On Wed, Jan 02, 2013 at 10:23:52AM +0000, Gary Stainburn wrote:
>
> Dan,
>
> The main cause of my confusion is that Bacula usually just creates a new
> volume file when the current one is full or marked as (unless a volume has
> already been recycled). This is why I only get the errror elsewhere if there
> is no space left on the other storage servers.
>
> I cannot understand why periodically Bacula stops and instead generates this
> email for just this one server. Admittedly it is the second busiest storage
> on the network.
>
> For the director I'm running 5.2.11 from RPM on a Fedora 17 server. This is
> box is also the main storage server.
>
> The storage server in question is running 5.0.3 from RPM on a Fedora 16
> server.
>
> The definition for Hales is:
>
> Pool {
> Name = Hales
> Pool Type = Backup
> Volume Use Duration = 23h
> Maximum Volume Bytes=5G
> Recycle = yes
> AutoPrune = yes
> Volume Retention = 3 months
> Label Format = "hales"
> }
>
>
Hi all,
I also have a feeling that there is definitely something wonky with
volume management & recycling ever since version 5.2.6. I cannot put
my finger on it, but for busy directors (we have one with 160 clients,
a 200GB mysql catalog, 40TB disk storage and a 24-slot lto5 library)
I've seen strange issues like copy jobs stopping to work for not
finding their storage / volume definitions any more (see previous
posts), bacula-sd just hanging so a "stat storage" won't return in
bconsole (on an otherwise idle machine with on-disk volumes and no
suspicious stuff in dmesg). For the last couple of months I've been
jumping between 5.2.6 and 2.12 via softlink (both compiled from source
on Centos 6), but neither version runs completely satisfactorily.
For some months now we've been having massive issues with win clients
of all variations which we have no problems with on the other
installations, backups break with errors like
deniow2021-fd JobId 50560: Error: /home/kern/bacula/k/bacula/src/\
lib/bsock.c:335 Socket has errors=1 on call to Storage
daemon:deniol186:9103
24-Nov 14:42 deniow2021-fd JobId 50560: Fatal error:
/home/kern/bacula/k/bacula/src/filed/backup.c:1270 Network send error to SD.
ERR=Input/output
error 24-Nov 14:42 deniow2021-fd JobId 50560: VSS Writer (BackupComplete):
"Task Scheduler Writer", State: 0x1 (VSS_WS_STABLE)
That won't go away no matter which win client we're trying.
That being said, I have three other bacula instances in other data
centers that tick along just fine for the most part (most on 5.2.12),
but then again they're nowhere near as big than the one that's
creating problems (and even this one ran fine until we hit a certain
size in both clients and storage attached).
I know this post probably isn't very helpful in solving your specific
problem, but maybe a "me too" type post won't exactly hurt, either.
All the best, Uwe
--
NIONEX --- Ein Unternehmen der Bertelsmann SE & Co. KGaA
------------------------------------------------------------------------------
Master Java SE, Java EE, Eclipse, Spring, Hibernate, JavaScript, jQuery
and much more. Keep your Java skills current with LearnJavaNow -
200+ hours of step-by-step video tutorials by Java experts.
SALE $49.99 this month only -- learn more at:
http://p.sf.net/sfu/learnmore_122612
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users
|
