On Thursday 29 November 2012 08:27:16 Dan Langille wrote:
> On 2012-11-29 08:18, Silver Salonen wrote:
> > On Thursday 29 November 2012 07:58:23 Dan Langille wrote:
> >> On 2012-11-29 07:02, Silver Salonen wrote:
> >> > Hi.
> >> >
> >> > I'm backing up some servers with multiple Bacula directors.
> >> >
> >> > Although I've set "Maximum Concurrent Jobs = 10" in FD's
> >> > configuration,
> >>
> >> You also have to look at the SD and Dir 'maximum concurrent jobs'
> >> settings as well.
> >
> > I have this kind of setup:
> >
> > sd1<--dir1 -------> client1 <------- dir2-->sd2
> >
> > At 23:05 dir1 contacts client1 and starts some job, backing up to
> > sd1.
> > At 23:05 dir2 contacts client1 and starts some job, backing up to
> > sd2.
> >
> > As both directors and storages are separate, SD and DIR 'maximum
> > concurrent jobs' settings don't matter here.
> > Anyhow, this setting on all these components is set to 99 :)
>
> OK, so both jobs are running on the FD. Concurrently.
>
> Thus, the problem, must be the SD. Let's look below..
I dug a bit deeper, made another test and found out that actually dir2 starts
the job a bit later than dir1, but this is a minor difference. What's important
is:
1) Both jobs start concurrently on separate directors
2) FD shows both jobs as "running" also
3) Some kind of queue seems to be in effect on FD side:
- log of the job that ended first shows:
28-Nov 23:05 dir1 JobId 83346: Start Backup JobId 83346 ...
28-Nov 23:08 dir1 JobId 83346: Pruning oldest volume ...
28-Nov 23:09 client1 JobId 83346: shell command: run ClientRunBeforeJob
...
28-Nov 23:09 dir1 JobId 83346: Sending Accurate information.
28-Nov 23:09 dir1 JobId 83346: Pruning oldest volume ...
28-Nov 23:10 sd1 JobId 83346: Job client1.2012-11-28_23.05.01_54 is
waiting. Cannot find any appendable volumes.
29-Nov 10:45 dir1 JobId 83346: Recycled volume ...
29-Nov 10:46 sd1 JobId 83346: Recycled volume ...
29-Nov 10:46 client1 JobId 83346: Generate VSS snapshots. Driver="VSS
Vista", Drive(s)="EC"
29-Nov 13:15 sd1 JobId 83346: Job write elapsed time = 02:29:14,
Transfer rate = 164.4 K Bytes/second
- log of the job that ended later shows:
28-Nov 23:55 dir2 JobId 11616: Start Backup JobId 11616
28-Nov 23:55 dir2 JobId 11616: All records pruned from Volume ...
28-Nov 23:55 dir2 JobId 11616: Recycled volume ...
28-Nov 23:55 dir2 JobId 11616: Sending Accurate information.
29-Nov 13:15 sd2 JobId 11616: Recycled volume ...
29-Nov 13:15 client1 JobId 11616: Generate VSS snapshots. Driver="VSS
Vista", Drive(s)="EC"
29-Nov 13:24 sd2 JobId 11616: Job write elapsed time = 00:09:24,
Transfer rate = 2.611 M Bytes/second
Any ideas what could be happening on FD side for the 2nd job while the 1st one
is running?
--
Silver------------------------------------------------------------------------------
Keep yourself connected to Go Parallel:
VERIFY Test and improve your parallel project with help from experts
and peers. http://goparallel.sourceforge.net
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
