On Fri, Jun 19, 2009 at 6:29 PM, Dirk Bartley<bartleyd2 AT chartermi DOT net>
wrote:
> Greetings
>
> I'm trying to understand Maximum Concurrent Jobs.
>
> Here is an example output from status dir
>
>
>
> Running Jobs:
> Console connected at 19-Jun-09 17:51
> JobId Level Name Status
> ======================================================================
> 9334 Full Alum3JobMain.2009-06-19_17.01.00_47 is running
> 9336 Full Alum2RootJob.2009-06-19_17.01.00_49 is waiting on Storage
> Alum2FileStorage
> 9337 Full Alum1Job.2009-06-19_17.01.00_50 is waiting on Storage
> Alum2FileStorage
> 9338 Full Alum3root.2009-06-19_17.01.00_51 is waiting on Storage
> Alum2FileStorage
> 9339 Full Alum3sharedlvm.2009-06-19_17.01.00_52 is waiting on
> Storage Alum2FileStorage
> 9340 Full DatabaseJob.2009-06-19_17.01.00_53 is waiting execution
> 9341 Full SubversionJob.2009-06-19_17.01.00_54 is waiting for
> higher priority jobs to finish
> ====
>
>
> These jobs are all being written to filestorage. Then when they are all
> done the low priority jobs will trigger a run after script that will
> start them all being copied to tape, but that is later.
>
> So it looks like only one job is running because I have a maximum
> concurrent directive relating to storage limiting the number of jobs
> running concurrently.
>
> Reading the storage daemons config documentation, there is only one and
> it is on the storage daemon itself, not on the Device resource. It is
> set to 50.
>
> On this storage daemon I have 2 tape devices and one Filestorage device
> on this storage Daemon.
>
> I can see from the director configuration, there is a maximum concurrent
> jobs directive for just about every resource. The director resource is
> set to 5. All jobs are set to 5 with the exception of the copy jobs.
> All client resources are set to 10. Each storage resource in the
> directors config is set to a Max of 10.
>
> So I'm not understanding why I can't have more jobs running at the same
> time. The pool resource for the File-xxx pool sets Maximum Volume Jobs
> = 1 so that each job will have it's own volume (or file on the hard
> drive) that it is writing to.
>
> Because I am writing to file first, I am not setting up spooling.
>
> Any possible assistance in helping me understand this is apprecieated.
> I guess I could be happy that everything is working, I'ts just that when
> someone asks me why I can't get more jobs running concurrently, I want
> to be able to answer. Because I know one person will ask.
>
Are the priorities the same for all jobs but the SubversionJob?
John
------------------------------------------------------------------------------
Are you an open source citizen? Join us for the Open Source Bridge conference!
Portland, OR, June 17-19. Two days of sessions, one day of unconference: $250.
Need another reason to go? 24-hour hacker lounge. Register today!
http://ad.doubleclick.net/clk;215844324;13503038;v?http://opensourcebridge.org
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users
|
