue never gets escalated to Engineering. At some point in time a manual backup kicked the hourly schedule back into place, but it seems to no longer work in recent versions.

The only solution is to change the global retry setting to 1 try per 1 hour.

[X2]

Thanks Marianne for the information.

Are you referring to the Global Attributes (highlighted in the attached)?

[Marianne]

Correct.

[X2]

Thanks.

I'm trying to reproduce it in my test environment and if everything is OK, will make change in production and mark earlier post as solution.

[X2]

Update: I recreated the setup and was able to reproduce the issue in testing environment. After changing the Global Attribute for attempting backup "1 tries every 1hours", the backup was not skipped.

Thanks for the pointer @Marianne.