Monitor MySQL instances and their databases in real time. Track availability, connections, query performance, storage engine health, replication status, and more from a single view.
Cloudmon provides deep monitoring for MySQL database servers running on both Windows and Linux. Once an instance is registered, all databases within it are automatically discovered and monitored at two levels: the instance level, covering the MySQL server as a whole, and the database level, covering each individual database running within it.
Navigate to Servers → Database → MySQL → MySQL Instance to view all monitored instances, or to Servers → Database → MySQL → MySQL Database for a cross-instance view of all databases.
Before registering a MySQL instance, a dedicated MySQL user must be created for the Cloudmon agent. The user can be restricted to localhost or allowed to connect from any host depending on whether the agent is installed locally or remotely.
Create the user and grant the required privileges by running the following on the MySQL server:
Replace username, hostname, and password with your chosen values. Use localhost as the hostname if the agent is installed on the same machine as the MySQL server, or % to allow connections from any host for remote agent setups. The MySQL server and listener must be reachable from the agent host.
Next, install the Cloudmon agent on the system that will perform the monitoring, whether that is the MySQL server itself or a separate remote host. With the agent installed, run the following command on that machine to register the MySQL instance for monitoring:
Adjust the path above if your agent is installed elsewhere. Once the command is run, the agent verifies connectivity to the database and begins collecting and forwarding metrics to the Cloudmon server. The newly registered instance will then appear under Servers → Database → MySQL → MySQL Instance in the Cloudmon web UI.
MySQL Instance monitoring covers the MySQL server as a whole. Cloudmon tracks availability, uptime, active and maximum connections, slow queries, network traffic, storage engine health, replication activity, command volumes, and the complete live server configuration from a single view.
The instance list ranks servers by connections, queries, slow queries, and network traffic, so the highest-load instances are always visible at a glance. An Instances Breakdown panel shows the hierarchical relationship between each server and its databases. Within an instance, Cloudmon monitors InnoDB in depth, covering buffer pool usage, row reads and writes, row lock waits, pages I/O, data bandwidth, and fsync activity. For environments using MyISAM, key buffer usage, key reads and writes, and key I/O are tracked separately.
Binary log disk growth, cache pressure, and replication command activity are monitored continuously, which is critical for any environment relying on MySQL replication or point-in-time recovery. The full live server configuration is surfaced alongside all performance data, so teams can correlate behaviour with the configuration driving it without querying the server directly.
MySQL Database monitoring provides visibility into each individual database within an instance. The database list page surfaces all databases across all monitored instances and ranks them by size, I/O, queries, latency, and throughput, making it easy to identify which database is driving the most load or consuming the most space across the entire environment.
Each database has its own availability tracking, query performance metrics, latency, throughput, I/O read and write activity, row statistics, and temp table counts. The Sequential versus Index Scans metric is particularly useful here: a high proportion of sequential scans relative to index scans is a strong signal that queries are missing indexes, which is often difficult to identify at the instance level alone. Errors and warnings are tracked at the database level as well, giving teams the context to distinguish whether an issue is isolated to one database or symptomatic of a broader server problem.
Cloudmon supports threshold-based alarms on any MySQL metric collected at the instance or database level, including connections, slow queries, InnoDB row lock waits, replication activity, and tablespace usage. Each alarm is built around a simple IF/THEN model, where you select a metric, set a threshold, and define what happens when it is breached. Learn more.
| Symptom | Likely Cause | Fix |
| Instance not appearing in the list | Instance not registered, or the agent cannot reach MySQL on the configured port | Re-run the registration command from the correct agent directory and verify the hostname, port, and credentials |
| High Slow Query count | Queries performing full table scans or missing indexes | Enable the Slow Query Log and review queries exceeding the Long Query Time threshold; add indexes where needed |
| Rising InnoDB Row Lock Waits | Concurrent transactions contending for the same rows | Identify long-running transactions holding locks and reduce transaction scope to minimise contention |
| High Aborted Connections count | Clients failing authentication or connections timing out before completing | Check MySQL error logs for authentication failures and review application connection pool timeout settings |
| Binlog disk growing rapidly | Binlog expiry not configured, or a replica lagging and preventing log purge | Set binlog_expire_logs_seconds to an appropriate retention period and investigate replica lag |
| Database not appearing in the list | The parent instance has not been registered or the agent is not running | Confirm the MySQL instance is registered and the Cloudmon agent is active on the host |
| High query latency on a specific database | Queries on that database performing full table scans or experiencing lock contention | Review the Sequential vs Index Scans metric for that database and add indexes where full scans are occurring |