[Pacemaker] Debbuging / Logging Pacemaker

[Andrew Beekhof]

On Mon, Jan 23, 2012 at 11:13 PM, Stallmann, Andreas
<AStallmann at conet.de> wrote:
> Hi!
>
>
>
> I try to set up MySQL HA with pacemaker according to:
>
>
>
> http://www.mysqlperformanceblog.com/2011/11/29/percona-replication-manager-a-solution-for-mysql-high-availability-with-replication-using-pacemaker/
>
>
>
> When I do that, I find the following error in crm_mon:
>
>
>
> ---
>
> Master/Slave Set: ms_MySQL
>
>      p_mysql:0  (ocf::heartbeat:mysql): Slave int-ipfuie-mgmt01 (unmanaged)
> FAILED
>
>      p_mysql:1  (ocf::heartbeat:mysql): Slave int-ipfuie-mgmt02 (unmanaged)
> FAILED
>
> writer_vip      (ocf::heartbeat:IPaddr2):       Started int-ipfuie-mgmt02
>
>
>
> Failed actions:
>
>     p_mysql:0_monitor_0 (node=int-ipfuie-mgmt01, call=10, rc=1,
> status=complete): unknown error
>
>     p_mysql:0_stop_0 (node=int-ipfuie-mgmt01, call=15, rc=1,
> status=complete): unknown error
>
>     p_mysql:1_monitor_0 (node=int-ipfuie-mgmt02, call=10, rc=1,
> status=complete): unknown error
>
>     p_mysql:1_stop_0 (node=int-ipfuie-mgmt02, call=15, rc=1,
> status=complete): unknown error
>
> ---
>
>
>
> Well, that isn’t the most detailed of all error messages.
>
>
>
> I then did the following grep on /var/log/messages (debugging is already
> turned on in /etc/corosync/corosync.conf):
>
>
>
> grep -B 2 -A 2 -i mysql /var/log/messages | grep -i error | less
>
>
>
> The only message, that is a little more “revealing” says:
>
>
>
> ---
>
> Jan 23 12:50:46 int-ipfuie-mgmt01 crmd: [10070]: info: process_lrm_event:
> LRM operation p_mysql:0_monitor_0 (call=36, rc=1, cib-update=44,
> confirmed=true) unknown error
>
> Jan 23 12:50:47 int-ipfuie-mgmt01 crmd: [10070]: info: process_lrm_event:
> LRM operation p_mysql:0_stop_0 (call=38, rc=1, cib-update=46,
> confirmed=true) unknown error
>
> ---
>
>
>
> Any ideas on how to get a little more information on why the resource
> failed? Where, for example, does the output of the resource scripts go by
> default? I feel they rather go to /dev/null, but definitely not to any
> logfile, am I right?

They're supposed to, but there was a bug:
   http://bugs.clusterlabs.org/show_bug.cgi?id=5024

Have you looked in syslog too?

>
>
>
> If you have any suggestions, where else to look for more obvious error
> messages, please tell me so!
>
>
>
> What other means of debugging my config do I have?
>
>
>
> This all leads me to a more general question:
>
>
>
> Can I redirect pacemakers log messages to any other file than
> /var/log/messages (without configuring filters in syslog-ng)? The logging of
> pacemaker (at least in the version I’m using) does not seem to be influenced
> by setting to_logfile: yes and logfile: /var/log/cluster.log in
> /etc/corosync.conf (which only get’s messages from corosync (TOTEM etc.).
>
>
>
>
>
> Thanks!
>
>
>
> Andreas
>
>
> ------------------------
> CONET Solutions GmbH, Theodor-Heuss-Allee 19, 53773 Hennef.
> Registergericht/Registration Court: Amtsgericht Siegburg (HRB Nr. 9136)
> Geschäftsführer/Managing Directors: Jürgen Zender (Sprecher/Chairman), Anke
> Höfer
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org
>