The MySQL cartridge provides a comprehensive set of preconfigured rules designed to monitor various aspects of MySQL server performance and health. You can modify these rules, including changing default threshold values. For more details, refer to Working with alarms templates or contact Quest Software PSO.
This section describes the following rules:
Alerts when the percentage of authentication errors exceeds a defined threshold relative to total connection attempts.
Alerts when a transaction has been waiting too long to complete.
Alerts when the number of active queries exceeds the defined limit.
Alerts when a connection to the MySQL database cannot be established.
Alerts when not all nodes in the Galera cluster are available.
Alerts when a MySQL server is disconnected from the Galera cluster.
Alerts when the average EVS latency is too high.
Alerts when the percentage of time flow control is paused exceeds acceptable levels.
Alerts when a MySQL server is not ready to accept queries.
Alerts when the average size of the Galera receive queue is too high.
Alerts when the average size of the Galera send queue is too high.
Alerts when the average wait time for a statement instance significantly increases compared to its historical average.
Alerts when the percentage of compression failures for a specific page size is too high.
Alerts when the success rate of connections to the MySQL server is too low.
Alerts when the index compression failure rate is too high.
Alerts when query sort operations are inefficient.
Alerts when the InnoDB Buffer Pool hit rate falls below acceptable levels.
Alerts when one or more nodes in an InnoDB cluster are down, with increasing severity for greater instability.
Alerts when the key buffer hit rate is too low.
Alerts for queries that exceed a specified execution time.
Sends email notifications for alarms with specified severities.
Alerts when the database server is at risk of running out of disk space based on current growth rates.
Alerts when the percentage of non-authentication errors exceeds the defined threshold.
Alerts when the query cache hit rate is too low.
Alerts when query cache is pruned due to low memory.
Alerts when queries wait too long for table locks.
Alerts when replication servers' clocks are out of sync.
Alerts when a replication slave lags behind the master.
Alerts when a replication slave cannot be reached by the agent.
Alerts when the replication server’s I/O state is reconnecting or waiting to reconnect.
Alerts when the I/O thread for a replication slave server is not running.
Alerts when the SQL thread for a replication slave server is not running.
Alerts when connections take too long to establish.
Alerts when the average execution time for slow queries exceeds acceptable thresholds.
Alerts when the maximum execution time for slow queries exceeds acceptable thresholds.
Alerts when the MySQL server does not utilize indexes efficiently.
Alerts when available tablespace is low in an InnoDB database without auto-extend enabled.
Alerts for agents using basic table collection when tables do not persist as topology objects.
Removes alarms triggered by basic table collection rules if they are not re-triggered within four hours.
Alerts when the thread cache is disabled.
Alerts when the thread pool hit rate is too low.
Alerts when the server has too many idle connections.
Alerts when the transaction purge lag exceeds acceptable levels.
Alerts when the percentage of unflushed log buffer entries is too high.
Alerts when the percentage of used connections approaches the maximum limit.