Hello,
if we add double clients config (IP oraz name) on Bacula 7.2 and we make
reload or just bacula-dir restart then director crash.
I don't know how look like on 7.0.x, but on 5.x works fine.
Before You make some changes, 1st check config :
$ sudo -u bacula /sbin/bacula-dir -t
Regards,
Adam
: Mittwoch, 24. Oktober 2012 09:51
An: ZIT-BB Büttner, Frank
Cc: bacula-users@lists.sourceforge.net
Betreff: Re: [Bacula-users] director crash when an backup starts
There has been a fix applied in 5.2.12 related to a Director crash,
could you please update?
If you want a binary package you can have a look
There has been a fix applied in 5.2.12 related to a Director crash,
could you please update?
If you want a binary package you can have a look here:
http://repos.fedorapeople.org/repos/slaanesh/bacula/README.txt
Regards,
--Simone
On 24 October 2012 09:14, ZIT-BB Büttner, Frank
wrote:
> Hello,
Hello,
I use bacula 5.2.11 on an SL 6 system(RHEL 6 clone).
As far I start an backup job, the director crash.
dbcheck have not find any errors.
Here the debug output:
system-name: next_vol.c:71-1227 find_next_vol_for_append: JobId=1227 PoolId=1,
MediaType=File
system-name: sql_find.c:462-1227 Rtn
No one has any ideas on what would have caused this. Based on the trace
dump it looks like there is a problem with the scheduler. Any pointers
as to what I can look at?
thanks,
---
Jerold Lowry
IT Manager / Software Engi
I really hate when I do that!!!
[?1034h[Thread debugging using libthread_db enabled]
[New Thread 0x7f8362bfd710 (LWP 9002)]
[New Thread 0x7f8363fff710 (LWP 3111)]
[New Thread 0x7f8368c49710 (LWP 3110)]
0x003377a0e91d in nanosleep () from /lib64/libpthread.so.0
$1 = '\000'
$2 = 0x1fe2068 "bac
Hi list,
I came in this morning and found that my director had died last night
after doing two of the backups. The traceback follows at the end.
This is the scenario:
I noticed yesterday that the only two jobs that were scheduled to
be performed last night were a monthly backup and the c
> Ok, I've missed that option. Thank's John.
>
I missed it about 4 years ago and the list helped me with that. Ever
since I always test the config first.
John
--
This SF.net email is sponsored by
Make an app they can't
> The director crash on reload command if syntax is broken. Do you know a
> workarround. ? This is really annoying as all running job failed for a so
> little syntax error ... can't be that way for a professional tool like
> Bacula.
>
Yes. Always test the configuration before you reload it.
bacul
Hi,
The director crash on reload command if syntax is broken. Do you know a
workarround. ? This is really annoying as all running job failed for a so
little syntax error ... can't be that way for a professional tool like
Bacula.
Best,
P.Lavoisier.
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> I don't believe so, but I could be wrong. I don't think there's
> presently any trapping for config problems. In my experience, it will
> die and you can get the error by attempting to restart it.
In 1.38.x you should test the changed conf wit
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I don't believe so, but I could be wrong. I don't think there's
presently any trapping for config problems. In my experience, it will
die and you can get the error by attempting to restart it.
Bill Moran wrote:
> On Wed, 24 May 2006 11:23:07 +0200 "R
On Wed, 24 May 2006 11:23:07 +0200
"Robert Wirth" <[EMAIL PROTECTED]> wrote:
> Hi!
>
> I'm using bacula 1.38.7. Today, when I changed the director's
> configuration a bit --just changed a MaxWaitTime entry in a JobDefs
> resource--, the director daemon terminated after the reload and sent
> th
Hi!
I'm using bacula 1.38.7. Today, when I changed the director's
configuration a bit --just changed a MaxWaitTime entry in a JobDefs
resource--, the director daemon terminated after the reload and sent
the email attached here.
The email has a subject which I don't understand, and an empty bo
Well, my solution (compiling with -O0) seemed to work for a while. I came in
this morning to no bacula reports, and I find a director hung. Tracebacks
for -dir and -sd are below. This is 1.36.3. I don't *really* want to
upgrade at the moment, but I can if need be.
Netstat shows:
tcp
FYI
I had some crashes - and then it is good to be able to find it in the bug
system, so you know how to avoid it again
In my case it was also everytime after a change in dir.conf, but I did not
change schedules, but added client, job and fileset.
Is it maybe a coincidence that the crash happen
16 matches
Mail list logo
