Supported Monitoring Metrics
Last updated:2026-02-11 14:40:26
TDSQL-C for MySQL provides monitoring metrics from both the instance dimension and the database proxy node dimension.
Monitored metrics
Note:
You can view the following monitoring metrics in the console or through TencentCloud API as instructed in Viewing Monitoring Data.
Parallel query monitoring metrics are supported only for TDSQL-C for MySQL 8.0 (kernel version 3.1.8) or later.
The monitoring metrics related to columnar indexes are supported by TDSQL-C for MySQL 8.0 kernel version 3.1.14.
For monitoring metrics related to the read-only analysis engine, see Viewing Monitoring Data.
Secondary cache monitoring metrics are supported in TDSQL-C for MySQL 5.7 kernel version 2.1.13.002 and above, and 8.0 kernel version 3.1.15.002 and above.
When retrieving monitoring data via API calls, refer to the Callable Metric Name column in the table below for the monitoring metric name to be used in the call parameters.
Monitoring Metrics at the Instance Dimension
Category | Monitoring Metric Name | Callable Metric Name | Unit | Data Aggregation Method | |
Resource Monitoring | | CPU Utilization | CpuUseRate | % | MAX |
| | Memory Utilization | MemoryUseRate | % | MAX |
| | Memory Usage | MemoryUse | MB | MAX |
| | Primary storage utilization | StorageUseRate | % | MAX |
| | Primary storage usage | StorageUse | GB | MAX |
| | Secondary storage usage | ArchiveStorageUsed | GB | MAX |
| | Data Table Space Usage | DataVolumeUsage | GB | MAX |
| | Temporary Table Space Usage | TmpVolumeUsage | GB | MAX |
| | Undo Table Space Usage | UndoVolumeUsage | GB | MAX |
| | CCU | Ccu | Individual | MAX |
| | Total Client Traffic Sent Per Second | BytesSent | MB/s | MAX |
| | Total Client Traffic Received Per Second | BytesReceived | MB/s | MAX |
| | Read Request IOPS | Readiops | per second | MAX |
| | Write Request IOPS | Writeiops | per second | MAX |
| | Total IOPS | Iops | per second | MAX |
| | Read I/O Throughput | ReadBandwidth | MB/s | MAX |
| | Write I/O Throughput | WriteBandwidth | MB/s | MAX |
| | Total I/O Throughput | IoBandwidth | MB/s | MAX |
| | Storage Space Occupied by Log | LogCapacity | GB | MAX |
| | Write status of a read-write instance | ReadOnly | 0: normal 1: abnormal | MAX |
Engine Monitoring | Connection | Operations Executed per Second | Qps | per second | MAX |
| | Transactions Executed per Second | Tps | per second | MAX |
| | Connection Utilization Rate | ConnectionUseRate | % | MAX |
| | Maximum Number of Connections | MaxConnections | Individual | MAX |
| | Currently Open Connections | ThreadsConnected | Individual | MAX |
| | Number of Threads Created | ThreadsCreated | Individual | SUM |
| | Number of Threads Running | ThreadsRunning | Individual | MAX |
| Access | Slow Queries | SlowQueries | Individual | SUM |
| | Full-Table Scans | SelectScan | Individual | SUM |
| | Number of Queries | ComSelect | Individual | SUM |
| | Number of Updates | ComUpdate | Individual | SUM |
| | Number of Deletions | ComDelete | Individual | SUM |
| | Number of Insertions | ComInsert | Individual | SUM |
| | Number of Overwrite | ComReplace | Individual | SUM |
| | Total Number of Requests | Queries | Individual | SUM |
| | Number of Submissions | ComCommit | Individual | SUM |
| | Number of Rollbacks | ComRollback | Individual | SUM |
| | Number of Full-Table Scan Composite Queries | SelectFullJoin | Individual | SUM |
| | Number of Range Scan Composite Queries | SelectFullRangeJoin | Individual | SUM |
| | Number of Sort Merge Passes | SortMergePasses | Individual | SUM |
| | Qcache Hit Rate | QcacheHitRate | % | MIN |
| | Qcache Utilization Rate | QcacheUseRate | % | MIN |
| | Number of Multi-Table Updates | ComUpdateMulti | Individual | SUM |
| | Number of Multi-Table Deletions | ComDeleteMulti | Individual | SUM |
| Table | Number of Temporary Tables | CreatedTmpTables | Individual | SUM |
| | Number of Times Waiting for Table Lock | TableLocksWaited | Individual | SUM |
| | Number of Tables Already Opened | OpenedTables | Individual | MAX |
| | Number of Table Locks Released Immediately | TableLocksImmediate | Individual | SUM |
| | Number of Table Open Cache Hits | TableOpenCacheHits | Individual | SUM |
| | Number of Table Open Cache Misses | TableOpenCacheMisses | Individual | SUM |
| InnoDB | InnoDB Engine Cache Hit Rate | InnodbCacheHitRate | % | MIN |
| | InnoDB Engine Cache Utilization | InnodbCacheUseRate | % | MIN |
| | Number of Disk Reads | InnodbOsFileReads | Individual | MAX |
| | Number of Disk Writes | InnodbOsFileWrites | Individual | MAX |
| | Number of InnoDB_fsyncs | InnodbOsFsyncs | Individual | MAX |
| | Number of InnoDB Open Tables | InnodbNumOpenFiles | Individual | MAX |
| | Number of InnoDB Reads | InnodbDataRead | Byte | SUM |
| | Total Number of InnoDB Reads | InnodbDataReads | Individual | SUM |
| | Total Number of InnoDB Writes | InnodbDataWrites | Individual | SUM |
| | Number of InnoDB Writes | InnodbDataWritten | Byte | SUM |
| | Number of InnoDB Rows Deleted | InnodbRowsDeleted | Individual | SUM |
| | Number of InnoDB Rows Inserted | InnodbRowsInserted | Individual | SUM |
| | Number of InnoDB Rows Updated | InnodbRowsUpdated | Individual | SUM |
| | Number of InnoDB Rows Read | InnodbRowsRead | Individual | SUM |
| | Average InnoDB Row Lock Acquiring Time | InnodbRowLockTimeAvg | Milliseconds | MAX |
| | Number of InnoDB Row Lock Waits | InnodbRowLockWaits | Individual | SUM |
| | Number of InnoDB Dirty Pages | InnodbBufferPoolPagesDirty | Individual | MAX |
| | Number of InnoDB Pending Writes | InnodbDataPendingWrites | Individual | MAX |
| | Number of InnoDB Pending Reads | InnodbDataPendingReads | Individual | MAX |
| | Number of InnoDB Log Wait Writes | InnodbLogWaits | Individual | SUM |
| | Number of InnoDB Log Physical Writes | InnodbLogWrites | Individual | SUM |
| | Number of InnoDB Log Physical Write Requests | InnodbLogWriteRequests | Individual | SUM |
| Tmp | Number of Temporary Tables | CreatedTmpDiskTables | Individual | SUM |
| | Number of Temporary Files | CreatedTmpFiles | Individual | SUM |
| Handler | Number of Requests to Read Next Line | HandlerReadRndNext | Individual | SUM |
| | Number of Internal Rollbacks | HandlerRollback | Individual | SUM |
| | Number of Internal Submissions | HandlerCommit | Individual | SUM |
| Buffer | Number of InnoDB Empty Pages | InnodbBufferPoolPagesFree | Individual | MAX |
| | Total Number of InnoDB Pages | InnodbBufferPoolPagesTotal | Individual | MAX |
| | Number of InnoDB Logical Reads | InnodbBufferPoolReadRequests | Individual | SUM |
| | Number of InnoDB Physical Reads | InnodbBufferPoolReads | Individual | SUM |
| | Number of InnoDB Buffer Pool Writes | InnodbBufferPoolWriteRequest | Individual | SUM |
| Parallel Query | Number of Parallelly Queried Threads | TxsqlParallelThreadsCurrentlyUsed | Individual | MAX |
| | Number of Failed Parallel Queries | TxsqlParallelStmtError | Individual | SUM |
| | Number of Executed Parallel Queries | TxsqlParallelStmtExecuted | Individual | SUM |
| | Number of Rollback Serial Queries | TxsqlParallelStmtFallback | Individual | SUM |
| Column Store Index | Number of Column Store Index Execution Statements | InnodbCsiSelectPushdown | Individual | SUM |
| | Number of Column Store Index Rollback Statements | InnodbCsiSelectFallback | Individual | SUM |
| | Column Store Index Synchronization Delay | InnodbCsiLagSeconds | Second | MAX |
| | Column Index Storage Usage | InnodbCsiDiskUsage | GB | MAX |
| | Number of CSI Creation Failures | InnodbCsiCreateFailed | Individual | SUM |
| Secondary cache | BP pages read from secondary cache | InnodbNcdbScReadFromSc | Individual | SUM |
| | BP pages read from storage | InnodbNcdbScReadFromCbs | Individual | SUM |
| | Pages evicted from secondary cache | InnodbNcdbScWriteWithoutFlush | Individual | SUM |
| | Pages invalidated in secondary cache | InnodbNcdbScFreeByLru | Individual | SUM |
| | Pages used in secondary cache | InnodbNcdbScUsedPageNum | Individual | SUM |
| | Secondary cache storage usage | InnodbNcdbScUsage | GB | MAX |
| | Secondary cache storage utilization | InnodbNcdbScUseRate | % | MAX |
| Other | Total Number of Open Files | OpenFiles | Individual | MAX |
| | Number of Queried Cache Hits | QcacheHits | Individual | SUM |
| | Number of Cache Searches | QcacheSearchTimes | Individual | SUM |
| | Total Number of Cache Queries | QcacheTotalTimes | Individual | SUM |
| | Query Remaining Cache Space | QcacheFreeMemory | MB | MAX |
| | Number of Successful Attempts to Cache Result Sets | QcacheInserts | Individual | SUM |
| | Number of Failed Attempts to Cache Result Sets | QcacheNotCached | Individual | SUM |
| | Number of Cache Queries | QcacheQueriesInCache | Individual | SUM |
| | Number of Deprecated Queries | QcacheLowmemPrunes | Individual | SUM |
Deployment Monitoring | | Replication Status | ReplicationStatus | 0-Yes,1-No | If there is 1, take 1. |
| | Replication Delay | ReplicationDelay | Milliseconds | MAX |
| | Replication Lag in lsn Distance | ReplicationDelayDistance | Bytes | MAX |
| | Cross-AZ primary/secondary synchronization gap | MasterSlaveSyncDistance | MB | MAX |
| | IO thread status for cross-AZ deployment | ClusterSlaveIoRunning | 0 - Yes 1 - No 2 - Connecting | MAX |
| | SQL thread status for cross-AZ deployment | ClusterSlaveSqlRunning | 0 - Yes 1 - No | MAX |
| | Cross-AZ latency | SecondsBehindMaster | Second | MAX |
| | Transaction wait time for cross-AZ cluster acknowledgment | RplSemiSyncMasterTxCurWaitTime | Microsecond | MAX |
Monitoring Metrics at Database Proxy Node Dimension
To obtain monitoring metrics at the database proxy node dimension, you need to first enable the database proxy. For detailed metric descriptions, refer to Viewing Database Proxy Monitoring Data.
Performance monitoring data analysis
You can set alarm policies to monitor and analyze performance metrics. Below are some examples:
CPU utilization: It describes how the CPU in the database server is used while running. If the CPU load is high, there will be problems such as slow response of the database system and query delay. Therefore, you can set a threshold alarm for this monitoring metric as needed. Generally, if the CPU utilization exceeds 90%, you need to check the request volume and query performance, and make timely adjustments. If the CPU utilization meets the application or database requirements such as throughput and concurrency and is expected, then it is appropriate.
Used storage space: If the used storage space is always greater than or equal to 85% of the total disk space, you need to check how the storage space is used and whether the data can be deleted from the instance or archived to another system to free up space. If the amount of stored data is too large, the disk is about to be full, and you can solve the problem of insufficient disk space through capacity expansion.
Traffic inbound to/outbound from client: You can check the fluctuations of these metrics in Resource Monitoring to determine whether the problem is caused by the business peak or database, and then optimize the business or adjust the database configuration.
Max connections:
max_connections is one of the parameters in TDSQL-C for MySQL, which is used to specify the maximum number of connections allowed by the database. When the database process opens a new connection, the number of currently connected clients will be queried. If the number has reached the set value of max_connections, a new connection cannot be established. If there is a large number of connections, the instance performance drops, and the response time increases, you need to consider restricting the maximum number of connections to the database. If the value of max_connections is set too high, it may cause problems such as insufficient memory and thread crash in the system. Therefore, you should carefully evaluate your application needs and set an appropriate value. The value generally falls between 200 and 500. You can also adjust the maximum number of connections based on the connection conditions of applications and databases.When a performance metric exceeds the configured alarm threshold, you may need to modify relevant parameters to optimize the database to improve its performance and availability.
Was this page helpful?
You can also Contact Sales or Submit a Ticket for help.
Yes
No
Feedback