Documentation | Tencent Cloud

Recent Pages

Monitoring Feature

Last updated: 2021-04-27 10:57:10

To make it easier for you to view and stay up to date with how instances work, TencentDB for MySQL provides a wide variety of performance monitoring metrics and convenient monitoring features (custom view, time comparison, merged monitoring metrics, etc). You can log in to the TencentDB for MySQL console, and view them in Instance Monitoring on the instance management page.

Note：

You can get instance monitoring metrics by calling the GetMonitorData API or using the TencentDB for MySQL monitoring metrics in Cloud Monitor.

You can create dashboards for monitoring metrics to analyze monitored data dynamically.

If the number of tables in a single instance exceeds one million, database monitoring may be affected. Please control this value appropriately and make sure that it is below one million.

Types of Instances for Monitoring

TencentDB for MySQL source instances, read-only replicas, and disaster recovery instances can be monitored, and each instance is provided with a separate monitoring view for easy query.

Types of Monitoring

Four types of monitoring are available for TencentDB for MySQL: resource monitoring, engine monitoring (general), engine monitoring (extended), and deployment monitoring. You can view the metrics of different monitoring types to gain a quick and accurate understanding of how instances perform and operate.

Resource monitoring provides monitoring data of CPU, memory, disk, and network.
Engine monitoring (general) provides monitoring data of the number of connections, locks, hotspot tables, and slow queries, helping you troubleshoot issues and optimize the performance.
Engine monitoring (extended) provides a wider variety of engine-related monitoring metrics so as to assist you in identifying existing or potential database problems as much as possible.
Deployment monitoring provides monitoring metrics with regard to source-replica delay. It divides into source monitoring and replica monitoring:
Deploy monitoring on the source: when the monitored instance is a source instance which is not a replica of any instance, replication-related monitoring data is invalid for the source, and the IO and SQL threads are disabled. Replication-related monitoring data is valid and the IO and SQL threads can be enabled only when the monitored instance is a disaster recovery or a read-only replica.
Deploy monitoring on the replica: the two-node or three-node source instance and disaster recovery instance come in a source/replica architecture by default. As a result, replication-related monitoring data is valid for the replica only when the monitored instance is a source or disaster recovery instance. Such monitoring data can reflect the replication delay distance and time between the source or disaster recovery instance and its hidden replica nodes. You are recommended to keep an eye out for such monitoring data of the replica. If the source or disaster recovery instance fails, its monitored hidden replica nodes can be promoted into the source instance quickly.

Monitoring Granularity

TencentDB for MySQL has adopted an adaptive policy for monitoring granularity since August 11, 2018, which means that you cannot select a monitoring granularity as desired for the time being. The adaptive policy is as follows:

Time Span	Monitoring Granularity	Adaptation Description	Retention Period
(0h, 4h]	5 seconds	The time span is below 4 hours, and the monitoring granularity is 5 seconds	1 day
(4h, 2d]	1 minute	The time span is above 4 hours but below 2 days, and the monitoring granularity is 1 minute	15 days
(2d, 10d]	5 minutes	The time span is above 2 days but below 10 days, and the monitoring granularity is 5 minutes	31 days
(10d, 30d]	1 hour	The time span is above 10 days but below 30 days, and the monitoring granularity is 1 hour	62 days

Note：
Currently, you can view monitoring data of TencentDB for MySQL in the past 30 days.

Monitored Metrics

Cloud Monitor provides the following monitoring metrics for TencentDB for MySQL instances in the instance dimension:

For more information on how to use TencentDB monitoring metrics, please see the TencentDB for MySQL APIs in Cloud Monitor.

Metric Name	Parameter	Unit	Description
Queries per Second	qps	Counts/second	Number of SQL statements (INSERT, SELECT, UPDATE, DELETE, and REPLACE) executed by the database per second. This metric mainly represents the actual processing capability of the TencentDB instance.
Transactions per Second	tps	Counts/second	The number of transactions executed per second in the database
Slow Queries	slow_queries	-	The number of queries that take more than `long_query_time` second(s) to be executed
Full Table Scans	select_scan	Counts/second	The number of full-table scans executed per second
SELECT Queries	select_count	Counts/second	The number of SELECT statements executed per second
UPDATE Queries	com_update	Counts/second	Number of UPDATE statements executed per second
DELETE Queries	com_delete	Counts/second	The number of DELETE statements executed per second
INSERT Queries	com_insert	Counts/second	The number of INSERT statements executed per second
REPLACE Queries	com_replace	Counts/second	The number of REPLACE statements executed per second
Total Queries	queries	Counts/second	All executed SQL statements such as SET and SHOW.
Open Connections	threads_connected	-	The number of currently open connections
Connection Utilization	connection_use_rate	%	The number of open connections/the maximum number of connections
Query Utilization	query_rate	%	Actual QPS/recommended QPS
Used Disk Space	capacity	MB	This includes MySQL's data directories and logs such as binlog, relaylog, undolog, errorlog, and slowlog.
Disk Space Used by Data	real_capacity	MB	This includes only MySQL's data directories.
Disk Space Used by Logs	disk_log_used	MB	This includes only MySQL's logs such as binlog, relaylog, undolog, errorlog, and slowlog.
Disk Space Used by Temp Files	disk_tmp_used	MB	This includes only MySQL's temp files.
Private Outbound Traffic	bytes_sent	Bytes/second	Bytes sent per second
Private Inbound Traffic	bytes_received	Bytes/second	Bytes received per second
Disk Utilization	volume_rate	%	Used disk capacity/purchased instance capacity
Query Cache Hit Rate	qcache_hit_rate	%	Query cache hit rate
Query Cache Utilization	qcache_use_rate	%	Query cache utilization
Table Lock Waits	table_locks_waited	Counts/second	The number of times that a request for a table lock could not be granted immediately and a wait was needed
Temp Tables	created_tmp_tables	Counts/second	The number of internal temporary tables created by the server while executing statements
InnoDB cache hit rate	innodb_cache_hit_rate	%	InnoDB engine cache hit rate
InnoDB cache utilization	innodb_cache_use_rate	%	InnoDB engine cache utilization
InnoDB Disk Reads	innodb_os_file_reads	Counts/second	The total number of file reads performed by read threads within InnoDB.
InnoDB Disk Writes	innodb_os_file_writes	Counts/second	The total number of file writes performed by write threads within InnoDB.
InnoDB fsync() Calls	innodb_os_fsyncs	Counts/second	The number of calls of the fsync function by InnoDB per second
InnoDB Opened Files	innodb_num_open_files	-	The number of files InnoDB currently holds open
MyISAM Cache Hit Rate	key_cache_hit_rate	%	MyISAM engine cache hit rate
MyISAM Cache Utilization	key_cache_use_rate	%	MyISAM engine cache utilization
CPU Utilization	cpu_use_rate	%	If overuse of idle resources is permitted, the CPU utilization may exceed 100%.
Memory Utilization	memory use rate	%	If overuse of idle resources is permitted, the memory utilization may exceed 100%.
Used Memory	memory_use	MB	If overuse of idle resources is permitted, the used memory may exceed the purchased specification.
Temp Files	created_tmp_files	Counts/second	The number of temp files created per second
Open Tables	opened_tables	-	Instance-level metric
COMMIT Statements	com_commit	Counts/second	The number of COMMIT statements executed
ROLLBACK Statements	com_rollback	Counts/second	The number of ROLLBACK statements executed
Created Threads	threads_created	-	The number of threads created to handle connections
Running Threads	threads_running	-	The number of threads that are not sleeping
Max. Connections	max_connections	-	The maximum number of connections
Temp Disk Tables	created_tmp_disk_tables	Counts/second	The number of internal on-disk temporary tables created by the server while executing statements
Requests to Read Next Row	handler_read_rnd_next	Counts/second	The number of requests to read the next row in the data file.
Internal ROLLBACK Statements	handler_rollback	Counts/second	The number of requests for a storage engine to perform a rollback operation
Internal COMMIT Statements	handler_commit	Counts/second	The number of internal COMMIT statements executed
InnoDB Free Pages	innodb_buffer_pool_pages_free	-	The number of free pages in the InnoDB buffer pool
Total InnoDB Pages	innodb_buffer_pool_pages_total	-	The total size of the InnoDB buffer pool in pages
InnoDB Logical Reads	innodb_buffer_pool_read_requests	Counts/second	The number of logical read requests
InnoDB Physical Reads	innodb_buffer_pool_reads	Counts/second	The number of logical reads that InnoDB could not satisfy from the buffer pool, and had to read directly from disk
Data Read in InnoDB	innodb_data_read	Bytes/second	The amount of data read per second
Total InnoDB Reads	innodb_data_reads	Counts/second	The total number of data reads per second
Total InnoDB Writes	innodb_data_writes	Counts/second	The total number of data writes per second
Data Written in InnoDB	innodb_data_written	Bytes/second	The amount of data written per second
InnoDB Rows Deleted	innodb_rows_deleted	Counts/second	The number of rows deleted from InnoDB tables
InnoDB Rows Inserted	innodb_rows_inserted	Counts/second	The number of rows inserted into InnoDB tables
InnoDB Rows Updated	innodb_rows_updated	Counts/second	The number of rows updated in InnoDB tables
InnoDB Rows Read	innodb_rows_read	Counts/second	The number of rows read from InnoDB tables
Avg. Time to Acquire an InnoDB Row Lock	innodb_row_lock_time_avg	Milliseconds	The average time to acquire a row lock for InnoDB tables
InnoDB Row Lock Waits	innodb_row_lock_waits	Counts/second	The number of times operations on InnoDB tables had to wait for a row lock
Unused Blocks in Key Cache	key_blocks_unused	-	The number of unused blocks in the MyISAM key cache
Used Blocks in Key Cache	key_blocks_used	-	The number of used blocks in the MyISAM key cache
Blocks Read from Key Cache	key_read_requests	Counts/second	The number of requests to read a key block from the MyISAM key cache
Blocks Read from Disk	key_reads	Counts/second	The number of physical reads of a key block from disk into the MyISAM key cache
Blocks Written into Key Cache	key_write_requests	Counts/second	The number of requests to write a key block to the MyISAM key cache.
Blocks Written into Disk	key_writes	Counts/second	The number of physical writes of a key block from the MyISAM key cache to disk.
Source-Replica Delay (in MB)	master_slave_sync_distance	MB	The amount of data by which the replica has lagged behind the source
Source-Replica Delay (in Seconds)	seconds_behind_master	Seconds	The amount of time by which the replica has lagged behind the source
IO Thread Status	slave_io_running	Status value (0: yes, 1: no, 2: connecting)	IO thread status
SQL Thread Status	slave_sql_running	Status value (0: yes, 1: no)	SQLA thread status