From: Mauro <mrsanna1 AT gmail DOT com<mailto:mrsanna1 AT gmail DOT com>>
Date: Wednesday, September 25, 2013 8:18 AM
To: Uwe Schuerkamp <uwe.schuerkamp AT nionex DOT net<mailto:uwe.schuerkamp AT
nionex DOT net>>, Mauro <mrsanna1 AT gmail DOT com<mailto:mrsanna1 AT gmail DOT
com>>, bacula-users <bacula-users AT lists.sourceforge DOT
net<mailto:bacula-users AT lists.sourceforge DOT net>>
Subject: Re: [Bacula-users] fatal error on full backup.
On 25 September 2013 13:58, Uwe Schuerkamp <uwe.schuerkamp AT nionex DOT
net<mailto:uwe.schuerkamp AT nionex DOT net>> wrote:
On Wed, Sep 25, 2013 at 01:01:41PM +0200, Mauro wrote:
> Start time: 25-Sep-2013 10:55:01
> End time: 25-Sep-2013 12:55:26
This looks suspiciously close to 7200 seconds which is the default
tcp_keepalive_timeout for most Linux distros.
Have you tried using the bacula dir / fd heartbeat feature? Is there a
firewall (iptables?) involved somehow?
I've set Heartbeat = 30 in the director and yes there is a firewall between dir
and fs, not iptables but a juniper, the firewall of the company.
Have you tried lowering the tcp_keepalive_timeout on both ends?
# echo 300 > /proc/sys/net/ipv4/tcp_keepalive_time
I'll try this.
>>>>>>>>
The heartbeat also needs to be on the client config, too. I have much fewer
timeouts, not completely free of them, though.
Patti Clark
Linux System Administrator
Research and Development Systems Support Oak Ridge National Laboratory
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60133471&iu=/4140/ostg.clktrk
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users
|
