Docs

RDS for MySQL

Monitoring and alerts

Monitoring metrics

This topic describes the monitoring items supported by RDS for MySQL and related information.

tip

For easier viewing, the RDS for MySQL monitoring dashboard automatically adjusts the unit of measurement based on the data volume. The actual unit displayed on the dashboard prevails. For example, when the InnoDB read volume is high, the unit changes from Byte/s to KB/s. If the read volume increases further, the unit changes to MB/s.

You can call the GetMetricData interface to query each monitoring indicator data and alarm information. When calling the query interface, you need to obtain the Namespace of the product, the SubNamespace to which the index to be queried belongs, and the MetricName of the index to be queried. The Namespace for RDS for MySQL is VCM_RDS_MySQL. The MetricName and SubNamespace for each metric are provided in the list of monitoring items below.

You can also obtain details about each monitoring indicator in Querying monitoring indicators for RDS for MySQL.

List of monitoring items

RDS for MySQL provides new monitoring items that allow you to view monitoring information at the instance and node levels under the resource monitoring, deployment monitoring, and engine monitoring categories. To use it, refer to List of monitoring items that support multi-dimensional viewing. The original monitoring items, which only support instance-level monitoring, are still available. Refer to List of monitoring items that only support the instance dimension for related information.

tip

If your instance does not support viewing monitoring information from node dimensions for resource monitoring, deployment monitoring, and engine monitoring, submit a ticket to apply for an upgrade of your instance version.

The new monitoring items provide all the information that the old items do. As support for the old items will be gradually discontinued, you should use the new monitoring items and migrate any business logic that relies on the old ones.

For instances of MySQL 8.4.7_20260415, MySQL 8.0.43_20251015, MySQL 5.7.44_20251015, and later versions, RDS for MySQL provides 5-second monitoring for some monitoring items. For a list of supported items, refer to List of monitoring items that support multi-dimensional viewing. This feature is currently in beta testing by invitation only. To use it for free, apply by submitting a ticket. If this capability is charged in the future, RDS for MySQL will announce the billing details one month in advance.

List of monitoring items that support multi-dimensional viewing

Monitoring Metrics	Name	MetricName	Subnamespace	Unit	Description of indicators	Supported viewing dimensions	Whether to support a 5-second aggregation cycle
CPU utilization	CPU utilization	CpuUtil	resource_monitor_new	Percent	CPU utilization.	Instance & Node	Yes
Memory utilization	Memory utilization	MemUtil	resource_monitor_new	Percent	Memory utilization.	Instance & Node	Yes
Disk utilization	Disk utilization	DiskUtil	resource_monitor_new	Percent	Disk space used/Purchased instance space first.	Instance & Node	No
network input traffic	network input traffic	NetworkReceiveThroughput	resource_monitor_new	Byte/s	Number of bytes received per second first.	Instance & Node	Yes
Network output traffic	Network output traffic	NetworkTransmitThroughput	resource_monitor_new	Byte/s	Number of bytes sent per second.	Instance & Node	No
IOPS usage rate	IOPS usage rate	IOPS	resource_monitor_new	Count/s	IOPS utilization.	Instance & Node	No
Total disk usage	-	DiskUsageBytes	resource_monitor_new	Bytes	Total disk usage. tip Only supports creating alarm strategies and does not provide a dashboard display.	None.	No
Total disk usage	Total disk usage	TotalDiskUsageBytes	resource_monitor_new	Bytes	Total disk usage.	Instance & Node	No
Slow log usage	Slow log usage	ErrLogUsageBytes	resource_monitor_new	Bytes	Disk usage of the slow log.	Instance & Node	No
Error log usage	Error log usage	ErrLogUsageBytes	resource_monitor_new	Bytes	Disk usage of the error log.	Instance & Node	No
Binlog usage	Binlog usage	BinlogUsageBytes	resource_monitor_new	Bytes	Binlog disk usage	Instance & Node	No
Redo log usage	Redo log usage	RedoLogUsageBytes	resource_monitor_new	Bytes	Disk usage of the redo log.	Instance & Node	No
Undo log usage	Undo log usage	UndoLogUsageBytes	resource_monitor_new	Bytes	Disk usage of the undo log first.	Instance & Node	No
Relay log usage	Relay log usage	RelayLogUsageBytes	resource_monitor_new	Bytes	Relay log disk usage	Instance & Node	No
System data usage	System data usage	SystemDataUsageBytes	resource_monitor_new	Bytes	Disk usage of system data.	Instance & Node	No
User data usage	User data usage	UserDataUsageBytes	resource_monitor_new	Bytes	Disk usage of user data	Instance & Node	No
Audit log usage	Audit log usage	AuditLogUsageBytes	resource_monitor_new	Bytes	Disk usage of the audit log.	Instance & Node	No
Temporary file usage	Temporary file usage	TempFileUsageBytes	resource_monitor_new	Bytes	Temporary file disk usage.	Instance & Node	No
Other log usage	Other log usage	OtherUsageBytes	resource_monitor_new	Bytes	Disk usage of other logs.	Instance & Node	No
ReplacationDelay	Slave replication latency	ReplicationDelay	deploy_monitor_new	Second	Primary-secondary delay time.	Instance & Node	Yes
TPS	TPS	ProxyTPS	proxy_monitor	Count	The number of transactions received by the database proxy service of the instance per second. Transactions include successfully committed transactions, failed rolled-back transactions, and single requests without explicit transactions.	Instance	No
QPS	QPS	ProxyQPS	proxy_monitor	Count	The number of requests that the database proxy of the instance receives per second.	Instance	No
Connections	Connections	ProxyConnection	proxy_monitor	Count	The number of connections for the database proxy service of the proxy instance. tip After MySQL reaches its max connections, the number of proxy connections is greater than the number of MySQL connections, with the proxy's max connections being twice that of MySQL.	Instance	No
CPU utilization	CPU utilization	ProxyCpuUtil	proxy_monitor	Percent	CPU usage rate of the database proxy service of the instance first.	Instance	No
Memory utilization	Memory utilization	ProxyMemUtil	proxy_monitor	Percent	Memory usage rate of the instance database proxy service.	Instance	No
Connection rate	Connection rate	ProxyNewConnections	proxy_monitor	Count/Second	Connection rate for the instance Database proxy.	Instance	No
Inbound traffic	Inbound traffic	ProxyFlowInBytes	proxy_monitor	Count	Total inbound traffic received by the database proxy service of the instance.	Instance	No
Outbound traffic	Outbound traffic	ProxyFlowOutBytes	proxy_monitor	Count	The outbound traffic of the database proxy service for the instance.	Instance	No
Total number of requests	Total requests	ProxyRequestNum	proxy_monitor	Count/Second	Total number of requests for the instance database proxy service.	Instance	No
Number of read requests	Number of read requests	ProxyReadReqestNum	proxy_monitor	Count/Second	Number of read requests for the instance Database proxy service first.	Instance	No
Write requests	Write requests	ProxyWriteRequestNum	proxy_monitor	Count/Second	Number of write requests for the instance database proxy service.	Instance	No
Connection utilization	-	ProxyConnectionUsage	proxy_monitor	Percent	Proxy connection usage rate. tip Only supports creating alarm strategies and does not provide a dashboard display.	None.	No
Connection utilization	Database proxy connection utilization	ProxyEndpointUsedConnRatio	proxy_monitor	Percent	Connection usage rate of the proxy endpoint.	Endpoint	No
Endpoint connections	Connections	ProxyEndpointConnection	proxy_monitor	Count	Proxy endpoint connections.	Endpoint	No
Endpoint TPS	TPS	ProxyEndpointTPS	proxy_monitor	Count	Number of transactions received per second by the proxy endpoint.	Endpoint	No
Endpoint QPS	QPS	ProxyEndpointQPS	proxy_monitor	Count	Number of requests received per second by the proxy endpoint.	Endpoint	No
Endpoint connection rate	Connection rate	ProxyEndpointNewConnections	proxy_monitor	Count/Second	Connection rate of the proxy endpoint.	Endpoint	No
Total endpoint request count	Total requests	ProxyEndpointRequestNum	proxy_monitor	Count/Second	Total number of requests for the proxy endpoint	Endpoint	No
Endpoint read requests	Number of read requests	ProxyEndpointReadReqNum	proxy_monitor	Count/Second	Number of read requests for the proxy endpoint.	Endpoint	No
Endpoint write requests	Write requests	ProxyEndpointWriteReqNum	proxy_monitor	Count/Second	Number of write requests for the Proxy endpoint.	Endpoint	No
Endpoint inbound traffic	Inbound traffic	ProxyEndpointFlowInBytes	proxy_monitor	Byte/Second	Inbound traffic of the proxy endpoint.	Endpoint	No
Endpoint egress traffic	Outbound traffic	ProxyEndpointFlowOutBytes	proxy_monitor	Byte/Second	Outbound traffic of the proxy endpoint.	Endpoint	No
Slow queries	Slow queries	SlowQueries	engine_monitor_new	Count/Second	Number of queries exceeding long_query_time Second	Instance & Node	No
Number of full table scans	Number of full table scans	SelectScan	engine_monitor_new	Count/Second	The number of queries that performed a full table Search	Instance & Node	Yes
Query Count	Query Count	OperationSelect	engine_monitor_new	Count/Second	QPS	Instance & Node	Yes
Updates	Updates	OperationUpdate	engine_monitor_new	Count/Second	Updates per second.	Instance & Node	No
Deleted count	Deleted count	OperationDelete	engine_monitor_new	Count/Second	Deletions per second.	Instance & Node	Yes
Insert count	Insert count	OperationInsert	engine_monitor_new	Count/Second	Number of INSERT requests per Second.	Instance & Node	Yes
Coverage count	Coverage count	OperationReplace	engine_monitor_new	Count/Second	Overwrites Per Second.	Instance & Node	No
Submit count	Submit count	OperationCommit	engine_monitor_new	Count/Second	Number of submits per second.	Instance & Node	Yes
Rollback count	Rollback count	OperationRollback	engine_monitor_new	Count/Second	Rollbacks per second first.	Instance & Node	Yes
Requests per second	Number of MySQL requests per second	QPS	engine_monitor_new	Count/Second	The quantity of queries sent by the MySQL client to MySQL per second, including queries within stored procedures. This metric counts COM_STMT_PREPARE, COM_STMT_CLOSE, and COM_STMT_RESET.	Instance & Node	Yes
TPS	TPS	TPS	engine_monitor_new	Count/Second	TPS (including succeeded transactions and failed rollback transactions).	Instance & Node	Yes
Number of Insert_Select	Insert select executions per Second	InsertSelect	engine_monitor_new	Count/Second	The average number of Insert_Select statements executed per second.	Instance & Node	No
Current number of open connections	Current number of open connections	ThreadsConnected	engine_monitor_new	Count	The number of currently open connections.	Instance & Node	Yes
Number of created threads	Number of created threads	ThreadsCreated	engine_monitor_new	Count	Number of threads created to handle connections.	Instance & Node	Yes
Running thread count	Running thread count	ThreadsRunning	engine_monitor_new	Count	The count of threads that are active and not in sleep state.	Instance & Node	No
Connection utilization percentage	MySQL connection usage rate	ConnUsage	engine_monitor_new	Percent	MySQL connection utilization ratio.	Instance & Node	Yes
Number of abnormally closed client connections	Number of abnormally closed client connections	AbortedClients	engine_monitor_new	Count	Number of connections aborted because the client terminated without closing the connection properly	Instance & Node	Yes
Number of failed connections	Number of failed connections	AbortedConnects	engine_monitor_new	Count	The number of failed attempts to connect to the MySQL server.	Instance & Node	Yes
Connections blocked by MDL lock	Connections waiting for an MDL lock	waitMDLConns	engine_monitor_new	Count	Connections blocked by MDL lock	Instance & Node	No
Number of transactions not submitted for more than 30 seconds	Number of transactions not submitted for more than 30 seconds	TrxOver30sAndSleep	engine_monitor_new	Count	Number of transactions in a sleep state that have been open for over 30 seconds first.	Instance & Node	No
Number of transactions that timed out after 1 minute and were not submitted	Number of transactions that timed out after 1 minute and were not submitted	TrxOver1mAndSleep	engine_monitor_new	Count	Number of transactions in a sleep state for over 1 minute and not closed first.	Instance & Node	Yes
Quantity of uncommitted transactions with a timeout of over 30 minutes	Quantity of uncommitted transactions with a timeout of over 30 minutes	TrxOver30mAndSleep	engine_monitor_new	Count	Number of transactions in a sleep State that have not been closed for more than 30 minutes.	Instance & Node	No
Quantity of large transactions with an execution time of more than 30 minutes	Quantity of large transactions with an execution time of more than 30 minutes	TrxOver30m	engine_monitor_new	Count	The number of large transactions with an execution time of more than 30 minutes.	Instance & Node	No
Quantity of large transactions with an execution time exceeding 1 hour	Quantity of large transactions with an execution time exceeding 1 hour	TrxOver1h	engine_monitor_new	Count	Quantity of large transactions with an execution time exceeding 1 hour.	Instance & Node	No
Quantity of large transactions exceeding 24 hours	Quantity of large transactions exceeding 24 hours	TrxOver24h	engine_monitor_new	Count	Number of large transactions with an execution time of more than 24 hours.	Instance & Node	No
Temporary table quantity	Temporary table quantity	CreatedTmpTables	engine_monitor_new	Count/Second	Quantity of temporary tables created	Instance & Node	No
Number of table lock waits	Number of table lock waits	TableLocksWaited	engine_monitor_new	Count/Second	The number of times that table locks could not be obtained Now	Instance & Node	Yes
Open tables	Number of open tables	OpenedTables	engine_monitor_new	Count	Number of open tables.	Instance & Node	No
Number of open files	Open file count	OpenFiles	engine_monitor_new	Count	Number of open files.	Instance & Node	Yes
Innodb cache hit rate	Innodb cache hit rate	InnodbCacheHitRate	engine_monitor_new	Percent	The cache hit rate of the InnoDB Engine.	Instance & Node	Yes
Innodb cache usage rate	Innodb cache usage rate	InnodbCacheUtil	engine_monitor_new	Percent	InnoDB Engine cache usage rate first.	Instance & Node	Yes
Innodb read volume	Innodb read volume	InnodbDataRead	engine_monitor_new	Byte/s	The number of bytes of data read by the InnoDB Engine per second	Instance & Node	No
Innodb write volume	Innodb write volume	InnodbDataWritten	engine_monitor_new	Byte/s	The number of bytes of data written by the InnoDB engine per second.	Instance & Node	Yes
Innodb reads	Innodb reads	InnodbDataReadCounts	engine_monitor_new	Count/Second	Number of data reads completed by the InnoDB Engine per second.	Instance & Node	Yes
Number of Innodb writes	Innodb writes	InnodbDataWriteCounts	engine_monitor_new	Count/Second	The number of writes completed by the InnoDB Engine per second.	Instance & Node	No
Number of InnoDB tables currently open	Number of InnoDB tables currently open	InnodbNumOpenFiles	engine_monitor_new	Count	Current number of open tables in the InnoDB engine.	Instance & Node	Yes
Number of Innodb rows read	Number of Innodb rows read	InnodbDataReadBytes	engine_monitor_new	Count/Second	Number of rows read by the InnoDB Engine per second.	Instance & Node	Yes
Innodb rows deleted	Innodb rows deleted	InnodbRowsDeleted	engine_monitor_new	Count/Second	The number of rows deleted per second by the InnoDB engine.	Instance & Node	No
Innodb row insert quantity	Innodb row insert quantity	InnodbRowsInserted	engine_monitor_new	Count/Second	The number of rows inserted by the InnoDB engine per second.	Instance & Node	Yes
Innodb rows updated	Innodb rows updated	InnodbRowsUpdated	engine_monitor_new	Count/Second	Number of rows updated by the InnoDB engine per second.	Instance & Node	Yes
Average Innodb row lock obtain time	Innodb average row lock wait time	InnodbRowsLockTimeAvg	engine_monitor_new	Millisecond	Average duration of InnoDB Engine row locks. tip This metric is the average wait time for row locks since MySQL started. It is calculated by dividing the total wait time for row locks by the total number of row lock waits. It is not a real-time Value and is updated only when a new row lock wait occurs. After MySQL restarts, this metric is cleared and recounted, which causes the Value to change abruptly.	Instance & Node	No
Number of InnoDB row lock waits	Innodb row lock wait count	InnodbRowLockWaits	engine_monitor_new	Count/Second	The number of times the InnoDB Engine waits for row locks per second.	Instance & Node	Yes
InnoDB table maximum row lock wait time	InnoDB table maximum row lock wait time	InnodbRowLockTimeMax	engine_monitor_new	Millisecond	Maximum InnoDB wait time for row locks.	Instance & Node	Yes
Average number of physical writes to the InnoDB redo log file per second	Average number of physical writes to the InnoDB redo log file per second	InnodbLogWrites	engine_monitor_new	Count/Second	Average number of physical writes to the Redo Log File per second by InnoDB.	Instance & Node	No
Average number of fsync writes completed to the log file per second	Average number of fsync writes completed to the log file per second	InnodbOsLogFsyncs	engine_monitor_new	Count/Second	The average number of fsync() writes completed to the Log file per second.	Instance & Node	No
Average Number of Completed Fsync Writes to Data Files Per Second	InnoDB average fsync operations per second	InnodbDataFsyncs	engine_monitor_new	Count/Second	The average number of fsync() writes completed to data files per second.	Instance & Node	No
Redo log usage	Redo log usage	RedologSize	engine_monitor_new	MByte	Disk usage of Redo log.	Instance & Node	No
Slow log usage	Slow log usage	SlowLogSize	engine_monitor_new	MByte	The disk usage of the slow log.	Instance & Node	No
Number of temporary files	Number of temporary files	CreatedTmpFiles	engine_monitor_new	Count/Second	Number of temporary files created per second.	Instance & Node	No
Quantity of temporary disk tables	Quantity of temporary disk tables	CreatedTmpDiskTables	engine_monitor_new	Count/Second	The number of temporary tables created on disk per second.	Instance & Node	No
Number of requests to read the next row	Number of requests to read the next row	HandlerReadRndNext	engine_monitor_new	Count/Second	Number of requests to read the next row per second.	Instance & Node	No
Internal rollback count	Internal rollback count	HandlerRollback	engine_monitor_new	Count/Second	The number of times transactions are rolled back per second.	Instance & Node	Yes
Internal submit count	Internal submit count	HandlerCommit	engine_monitor_new	Count/Second	Number of transaction submits per second.	Instance & Node	No
Quantity of empty Innodb pages	Quantity of empty Innodb pages	InnodbBufferPoolPagesFree	engine_monitor_new	Count	Number of empty pages in the InnoDB Engine memory.	Instance & Node	Yes
Total Innodb pages	Total Innodb pages	TotalInnodbBufferPoolPages	engine_monitor_new	Count	Total number of memory pages occupied by the InnoDB engine.	Instance & Node	Yes
Innodb logical reads	Innodb logical reads	InnodbBufferPoolReadRequests	engine_monitor_new	Count/Second	The number of logical read requests that the InnoDB Engine has Completed per second.	Instance & Node	No
Innodb physical read	Innodb physical read	InnodbBufferPoolReads	engine_monitor_new	Count/Second	The number of physical read requests completed by the InnoDB engine per second.	Instance & Node	Yes
InnoDB buffer pool dirty page ratio	InnoDB buffer pool dirty page ratio	InnodbBpDirtyPct	engine_monitor_new	Percent	The ratio of dirty pages in the InnoDB buffer pool, calculated by the formula: InnoDB_buffer_pool_pages_dirty/InnoDB_buffer_pool_pages_data* 100%.	Instance & Node	Yes
InnoDB Buffer Pool page flush request quantity	InnoDB Buffer Pool page flush request quantity	InnodbBufferPoolPagesFlushed	engine_monitor_new	Count/Second	Number of InnoDB Buffer Pool page flush requests.	Instance & Node	No
RocksDB read data volume	RocksDB disk reads	RocksDBBytesRead	engine_monitor_new	Byte/Second	Bytes read from disk per second by the RocksDB Engine.	Instance & Node	No
RocksDB write data volume	RocksDB Disk Write Volume	RocksDBBytesWritten	engine_monitor_new	Byte/Second	The number of bytes written to disk per second by the RocksDB engine.	Instance & Node	No
RocksDB read cache amount	RocksDB read cache amount	RocksDBBlockCacheBytesRead	engine_monitor_new	Byte/Second	Data volume read from Cache by the RocksDB engine.	Instance & Node	No
RocksDB write cache amount	RocksDB write cache amount	RocksDBBlockCacheBytesWrite	engine_monitor_new	Byte/Second	Data volume written to the cache by the RocksDB engine.	Instance & Node	No
RocksDB WAL write volume	RocksDB WAL log write size	RocksDBWalLogCapacity	engine_monitor_new	Byte/Second	Data volume written to the WAL log by the RocksDB engine	Instance & Node	No

List of monitoring items that only support the instance dimension

Monitoring Metrics	Name	MetricName	Subnamespace	Unit	Description of indicators
CPU utilization	CPU utilization	CpuUtil	resource_monitor	Percent	CPU utilization.
Memory utilization	Memory utilization	MemUtil	resource_monitor	Percent	Memory utilization.
Disk utilization	Disk utilization	DiskUtil	resource_monitor	Percent	Disk space used/Purchased instance space first.
network input traffic	network input traffic	NetworkReceiveThroughput	resource_monitor	Byte/s	Number of bytes received per second first.
Network output traffic	Network output traffic	NetworkTransmitThroughput	resource_monitor	Byte/s	Number of bytes sent per second.
IOPS	IOPS	IOPS	resource_monitor	Count/s	IOPS。
Disk usage	Disk usage	DiskUsageBytes	resource_monitor	Bytes	Disk usage.
ReplacationDelay	Slave replication latency	ReplicationDelay	deploy_monitor	Second	Primary-secondary delay time.
TPS	TPS	ProxyTPS	proxy_monitor	Count	The number of transactions the Proxy receives per Second. Transactions include successfully committed transactions, failed rolled-back transactions, and single requests without explicit transactions.
QPS	QPS	ProxyQPS	proxy_monitor	Count	Number of requests received by the proxy per second.
Connections	Connections	ProxyConnection	proxy_monitor	Count	Proxy connection count. tip After MySQL reaches its max connections, the number of proxy connections is greater than the number of MySQL connections, with the proxy's max connections being twice that of MySQL.
CPU utilization	CPU utilization	ProxyCpuUtil	proxy_monitor	Percent	The CPU usage rate of the database proxy.
Memory utilization	Memory utilization	ProxyMemUtil	proxy_monitor	Percent	Proxy memory usage rate
Connection utilization	Database proxy connection utilization	ProxyConnectionUsage	proxy_monitor	Percent	Proxy connection usage rate.
Slow queries	Slow queries	SlowQueries	engine_monitor	Count/Second	Number of queries exceeding long_query_time Second
Number of full table scans	Number of full table scans	SelectScan	engine_monitor	Count/Second	The number of queries that performed a full table Search
Updates	Updates	OperationUpdate	engine_monitor	Count/Second	Updates per second.
Deleted count	Deleted count	OperationDelete	engine_monitor	Count/Second	Deletions per second.
Insert count	Insert count	OperationInsert	engine_monitor	Count/Second	Number of INSERT requests per Second.
Coverage count	Coverage count	OperationReplace	engine_monitor	Count/Second	Overwrites Per Second.
Submit count	Submit count	OperationCommit	engine_monitor	Count/Second	Number of submits per second.
Rollback count	Rollback count	OperationRollback	engine_monitor	Count/Second	Rollbacks per second first.
Requests per second	Number of MySQL requests per second	QPS	engine_monitor	Count/Second	The quantity of queries sent by the MySQL client to MySQL per second, including queries within stored procedures. This metric counts COM_STMT_PREPARE, COM_STMT_CLOSE, and COM_STMT_RESET.
TPS	TPS	TPS	engine_monitor	Count/Second	TPS (including succeeded transactions and failed rollback transactions).
Insert_Selects per second	Average number of insert select executions per second	InsertSelect	engine_monitor	Count/Second	The average number of Insert_Select statements executed per second.
Current number of open connections	Current number of open connections	ThreadsConnected	engine_monitor	Count	The number of currently open connections.
Number of created threads	Number of created threads	ThreadsCreated	engine_monitor	Count	Number of threads created to handle connections.
Running thread count	Running thread count	ThreadsRunning	engine_monitor	Count	The count of threads that are active and not in sleep state.
Connection utilization percentage	MySQL connection usage rate	ConnUsage	engine_monitor	Percent	MySQL connection utilization ratio.
Temporary table quantity	Temporary table quantity	CreatedTmpTables	engine_monitor	Count/Second	Quantity of temporary tables created
Number of table lock waits	Number of table lock waits	TableLocksWaited	engine_monitor	Count/Second	The number of times that table locks could not be obtained Now
Open tables	Number of open tables	OpenedTables	engine_monitor	Count	Number of opened tables.
Number of open files	Open file count	OpenFiles	engine_monitor	Count	Number of open files.
Innodb cache hit rate	Innodb cache hit rate	InnodbCacheHitRate	engine_monitor	Percent	The cache hit rate of the InnoDB Engine.
Innodb cache usage rate	Innodb cache usage rate	InnodbCacheUtil	engine_monitor	Percent	InnoDB Engine cache usage rate first.
Innodb read volume	Innodb read volume	InnodbDataRead	engine_monitor	Byte/s	The number of bytes of data read by the InnoDB Engine per second
Innodb write volume	Innodb write volume	InnodbDataWritten	engine_monitor	Byte/s	The number of bytes of data written by the InnoDB engine per second.
Innodb reads	Innodb reads	InnodbDataReadCounts	engine_monitor	Count/Second	Number of data reads completed by the InnoDB Engine per second.
Number of Innodb writes	Innodb writes	InnodbDataWriteCounts	engine_monitor	Count/Second	The number of writes completed by the InnoDB Engine per second.
Number of InnoDB tables currently open	Number of InnoDB tables currently open	InnodbNumOpenFiles	engine_monitor	Count	Current number of open tables in the InnoDB engine.
Number of Innodb rows read	Number of Innodb rows read	InnodbDataReadBytes	engine_monitor	Count/Second	Number of rows read by the InnoDB Engine per second.
Innodb rows deleted	Innodb rows deleted	InnodbRowsDeleted	engine_monitor	Count/Second	The number of rows deleted per second by the InnoDB engine.
Innodb row insert quantity	Innodb row insert quantity	InnodbRowsInserted	engine_monitor	Count/Second	The number of rows inserted by the InnoDB engine per second.
Innodb rows updated	Innodb rows updated	InnodbRowsUpdated	engine_monitor	Count/Second	Number of rows updated by the InnoDB engine per second.
Average Innodb row lock obtain time	Innodb average row lock wait time	InnodbRowsLockTimeAvg	engine_monitor	Millisecond	Average duration of InnoDB Engine row locks. tip This metric is the average wait time for row locks since MySQL started. It is calculated by dividing the total wait time for row locks by the total number of row lock waits. It is not a real-time Value and is updated only when a new row lock wait occurs. After MySQL restarts, this metric is cleared and recounted, which causes the Value to change abruptly.
Number of InnoDB row lock waits	Innodb row lock wait count	InnodbRowLockWaits	engine_monitor	Count/Second	The number of times the InnoDB Engine waits for row locks per second.
InnoDB table maximum row lock wait time	InnoDB table maximum row lock wait time	InnodbRowLockTimeMax	engine_monitor	Millisecond	Maximum InnoDB wait time for row locks.
Average number of physical writes to the InnoDB redo log file per second	Average number of physical writes to the InnoDB redo log file per second	InnodbLogWrites	engine_monitor	Count/Second	Average number of physical writes to the Redo Log File per second by InnoDB.
Average number of fsync writes completed to the log file per second	Average number of fsync writes completed to the log file per second	InnodbOsLogFsyncs	engine_monitor	Count/Second	The average number of fsync() writes completed to the Log file per second.
Average Number of Completed Fsync Writes to Data Files Per Second	InnoDB average fsync operations per second	InnodbDataFsyncs	engine_monitor	Count/Second	The average number of fsync() writes completed to data files per second.
Redo log usage	Redo log usage	RedologSize	engine_monitor	MByte	Disk usage of Redo log.
Slow log usage	Slow log usage	SlowLogSize	engine_monitor	MByte	The disk usage of the slow log.
Number of temporary files	Number of temporary files	CreatedTmpFiles	engine_monitor	Count/Second	Number of temporary files created per second.
Quantity of temporary disk tables	Quantity of temporary disk tables	CreatedTmpDiskTables	engine_monitor	Count/Second	The number of temporary tables created on disk per second.
Number of requests to read the next row	Number of requests to read the next row	HandlerReadRndNext	engine_monitor	Count/Second	Number of requests to read the next row per second.
Internal rollback count	Internal rollback count	HandlerRollback	engine_monitor	Count/Second	The number of times transactions are rolled back per second.
Internal submit count	Internal submit count	HandlerCommit	engine_monitor	Count/Second	Number of transaction submits per second.
Quantity of empty Innodb pages	Quantity of empty Innodb pages	InnodbBufferPoolPagesFree	engine_monitor	Count	Number of empty pages in the InnoDB Engine memory.
Total Innodb pages	Total Innodb pages	TotalInnodbBufferPoolPages	engine_monitor	Count	Total number of memory pages occupied by the InnoDB engine.
Innodb logical reads	Innodb logical reads	InnodbBufferPoolReadRequests	engine_monitor	Count/Second	The number of logical read requests that the InnoDB Engine has Completed per second.
Innodb physical read	Innodb physical read	InnodbBufferPoolReads	engine_monitor	Count/Second	The number of physical read requests completed by the InnoDB engine per second.
InnoDB buffer pool dirty page ratio	InnoDB buffer pool dirty page ratio	InnodbBpDirtyPct	engine_monitor	Percent	The ratio of dirty pages in the InnoDB buffer pool, calculated by the formula: InnoDB_buffer_pool_pages_dirty/InnoDB_buffer_pool_pages_data* 100%.
InnoDB Buffer Pool page flush request quantity	InnoDB Buffer Pool page flush request quantity	InnodbBufferPoolPagesFlushed	engine_monitor	Count/Second	Number of InnoDB Buffer Pool page flush requests.

Alert policy configuration recommendations

We recommend that you configure trigger conditions for the following resource-level monitoring items in your alert policy as shown in the table below.

MetricName	Subnamespace	Number of continuous cycles	Statistical method	Recommended threshold
CPU utilization	CpuUtil	3	Average value	> 80%
System data usage	SystemDataUsageBytes	3	Average value	> 80%
User data usage	UserDataUsageBytes	3	Average value	> 80%
Memory utilization	MemUtil	3	Average value	> 90%

Last updated: 2026.07.03 18:15:19