On Tue, Jun 30, 2009 at 8:09 PM, Jon Schewe<jpschewe AT mtu DOT net> wrote:
> I have a long running job that I want to run before backups. It backs up
> the databases on the client. I've added the ClientRunBeforeJob line and
> noticed that I was getting errors like this:
>
> 29-Jun 09:11 jon-dir JobId 4073: Start Backup JobId 4073, 
> Job=mtu.2009-06-29_09.11.03
> 29-Jun 09:11 jon-dir JobId 4073: Using Device "FileStorage"
> 29-Jun 09:11 mtu-fd JobId 4073: ClientRunBeforeJob: run command 
> "/etc/bacula/before-full-backup.sh"
> 29-Jun 11:11 jon-dir JobId 4073: Fatal error: Socket error on 
> ClientRunBeforeJob command: ERR=Connection reset by peer
> 29-Jun 11:11 jon-dir JobId 4073: Fatal error: Client "mtu-fd" RunScript 
> failed.
> 29-Jun 11:11 jon-dir JobId 4073: Error: Bacula jon-dir 2.4.2 (26Jul08): 
> 29-Jun-2009 11:11:37
>
>
> I changed my heartbeat interval for the client in bacula-dir.conf to 15
> seconds and on the client in it's bacula-fd.conf to 5 seconds. After
> restarting all services, I'm still getting the connection reset error.
> Between my bacula server and the client is a Linksys router running
> DDWRT. The client is a Linux server using iptables for the firewall.
>

It could be that the heartbeat interval does not apply until the job
starts. Anyone know for sure?

John

------------------------------------------------------------------------------
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users