I'm trying to configure Pacemaker on Ubuntu 18.04 so I can run a master and a slave and I've been smashing my head against a brick wall all day.
I have two nodes,
mysql2 both with MariaDB 10.13 & Pacemaker 1.18 installed.
bind-address=0.0.0.0 server_id=1 log-basename=master log-bin=/var/log/mysql/mariadb-bin binlog-format=row binlog-do-db=masterdb
bind-address=0.0.0.0 server-id=2 replicate-do-db=masterdb
Both installations start no problem using
systemctl start mysql and I've got master/slave replication configured on them. However, when trying to configure the Pacemaker configuration using
mysql1 starts no problem.
mysql2, again starts no problem but eventually pacemaker reports a timeout -
root@mysql1:~# pcs status Cluster name: mysqlcluster Stack: corosync Current DC: mysql2 (version 1.1.18-2b07d5c5a9) - partition with quorum Last updated: Tue May 28 13:00:21 2019 Last change: Tue May 28 12:56:15 2019 by root via crm_attribute on mysql1 2 nodes configured 2 resources configured Online: [ mysql1 mysql2 ] Full list of resources: Master/Slave Set: mysql-master [mysql] Masters: [ mysql1 ] Stopped: [ mysql2 ] Failed Actions: * mysql_start_0 on mysql2 'unknown error' (1): call=57, status=Timed Out, exitreason='', last-rc-change='Tue May 28 12:54:15 2019', queued=0ms, exec=120002ms Daemon Status: corosync: active/enabled pacemaker: active/enabled pcsd: active/enabled
I've tried a number of things to get this working but nothing has been of use. Here's the pcs command I'm running -
pcs resource create mysql ocf:heartbeat:mysql binary="/usr/sbin/mysqld" config="/etc/mysql/my.cnf" op monitor role="Master" timeout="30s" interval="10s" op monitor role="Slave" timeout="30s" interval="30s" meta notify=true --master
Again, as I mentioned MySQL seems to start fine on both machines and there are no errors recorded in the MySQL log, just the service suddenly gets killed after the 120s timeout. I've also checked pacemaker.log, corosync.log and Googled heavily but there doesn't seem to be much advice out there... I've also tried this setup on CentOS 7 and I'm getting the exact same issue.