The MySQL database is a crucial part of a wide variety of products, particularly web applications. Naturally, it is very important to monitor the health status of MySQL. However, there is constant disagreement on which of the many MySQL status variables provide the best overview on MySQL health status and indicate that something is not right with a server.
It certainly depends on what your application does – tuning read performance is different than optimizing write operations and everything changes when you have a cluster. The average user can use small subset of variables while advanced user want to get more detailed picture of the situation. So there cannot be one set of “magic variables” to quietly optimize every situation. However, it is possible to have a more-or-less optimal set of metrics that will allow to get a “good enough” notion about the general health status of MySQL Server.
It consists of four chapters and four appendices. The first three chapters give an overview of MySQL health monitoring. They discuss the way information is gathered, as well as which of the MySQL raw data variables are used. It briefly discusses how monitoring tools present the database health using calculated metrics — rather than raw data available from the database engine.
Chapter four describes calculated data in the Paid Monitor MySQL monitoring metrics, possible causes of metrics alerts and suggests the initial warning and critical threshold values and the best ways to address database issues, causing alerts.
You can download the full Paid Monitor white paper free here.
Should you have any questions, please don’t hesitate contacting us at [email protected] You can also try Paid Monitor MySQL tool at https://www.monitor.us/free_signup.jsp