In my runcopy.sh script I put a 'sleep 30' delay in at the beginning to
give the backup job a chance to completely finish. This is a somewhat
arbitrary choice and may not be enough for long jobs. Is there a way to
have the script check for completion prior to running the copy job or is
that not necessary?

I set the priorities of both backup and copy jobs = 10 as you suggested.

-Chris-

On Thu, 14 Nov 2024, 18:54 Bill Arlofski, <w...@protonmail.com> wrote:

> On 11/14/24 9:28 AM, Chris Wilkinson wrote:
>  >
> > That's great, thanks Bill. It seems I had the correct answer but didn't
> know how to use it correctly. The level thing threw
> > me off completely.
>
> Ahh!  Good to hear that my "book" about copy jobs was helpful. I have a
> habit of writing long-winded answers.😆
>
>
> > My plan is to run the copy job immediately after the backup in a run
> script
> >
> > runcopy.sh %n %p
> >
> > with some text substitutions within to change <job>-backup to <job>-copy
>
> Make sure that your Copy Control job has the same priority as the Backup
> job(s) that call it, or you will end up in a chicken
> and egg situation where the Backup Job is waiting to the Copy control job
> to finish, but the Copy control job is waiting for
> the Backup job to finish before it starts. :)
>
> Also, have a look at `MaximumConcurrentJobs` in the Copy Control job, and
> the Backup jobs being copied.
>
> When a Copy control job triggers a copy, two jobs are started: Another
> Copy control job to manage the Copying of one backup
> job, and a "backup" job representing the new copy job. So, you will need
> to increase the MaximumConcurrentJobs setting for each.
>
> Also, there is a limiting setting in a Copy control job called
> `MaximumSpawnedJobs`.
>
> If your Copy control job, using `PoolUncopiedJobs` finds 1,000 backup jobs
> in a pool that have not been copied yet, by
> default it will spawn 999 new copy control jobs, and take that last jobid
> and Copy it itself.
>
> You might not want this to be unlimited, especially when you first start
> using Copy jobs on Pools that have existed for a
> while and may have a lot of jobs in them. :)    Later, once everything is
> "all caught up", this setting may be left alone, or
> removed since there should only ever be a few Backup jobs each day that
> need to be copied.
>
>
> One other note: Since you are planning to trigger the Copy control job
> from the Backup job(s) when they are finished, you can
> specify the Backup jobid on the command line that calls your script using
> `%i`.
>
> Then, in your `runcopy.sh` script when you issue the `run job=CopyJobName`
> command, you can add `jobid=xxxx` and the
> `PoolUncopiedJobs` setting is ignored and Bacula will only copy this one
> specific JobId.
>
> Hope this helps!
>
>
> > Thanks
>
> Welcome!
>
>
> Best regards,
> Bill
>
> --
> Bill Arlofski
> w...@protonmail.com
>
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Reply via email to