Tape job priority v12

[JaySt]

i'm running 10 tape jobs, all to the same set of two LTO drives. All having different source jobs to process.
I wanted to make sure 2 of those 10 jobs start first, the rest can follow. I'm doing this by starting them earlier compared to the rest. So i start those on 8:00amd and 8:05am, the rest starts at 9:00am.
When the job scheduled to start at 8:00am starts, it runs and finishes in the afternoon. However, i dont see the job scheduled for 8:05 start after that. Instead, one of the jobs scheduled at 9:00am starts before the one schedule at 08:05am.

i dont want to link jobs to run after one another. it should run without linking.

what to expect from the order of starting jobs in this case?

[Dima P.]

Hello JaySt,

Can you please clarify if that's what tape job type you are referring to (backup to tape / file to tape or backup to tape pointed to the GFS media pool)? What are the source objects in the tape jobs? For backup to tape is there any chance you've enabled 'If some linked backup jobs are still running, wait for up to X' option? Thank you!

[JaySt]

i'm doing backup to tape. It is sending only the full backup created weekly to a standard (non-GFS) pool. Source are VMware backupjobs for all tape jobs.
I checked the option 'If some linked backup jobs are still running, wait for up to X' . It seems not like something you can enable/disable. I'ts just set at 3 hours by default. I don't think it's used, because we don't link the jobs. We just start them at different times.

[Dima P.]

Hello Jay,

We've discussed the mentioned behavior with RnD folks and agreed that it looks unexpected. Can you please raise a support case and ask the team to check you logs? Also please share the case id with us. Thank you!

[JaySt]

thanks for check with RnD.
i will! We'll just need to wait how this week's run is going. We're now seeing the following behavior as well in that run:
Tape job 1 is started, claims the two drives available and at some point (near it's completion) releases a single drive for processing of next job (and its tasks). Tape job 2 (scheduled to start a bit later than Tape job 1) was waiting and sees a free drive it can use, so it starts to run. However, after Tape Job 1 releases the second drive, Tape Job 3 starts using it, even though tape job 2 is already running and using the other drive and could use a drive extra for faster processing. Tape job 3 is scheduled to start a bit later than Tape job 2.
So not sure what's happening/should happen there either.
Ah well. i guess it's support time later this week

[Dima P.]

Thank you! The only guess we have is that tape job await for the latest restore points being still locked by the backup job, but to be sure we need logs

[bct44]

Hello,

May i ask why you cannot put all from source jobs in the same backup to tape jobs and prioritize the list of source jobs? If you put your priority jobs on the top of the list, he will prioritized and process it first.

"The sources of the backup to tape job are processed in order the displayed here. To move a source up or down in the list, use the Up and Down buttons on the right."
https://helpcenter.veeam.com/docs/backu ... ml?ver=120

[JaySt]

we've split the jobs because of more granular control, reporting and less chance of source jobs kicking off and interfering with the tape job. The tape job runs for almost 2 days, sometimes more, when put into single job. we've had bad experience with the consequences of the tape job being interupted by the source jobs and try to "resume" from that point.

[bct44]

I can understand , you have some tape jobs interrupted by the source. If you are in V12, please check this topic and the registry key suggested by Mildur:
tape-f29/tapes-fail-when-primary-backup ... 89466.html