Monitoring MySQL
Ronald Bradford wants to know what kind of Monitoring you use..
He specifically wants to know about Alerting tools
There's different cases , looking at it from a full infrastructure point my current favourite is Zabbix or good old Nagios,
But when looking at it from a debugging perspective you have MySQLAR or Hyperic, but those aren't in the alerting list.
However, when you are building HA clusters, you have custom scripts running either from mon or from pacemaker ..
Still .. Ronald probably wants more input :)
Comments
#1 Anonymous : In our clusters we use custom
In our clusters we use custom written check scripts via Nagios and collect and store metrics using RRDTool. This is all deployed automatically using Puppet so it requires no setup when adding or removing machines from the cluster.
MySQL specific metrics we check for are;
- Is MySQL running and uptime
- Number of queries executed per second
- SQL related errors in log
- Replication lag
- Crashed tables
- Number of connections present
- Slow queries monitoring
- Query cache stats