MySQL Enterprise Monitor

Overview

MySQL Enterprise Monitor is used to to monitor MySQL servers. It automatically checks server operating status, security, and performance etc. - which are usually conducted manually by database administrators, and notifies any potential problem or increasing load before it develops. These notifications help administrators prevent from database crashing and grasp the necessity of tuning.
By utilizing various features that MySQL Enterprise Monitor provides, database administrators can reduce efforts, time, and costs while safety and efficiency increase.

MySQL Enterprise Monitor includes:

MySQL Enterprise Monitor Dashboard
MySQL Query Analyzer
MySQL Replication Monitor

MySQL Enterprise Monitor Dashboard

MySQL Enterprise Monitor Dashboard is used to conduct basic management and monitor security, performance, and availability of one or more MySQL servers.

Figure1. MySQL Enterprise Monitor Dashboard (click to enlarge)

MySQL Enterprise Monitor Dashboard includes the followings.

Replication and Scale-out detection	The replication feature of MySQL databases is used in many systems, and additionally the case that server-related things, such as conducting scale-out, become complicated occurrences. However, such related things are not changed into documents, so it is not rare to need a lot of time for grasping such things when database managers change something or scale-out etc. The detecting function solves such problems. Thanks to this, it is possible to grasp the relationship between master and slave and scale-out of each group easily. As a result, you can conduct the works of replication and scale-out faster than before.
MySQL Adviser	The right side of Figure 1 is called hit chart, and it is possible to grasp on or offline situation, OS condition, and performance situation of all the servers or only servers divided into groups. By grasping this, it is possible to define which server must be dealt with and which part the work must be done to, and it is useful for server maintenance.
Availability check and Performance diagnosis	On normal operation, database managers always supervise MySQL’s server setting, security, and performance and conduct problem correspondence and tuning. But, this requires labor, and it is likely for mistakes such as checking the delay of problems to occur. Then, in MySQL Enterprise Monitor, the rules to stably operate MySQL servers called Advisor Rules are prepared, and checks based on the Advisor Rule are automatically conducted. As a result, content problems are detected, and the problems, tuning necessity, and correction action (correcting places and methods) are noticed. MySQL advisor supports managers on management, security, upgrade, customization, replication, memory, schema, and performance. Figure2. MySQL Adviser notification example

Replication and Scale-out detection

The replication feature of MySQL databases is used in many systems, and additionally the case that server-related things, such as conducting scale-out, become complicated occurrences. However, such related things are not changed into documents, so it is not rare to need a lot of time for grasping such things when database managers change something or scale-out etc. The detecting function solves such problems.
Thanks to this, it is possible to grasp the relationship between master and slave and scale-out of each group easily. As a result, you can conduct the works of replication and scale-out faster than before.

MySQL Adviser

The right side of Figure 1 is called hit chart, and it is possible to grasp on or offline situation, OS condition, and performance situation of all the servers or only servers divided into groups. By grasping this, it is possible to define which server must be dealt with and which part the work must be done to, and it is useful for server maintenance.

Availability check and Performance diagnosis

On normal operation, database managers always supervise MySQL’s server setting, security, and performance and conduct problem correspondence and tuning. But, this requires labor, and it is likely for mistakes such as checking the delay of problems to occur.
Then, in MySQL Enterprise Monitor, the rules to stably operate MySQL servers called Advisor Rules are prepared, and checks based on the Advisor Rule are automatically conducted.
As a result, content problems are detected, and the problems, tuning necessity, and correction action (correcting places and methods) are noticed. MySQL advisor supports managers on management, security, upgrade, customization, replication, memory, schema, and performance.

SQL Adviser

Figure2. MySQL Adviser notification example

MySQL Query Analyzer

It is possible to reveal query details and action times. The analyzer can conduct the detection of the queries taking the longest time, queries implemented excessively, and queries with errors. Additionally, it is possible to check the influences that queries give to systems. Thanks to these things, developers can specialize and correct the SQL codes that become problems, and the systems can exert their best performances.

MySQL Query Analyzer has the following capabilities.

Analyze queries using the MySQL connector plug-in
Scan all tables.
Detect queries with problems such as invalid index etc.
Monitor servers running in real time, query hitories
Independent from MySQL log or SHOW PROCESSLIST
Superior in visibility and operability
Check usage of access path index through EXPLAIN

Figure3. MySQL Query Analyzer (click to enlarge)

MySQL Replication Moniter

In MySQL Replication Monitor, it is possible to check the operation situation, performance, and availability on master and slave in real time. Also, based on the check rule called Advisor Rule, the Replication Monitor checks and notices any alerts in case that there is any problem. From this notice, the database managers analyze the problems of master or slave, investigate (I/O, SQL threads, logs, errors, and so on), and cope with what they found. And, it is possible to reduce the labor for protection by creating the scripts for collecting and supervising replication situation and data.