Cedric Devillers wrote:
> Arno Lehmann wrote:
>> Hi,
>>
>> 13.11.2007 12:54,, Cedric Devillers wrote::
>>> Hello,
>>>
>>> I have a little problem with one of our bacula installation.
>>>
>>> Let me explain the setup first.
>>>
>>> There is two server, the first has the data and the storage daemon
>>> (meia). The second is the director/DB server (lucita). There is alsa two
>>> client only servers (hr-accentv2 a windows client and darla).
>> Ok. The catalog database is on lucita, right?
>
>
> That's right.
>
>
>>> All the jobs are running fine, except the Catalog Backup. The strange
>>> thing here is that i have nothing in logs about it. If i run it
>>> manually, it is fine.
>>>
>>> Director ans storage version : 1.38.11 (can't upgrade right now).
>> You should plan for that, though :-)
>
>
> It is planned yes, i have to backport the packages :)
>
>
>>> I suppose i have messed something with scheduling or concurency, but i
>>> can't find what.
>> I hope I can...
>>
>> ...
>>> Here is the relevant part of my config :
>>>
>>> ### Jobs definitions :
>>>
>>> JobDefs {
>>> Name = "DefaultJob"
>>> Type = Backup
>>> Level = Incremental
>>> Client = lucita-fd
>>> FileSet = "Full Set"
>>> Schedule = "WeeklyCycle"
>>> Storage = meia-sd
>>> Messages = Standard
>>> Pool = Default
>>> Priority = 10
>>> }
>>>
>>> JobDefs {
>>> Name = "Daily"
>>> Type = Backup
>>> Level = Differential
>>> Client = meia-fd
>>> FileSet = "Full Set"
>>> Schedule = "DailyCycle"
>>> Storage = meia-sd
>>> Messages = Standard
>>> Pool = Default #overwrited by schedule config, but needed to start
>>> bacula
>>> Max Wait Time = 1 hours
>>> Max Start Delay = 4 hours
>>> RunBeforeJob = "etc/bacula/before.sh"
>>> Priority = 10
>>> }
>>>
>>> JobDefs {
>>> Name = "Weekly"
>>> Type = Backup
>>> Level = Full
>>> Client = meia-fd
>>> FileSet = "Full Set"
>>> Schedule = "WeeklyCycle"
>>> Storage = meia-sd
>>> Messages = Standard
>>> Pool = Default #overwrited by schedule config, but needed to
>>> start bacula
>>> Max Wait Time = 1 hours
>>> Max Start Delay = 4 hours
>>> RunBeforeJob = "etc/bacula/before.sh"
>>> Priority = 10
>>> }
>>>
>>>
>>>
>>> Job {
>>> Name = "Daily-meia"
>>> JobDefs = "Daily"
>>> Write Bootstrap = "/var/bacula/incremental.bsr"
>>> }
>>>
>>> Job {
>>> Name = "Weekly-meia"
>>> JobDefs = "Weekly"
>>> Write Bootstrap = "/var/bacula/full.bsr"
>>> }
>>>
>>> Job {
>>> Name = "DARLABackup"
>>> JobDefs = "Weekly"
>>> Client = darla-fd
>>> FileSet="DARLA"
>>> Schedule = "DARLACycle"
>>> Max Wait Time = 1 hours
>>> Max Start Delay = 4 hours
>>> RunBeforeJob = "/etc/bacula/before.sh"
>>> Write Bootstrap = "/var/bacula/darla.bsr"
>>> }
>>>
>>>
>>> Job {
>>> Name = "HRBackup"
>>> Client = "hr-accentv2-fd"
>>> JobDefs = "Daily"
>>> Level = Full
>>> FileSet = "HRSet"
>>> Schedule = "HRSchedule"
>>> Max Wait Time = 1 hours
>>> Max Start Delay = 4 hours
>>> Write Bootstrap = "/var/bacula/hraccent.bsr"
>>> Priority = 11 # run after main backup
>>> }
>>>
>>> #
>>> # Backup the catalog database (after the nightly save)
>>> Job {
>>> Name = "BackupCatalog"
>>> JobDefs = "Weekly"
>>> Level = Full
>>> FileSet="Catalog"
>>> Client = lucita-fd
>> Ok. This is looking right.
>>
>>> Schedule = "WeeklyCycleAfterBackup"
>>> # This creates an ASCII copy of the catalog
>>> RunBeforeJob = "/etc/bacula/scripts/make_catalog_backup bacula bacula
>>> Ya2AhGho"
>>> RunBeforeJob = "/etc/bacula/before.sh"
>>> # This deletes the copy of the catalog
>>> RunAfterJob = "/etc/bacula/scripts/delete_catalog_backup"
>>> RunAfterJob = "/etc/bacula/after.sh"
>>> RunAfterJob = "ssh -i /etc/bacula/Bacula_key [EMAIL PROTECTED]"
>> I *believe* that 1.38 could only handle one Run After Job and Run
>> Before Job option per job. See below how to verify this.
>>
>>> Write Bootstrap = "/var/lib/bacula/BackupCatalog.bsr"
>>> Priority = 11 # run after main backup
>>> }
>> In bconsole, use the command "show jobs=BackupCatalog". Search for the
>> lines with the Run before/after Job commands.
>>
>> If you only see one each, you'll have to put the commands you need to
>> execute into one script, and then reference that script.
>
> You make the point here, multiple runbefore and runafter directive are
> not supported on this version.
>
> I hope it is supported on 2.2.5, because all my other setup use this :)
> (i checked of course, it is supported).
>
> I've made the changes and see if the planned backup of tonight run fine.
>
> But one thing i don't understand is why i don't have anything about this
> job in my logs. And also the fact that manually running the job is
> working fine. Of course, the different runbefore and runafter scripts
> where not running, but the job was executed without issuing any errors.
>
> I'm wondering if there is not a problem with my "Max Wait Time" and "Max
> Start Delay" directive. But as far i undrestand thems, they should be good.
>
Ok, the runbefore runafter scripts are working fine now.
But i still have the exact same problem as before. The job is showed as
canceled in bconsole, but there is absolutely nothing in the logs about it.
I have turned trace on and set debug level 200 (maybe a little high ?)
and i'll see if i can catch some informations.
>> By the way: If you posted the real password to the catalog above
>> you'll want to change that soon :-)
>
>
> I've noticed that right after i compulsively clicked the "send" button :)
>
>> Hope that helps,
>
> It is helping a lot, thanks for your time.
>
>> Arno
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
