ordergroove / check_mariadb_slaves Goto Github PK

Currently a connection name must be provided. There should be a way to check all slave connections of a MariaDB instance.

Considerations:

• What would the output look like in the Nagios admin?
  -- Should all connections and their respective states be enumerated?
  -- Should it simply return "OK" if everything is fine and only report on slaves that are failing there checks?

This was initiated by #4

Thanks for the plugin, it was easy to get going!

Nagios plugins may optionally return "performance data" in their output.
This would be useful in "replication_lag" mode, to track the number of seconds that the slave is behind the master. All that would be required would be this:
OK - Slave is 0 seconds behind master | slave_lag=0;<w>;<c>
instead of this:
OK - Slave is 0 seconds behind master
where <w> is the warning threshold and <c> is the critical threshold

Make error messages appear with sanitized passwords, such as

CRITICAL - Slave sql is not running. Last error: Error Operation CREATE USER failed for [email protected] on query. Default database: . Query: CREATE USER [email protected] IDENTIFIED BY *****