[Pacemaker] MySQL not starting

[Yves Trudeau]

Hi Andreas,
   make sure the pid defined in pacemaker is the same has the one define 
in /etc/mysql/my.cnf.  Although it may not be that since you have the 
replication info set.  Have you verified that MySQL is indeed not 
running? maybe you just forgot to cleanup the errors.  The updated 
version is here:

https://github.com/y-trudeau/resource-agents-prm

This will be in resource-agents 3.9.3.  Be aware there are some slight 
changes like for the node IP attribute.  It is documented in the 
meta-data.  Also, if you do:

mkdir /tmp/mysql.ocf.ra.debug/
touch /tmp/mysql.ocf.ra.debug/log

This will starts tracing to the "log" file.  It is very verbose so don't 
forget it... I am writing documentation at the moment.

Do not hesitate if you other questions.

Regards,

Yves

Le 2012-05-16 08:29, Stallmann, Andreas a écrit :
> Hi!
>
> I try to get a mysql master / slave setup running, according to
> http://www.e-rave.nl/prm-mysql-ha-and-pacemaker-in-the-mix. I had this
> setup running already, but now as we are switching from SLES to Ubuntu,
> I have to set it up anew.
>
> Here’s the primitive for mysql:
>
> primitive p_mysql ocf:heartbeat:mysql \
>
> params config="/etc/mysql/my.cnf" log="/var/log/mysql.err"
> additional_parameters="--debug" pid="/var/run/mysqld/mysqld.pid"
> socket="/var/run/mysqld/mysqld.sock" replication_user="repl"
> replication_passwd="***" max_slave_lag="15"
> evict_outdated_slaves="false" binary="/usr/sbin/mysqld" test_user="root"
> test_passwd="***" \
>
> op monitor interval="20s" role="Master" OCF_CHECK_LEVEL="1" \
>
> op monitor interval="30s" role="Slave" timeout="30s" OCF_CHECK_LEVEL="1" \
>
> op start interval="0" timeout="120s" \
>
> op stop interval="0" timeout="120s"
>
> and here’s the master-slave-resource:
>
> ms ms_MySQL p_mysql \
>
> meta master-max="1" master-node-max="1" clone-max="2" clone-node-max="1"
> notify="true" globally-unique="false
>
> Some part of it works, as the replication info is available:
>
> property $id="mysql_replication" \
>
> replication_info="10.10.0.101|mysql-bin.000005|107"
>
> Still, the agent fails, as crm_mon reports:
>
> Failed actions:
>
> p_mysql:0_start_0 (node=proto-cms-appl01, call=22, rc=-2, status=Timed
> Out): unknown exec error
>
> p_mysql:1_start_0 (node=proto-cms-appl02, call=21, rc=-2, status=Timed
> Out): unknown exec error
>
> Starting mysql by hand with service mysql startworks nicely, while
> ocf-tester –v –n p_mysql –o … /usr/lib…/mysql keeps on reporting
>
> mysql[20688]: ERROR: MySQL is not running
>
> and never (!) ends, unless it is killed via CTRL-C.
>
> I find nothing of any (obvious) relevance in /var/log/syslog besides
>
> May 16 13:27:20 proto-cms-appl01 mysql[1300]: ERROR: ERROR 1210 (HY000)
> at line 1: Incorrect arguments to MASTER_HOST
>
> May 16 13:27:20 proto-cms-appl01 mysql[1300]: ERROR: Failed to set master
>
> which is from one hour ago.
>
> mysqld.log, mysql.err, and mysql.log under /var/log/ stay empty, as well
> as the directory /var/log/mysql.
>
> Any ideas?
>
> Cheers,
>
> Andreas
>
> BTW: Does anyone else feel, that “unknown exec error” is not the most
> informative way to report an error? It would be nice if the “verbosity”
> could somehow be increased. What do you think?
>
> PS: Yves, if you happen to read this: Are there any updates due for your
> mysql resource agent? When will it be included into the regular
> pacemaker-resource-agents package?
>
> --
> CONET Solutions GmbH
> Andreas Stallmann,
> Theodor-Heuss-Allee 19, 53773 Hennef
> Tel.: +49 2242 939-677, Fax: +49 2242 939-393
> Mobil: +49 172 2455051
> Internet: http://www.conet.de, mailto: AStallmann at CONET.DE
> <mailto:AStallmann at CONET.DE>
>
> ----------------------------
> CONET Solutions GmbH, Theodor-Heuss-Allee 19, 53773 Hennef.
> Registergericht/Registration Court: Amtsgericht Siegburg (HRB Nr. 9136)
> Geschäftsführer/Managing Director: 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