[Bacula-users] RunScript fails

2009-12-11 Thread Sascha Retzki
Good morning,

I've got the following RunScript-directive in my Job definition. What I 
want to do is run a script before the job is started, on the client 
obviously, that dumps the databases into a directory backed up by bacula 
later on.

 RunScript {
 RunsWhen = Before
 FailJobOnError = No
 Command = /usr/local/sbin/db_sql_backup.sh
 }


I've got two problems, first of the Job fails because the script can't 
be run - I expected the FailJobOnError = No directive to make bacula 
inform me about the script not beeing run but still continues backing up 
the other data.
I named it already, the script fails. I tried starting it on the client 
machine and it works fine. I suspect permission problems because the 
directory the script writes into is just writeable by root. Then again 
the FD is running as root too sooo.. yeah dunno.
Yet another question, is there a way to get stderr from the script? 
Should I (or rather, can I?) pipe stderr into some directory that is 
backed up (like in Command = myscript.sh 2 /var/www/myscript.err)?


Either way here's the log output:

11-Dec 02:00 datensenke-dir JobId 35: Fatal error: Bad response to 
RunScript command: wanted 2000 OK RunScript, got 2999 Invalid command

11-Dec 02:00 datensenke-dir JobId 35: Fatal error: Client vash-fd 
RunScript failed.
11-Dec 02:00 datensenke-dir JobId 35: Fatal error: No Job status 
returned from FD.



-- 
__
Sascha Retzki   n...@work Internet Informationssyteme
EMail: sas...@work.de   Wandalenweg 5 - 20097 Hamburg
Tel.: +49 (0)40-23 88 09-338Germany -   http://www.work.de
HR B 61 668 - Amtsgericht Hamburg Geschaeftsfuehrer Jan Diegelmann
__

--
Return on Information:
Google Enterprise Search pays you back
Get the facts.
http://p.sf.net/sfu/google-dev2dev
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


Re: [Bacula-users] RunScript fails

2009-12-11 Thread Sascha Retzki
Sascha Retzki wrote:
 Good morning,
 

ClientRunBeforeJob = /usr/local/sbin/db_sql_backup.sh

this works. Apperently the RunScript {} thing runs on the director?


 I've got the following RunScript-directive in my Job definition. What I 
 want to do is run a script before the job is started, on the client 
 obviously, that dumps the databases into a directory backed up by bacula 
 later on.
 
  RunScript {
  RunsWhen = Before
  FailJobOnError = No
  Command = /usr/local/sbin/db_sql_backup.sh
  }
 
 
 I've got two problems, first of the Job fails because the script can't 
 be run - I expected the FailJobOnError = No directive to make bacula 
 inform me about the script not beeing run but still continues backing up 
 the other data.
 I named it already, the script fails. I tried starting it on the client 
 machine and it works fine. I suspect permission problems because the 
 directory the script writes into is just writeable by root. Then again 
 the FD is running as root too sooo.. yeah dunno.
 Yet another question, is there a way to get stderr from the script? 
 Should I (or rather, can I?) pipe stderr into some directory that is 
 backed up (like in Command = myscript.sh 2 /var/www/myscript.err)?
 
 
 Either way here's the log output:
 
 11-Dec 02:00 datensenke-dir JobId 35: Fatal error: Bad response to 
 RunScript command: wanted 2000 OK RunScript, got 2999 Invalid command
 
 11-Dec 02:00 datensenke-dir JobId 35: Fatal error: Client vash-fd 
 RunScript failed.
 11-Dec 02:00 datensenke-dir JobId 35: Fatal error: No Job status 
 returned from FD.
 
 
 


-- 
__
Sascha Retzki   n...@work Internet Informationssyteme
EMail: sas...@work.de   Wandalenweg 5 - 20097 Hamburg
Tel.: +49 (0)40-23 88 09-338Germany -   http://www.work.de
HR B 61 668 - Amtsgericht Hamburg Geschaeftsfuehrer Jan Diegelmann
__

--
Return on Information:
Google Enterprise Search pays you back
Get the facts.
http://p.sf.net/sfu/google-dev2dev
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


Re: [Bacula-users] RunScript fails

2009-12-11 Thread Sascha Retzki
Thomas Mueller wrote:
 Am Fri, 11 Dec 2009 10:38:08 +0100 schrieb Sascha Retzki:
 
 Sascha Retzki wrote:
 Good morning,
 
 
 ClientRunBeforeJob = /usr/local/sbin/db_sql_backup.sh
 
 this works. Apperently the RunScript {} thing runs on the director?
 
 you need to set RunsOnClient = yes for RunScript resource.
 

hmm tested that, doesn't work for me :-/



 
 - Thomas
 
 
 --
 Return on Information:
 Google Enterprise Search pays you back
 Get the facts.
 http://p.sf.net/sfu/google-dev2dev
 ___
 Bacula-users mailing list
 Bacula-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/bacula-users
 


-- 
__
Sascha Retzki   n...@work Internet Informationssyteme
EMail: sas...@work.de   Wandalenweg 5 - 20097 Hamburg
Tel.: +49 (0)40-23 88 09-338Germany -   http://www.work.de
HR B 61 668 - Amtsgericht Hamburg Geschaeftsfuehrer Jan Diegelmann
__

--
Return on Information:
Google Enterprise Search pays you back
Get the facts.
http://p.sf.net/sfu/google-dev2dev
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


Re: [Bacula-users] Best Practice tip: automatically cancel Jobs

2009-11-26 Thread Sascha Retzki
Arno Lehmann wrote:
 Hi,

 25.11.2009 14:20, Sascha Retzki wrote:
   
 Hello List,

 I've got a best practice question. When a Job is started it changes to 
 Status=Running and the FD tries to connect to the SD. Let's assume this 
 fails due to network outages, or in my case I configured an IP for the 
 SD which is not accessible from the Client.
 

 I suggest to first upgrade... the versions you're running are really 
 old, and espcially regarding the different max ... time settings, much 
 has been improved in between.

   
On an interesting sidenote, after playing arround with the syntax 
(changed it to 1 min everywhere) bacula indeedly automatically 
canceled the job, but after 30 minutes and not 1 minute. As 30 Minutes 
beeing the default I kinda figured the same, maybe I was just running 
into some kind of bug.

So to get this straight I am doing all right, yeah? I am kinda puzzled 
by the syntax as the Documentation is vague on it on this case:
SD Connect Timeout = 1 min
FD Connect Timeout = 1 min

and in the SD config I got:
Client Connect Wait = 1 min



 Furthermore, it's unlikely anybody has a 2.2.8 reference installation 
 available any more :-)

   

Derailing my own thread, what's the best practice for Bacula regarding 
upgrading? Do you guys just follow your Distributions/Packet Managements 
and upgrade the Bacula server? Do you compile from source and read the 
Bacula docs? Do you upgrade at all or do you just upgrade when you need 
new features/bugfixes?

 Cheers,

 Arno

   



-- 
__
Sascha Retzki   n...@work Internet Informationssyteme
EMail: sas...@work.de   Wandalenweg 5 - 20097 Hamburg
Tel.: +49 (0)40-23 88 09-338Germany -   http://www.work.de
HR B 61 668 - Amtsgericht Hamburg Geschaeftsfuehrer Jan Diegelmann
__


--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users


[Bacula-users] Best Practice tip: automatically cancel Jobs

2009-11-25 Thread Sascha Retzki
Hello List,

I've got a best practice question. When a Job is started it changes to 
Status=Running and the FD tries to connect to the SD. Let's assume this 
fails due to network outages, or in my case I configured an IP for the 
SD which is not accessible from the Client.

What happens is the Job sits in Status=Running and no Data is 
transfered. On canceling the Job,  I see things like:
25-Nov 11:25 bacula-dir JobId 27: Start Backup JobId 27, 
Job=bacula-test.2009-11-25_11.25.40
25-Nov 11:25 bacula-dir JobId 27: Created new Volume bacula-test-0017 
in catalog.
25-Nov 11:25 bacula-dir JobId 27: Using Device bacula-test-file
25-Nov 11:34 bacula-test-fd JobId 27: Fatal error: Failed to connect to 
Storage daemon: 192.168.1.xxx:9103
25-Nov 11:31 bacula-test-fd JobId 27: Warning: bsock.c:123 Could not 
connect to Storage daemon on 192.168.1.xxx:9103. ERR=Connection timed out
Retrying ...
25-Nov 11:34 bacula-dir JobId 27: Fatal error: Socket error on Storage 
command: ERR=No data available
25-Nov 11:34 bacula-dir JobId 27: Bacula bacula-dir 2.2.8 (26Jan08): 
25-Nov-2009 11:34:59
  Build OS:   x86_64-pc-linux-gnu debian lenny/sid
  JobId:  27
  Job:bacula-test.2009-11-25_11.25.40
  Backup Level:   Full
  Client: bacula-test-fd 2.4.4 (28Dec08) 
i486-pc-linux-gnu,debian,5.0
  FileSet:bacula-test-fileset 2009-11-25 11:25:10
  Pool:   bacula-test-pool (From Job resource)
  Storage:bacula-test-sd (From Pool resource)
  Scheduled time: 25-Nov-2009 11:24:58
  Start time: 25-Nov-2009 11:25:12
  End time:   25-Nov-2009 11:34:59
  Elapsed time:   9 mins 47 secs
  Priority:   10
  FD Files Written:   0
  SD Files Written:   0
  FD Bytes Written:   0 (0 B)
  SD Bytes Written:   0 (0 B)
  Rate:   0.0 KB/s
  Software Compression:   None
  VSS:no
  Storage Encryption: no
  Volume name(s):
  Volume Session Id:  2
  Volume Session Time:1259081276
  Last Volume Bytes:  0 (0 B)
  Non-fatal FD errors:0
  SD Errors:  0
  FD termination status:
  SD termination status:  Waiting on FD
  Termination:Backup Canceled



I've already set the following timeouts in my bacula-dir.conf:
Director {
...
FD Connect Timeout = 5
SD Connect Timeout = 1
...
}
(I had n postfixed to the numbers because I somewhere read it means 
minutes, so originally I had XD Connect Timeout = 5n, but it doesnt work 
either way)


Okay so I want it to timeout so it triggers a log message like it does 
when I cancel the job. I want the people who're responsible for these 
backup to come to the office in the morning and see their daily backup 
job succeeded or failed. I don't actually care if it cancels or not, at 
least if all the other jobs are succeeding regardlessly ;-)


Soo, give me your best practice advice, or tell me if I just did 
something wrong with teh configuration and I am on the wrong track!


Thanks in advance, Sascha

-- 
__
Sascha Retzki   n...@work Internet Informationssyteme
EMail: sas...@work.de   Wandalenweg 5 - 20097 Hamburg
Tel.: +49 (0)40-23 88 09-338Germany -   http://www.work.de
HR B 61 668 - Amtsgericht Hamburg Geschaeftsfuehrer Jan Diegelmann
__


--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
___
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users