Monitoring items and metrics supported by the basic monitoring, advanced monitoring, and performance trends features - ApsaraDB for MongoDB

This topic describes the monitoring items and metrics supported by the basic monitoring, advanced monitoring, and performance trends features of ApsaraDB for MongoDB.

Basic monitoring

Note

The monitoring items vary based on the instance architecture. The monitoring items displayed in the ApsaraDB for MongoDB console shall prevail.

Instance architecture	Monitoring item	Metric and description
Standalone instance Replica set instance Sharded cluster instance	CPU utilization	cpu_usage: the CPU utilization of the instance.
	Memory usage	mem_usage: the memory usage of the instance.
	Used disk space	The total disk space occupied by the instance. Related metrics: ins_size: the total disk space occupied by the instance. data_size: the disk space occupied by data files. log_size: the disk space occupied by log files.
	cursors	The number of cursors used by the instance. Related metrics: total_open: the number of open cursors. timed_out: the number of cursors that timed out.
	Read and write queue length	The total number of operations queued waiting for a global lock. Related metrics: gl_cq_readers: the number of operations queued waiting for a global read lock. gl_cq_writers: the number of operations queued waiting for a global write lock. gl_cq_total: the number of operations queued waiting for a global read and write lock.
	Number of WiredTiger transactions	The number of and the available number of concurrent read and write transactions allowed into the WiredTiger storage engine. Related metrics: write_concurrent_trans_out: the number of concurrent write transactions allowed into the WiredTiger storage engine. read_concurrent_trans_out: the number of concurrent read transactions allowed into the WiredTiger storage engine. write_concurrent_trans_available: the available number of concurrent write transactions allowed into the WiredTiger storage engine. read_concurrent_trans_available: the available number of concurrent read transactions allowed into the WiredTiger storage engine.
	I/O latency	iocheck_cost: indicates how fast the current I/O system responds. Note This metric is supported only for standalone instances and replica set instances that run MongoDB 4.0 or earlier.
	WiredTiger	The cache metrics of the WiredTiger engine used by the instance. Related metrics: bytes_read_into_cache: the amount of data that is read into the cache. bytes_written_from_cache: the amount of data that is written from the cache to the disk. maximum_bytes_configured: the maximum available disk size that you can configure.
Standalone instance Replica set instance Sharded cluster instance Serverless instance	Disk usage	disk_usage: the proportion of the total disk space occupied by the instance to the maximum disk space available.
	QPS	The queries per second (QPS) of the instance. Related metrics: insert: the number of insert operations. query: the number of query operations. update: the number of update operations. delete: the number of delete operations. getmore: the number of read operations. command: the number of command operations.
	Connections	The number of connections to the instance. Related metrics: current_conn: the total number of connections to the instance. active_conn: the number of active connections to the instance.
	Network traffic	The network traffic of the instance. Related metrics: bytes_in: the inbound network traffic. bytes_out: the outbound network traffic. num_requests: the number of requests that are processed.
Replica set instance Sharded cluster instance	IOPS usage	The IOPS usage of the instance. Related metrics: data_iops: the IOPS usage of the data disk. log_iops: the IOPS usage of the log disk. Note The IOPS usage and IOPS utilization of a replica set or sharded cluster instance that runs MongoDB 4.2 and uses cloud disks cannot be queried.
	IOPS utilization	iops_usage: the proportion of the IOPS used by the instance to the maximum available IOPS. Note The IOPS usage and IOPS utilization of a replica set or sharded cluster instance that runs MongoDB 4.2 and uses cloud disks cannot be queried.
	Primary/secondary latency	repl_lag: the latency in data synchronization between the primary and secondary nodes of the instance. Note The latency between the primary and secondary nodes of a replica set instance and the latency between the shard and Configserver nodes of a sharded cluster instance can be -2 to 2 seconds. This is because the heartbeat interval between replica set instances is 2 seconds.
Serverless instance	Number of requests	sl_qps: the number of requests per second.

Advanced monitoring

Note

The advanced monitoring feature is supported only by standalone, replica set, and sharded cluster instances.
The monitoring items vary based on the instance architecture. The monitoring items displayed in the ApsaraDB for MongoDB console shall prevail.

Instance architecture	Monitoring item	Metric and description
Standalone instance Replica set instance Sharded cluster instance	CPU utilization	cpu_usage: the CPU utilization of the instance.
	Memory usage	mem_usage: the memory usage of the instance.
	Used disk space	The total disk space occupied by the instance. Related metrics: ins_size: the total disk space occupied by the instance. data_size: the disk space occupied by data files. log_size: the disk space occupied by log files.
	Connection utilization	conn_usage: the proportion of the current connections to the maximum connections.
	cursors	The number of cursors used by the instance. Related metrics: open_no_timedout: the number of open cursors that have timeout disabled. open_with_timedout: the number of open cursors that have timeout enabled. open_total: the number of open cursors. closed_by_timeout: the number of cursors that are closed due to timeout. closed_by_session: the number of cursors that are closed due to terminated sessions.
	Read and write queue length	The total number of operations queued waiting for a global lock. Related metrics: currentQueue_total: the number of operations queued waiting for a global read and write lock. currentQueue_readers: the number of operations queued waiting for a global read lock. currentQueue_writers: the number of operations queued waiting for a global write lock. active_readers: the number of operations that trigger a global read lock. active_writers: the number of operations that trigger a global write lock.
	Average response time	avgRT: the average response time of the instance.
	Number of WiredTiger transactions	The number of and the available number of concurrent read and write transactions allowed into the WiredTiger storage engine. Related metrics: write_concurrent_trans_out: the number of concurrent write transactions allowed into the WiredTiger storage engine. read_concurrent_trans_out: the number of concurrent read transactions allowed into the WiredTiger storage engine. write_concurrent_trans_available: the available number of concurrent write transactions allowed into the WiredTiger storage engine. read_concurrent_trans_available: the available number of concurrent read transactions allowed into the WiredTiger storage engine.
	I/O latency	iocheck_cost: indicates how fast the current I/O system responds.
	WiredTigerCache usage	The cache usage of the WiredTiger engine for the instance. Related metrics: wt_cache_usage: the cache usage of the WiredTiger engine for the instance. wt_cache_dirty_usage: the dirty cache usage of the WiredTiger engine.
	Number of affected documents	The number of documents of the instance. Related metrics: deletedDocuments: the number of documents that are deleted. insertedDocuments: the number of documents that are inserted. returnDocuments: the number of documents that are returned. updatedDocuments: the number of documents that are updated.
	Operation details	The details of the operations performed on the instance. Related metrics: scanAndOrder: the number of requests that cannot be sorted by index. writeConflicts: the number of write conflicts that occur. exactIDCount: the number of requests that are sent to query matched IDs by using broadcast.
	Number of indexes and documents that are queried and scanned	The number of indexes and documents that are queried and scanned. Related metrics: scannedKeys: the number of indexes that are queried and scanned. scannedDocs: the number of documents that are queried and scanned.
	TTL	The TTL operation performed on the instance. Related metrics: ttlDeletedDocuments: the number of documents that are deleted due to TTL indexes. ttlPasses: the number of times that the TTL thread deletes expired documents in the background.
Standalone instance Replica set instance Sharded cluster instance Serverless instance	Disk usage	disk_usage: the proportion of the total disk space occupied by the instance to the maximum disk space available. Note The metric is not supported for serverless instances.
	QPS	The queries per second (QPS) of the instance. Related metrics: insert: the number of insert operations. query: the number of query operations. update: the number of update operations. delete: the number of delete operations. getmore: the number of read operations. command: the number of command operations. Note The metric is not supported for serverless instances.
	Connections	The number of connections to the instance. Related metric: current_conn: the total number of connections to the instance. active_conn: the number of active connections to the instance. Note The metric is not supported for serverless instances.
	Network traffic	The network traffic of the instance. Related metrics: bytes_in: the inbound network traffic. bytes_out: the outbound network traffic. Note The metric is not supported for serverless instances.
Replica set instance Sharded cluster instance	IOPS usage	The IOPS usage of the instance. Related metrics: data_iops: the IOPS usage of the data disk. log_iops: the IOPS usage of the log disk. Note The IOPS usage and IOPS utilization of a replica set or sharded cluster instance that runs MongoDB 4.2 and uses cloud disks cannot be queried.
	IOPS utilization	iops_usage: the proportion of the IOPS used by the instance to the maximum available IOPS. Note The IOPS usage and IOPS utilization of a replica set or sharded cluster instance that runs MongoDB 4.2 and uses cloud disks cannot be queried.
	Primary/secondary latency	repl_lag: the latency in data synchronization between the primary and secondary nodes of the instance. Note The latency between the primary and secondary nodes of a replica set instance and the latency between the shard and Configserver nodes of a sharded cluster instance can be -2 to 2 seconds. This is because the heartbeat interval between replica set instances is 2 seconds.
Sharded cluster instance	Total QPS of shard nodes in a sharded cluster instance	Total QPS of shard nodes in the instance. Related metrics: insert_primary: the number of insert operations performed on the primary node. query_primary: the number of query operations performed on the primary node. update_primary: the number of update operations performed on the primary node. delete_primary: the number of delete operations performed on the primary node. getmore_primary: the number of read operations performed on the primary node. command_primary: the number of command operations performed on the primary node. insert_secondary: the number of insert operations performed on the secondary node. query_secondary: the number of query operations performed on the secondary node. insert_hidden: the number of insert operations performed on the hidden node. query_hidden: the number of query operations performed on the hidden node. update_hidden: the number of update operations performed on the hidden node. delete_hidden: the number of delete operations performed on the hidden node. getmore_hidden: the number of read operations performed on the hidden node. command_hidden: the number of command operations performed on the hidden node.
	Total QPS of mongos nodes in a sharded cluster instance	Total QPS of mongos nodes in the instance. Related metrics: insert: the number of insert operations. query: the number of query operations. update: the number of update operations. delete: the number of delete operations. getmore: the number of read operations. command: the number of command operations.
	moveChunk	The number of moveChunk operations performed on the current node that serves as the source or destination node. Related metrics: donorCount: the number of moveChunk operations performed on the current node that serves as the source node. recipientCount: the number of moveChunk operations performed on the current node that serves as the destination node.

Performance trends

Note

The monitoring items vary based on the instance architecture. The monitoring items displayed in the ApsaraDB for MongoDB console shall prevail.

Instance architecture	Monitoring item	Metric and description
Standalone instance Replica set instance Sharded cluster instance	CPU utilization	mongodb.cpu_usage: the CPU utilization of the instance.
	Memory usage	mongodb.mem_usage: the memory usage of the instance.
	Used disk space	The performance trends of the disk space used by the instance. Related metrics: mongodb.detailed_space_usage.ins_size: the total disk space used by the instance. mongodb.detailed_space_usage.data_size: the disk space used by data files. mongodb.detailed_space_usage.log_size: the disk space used by log files.
	cursors	The performance trends of cursors of the instance. Related metrics: mongodb.metrics.cursor.timedOut: the number of cursors that timed out. mongodb.metrics.cursor.open.total: the number of open cursors.
	Number of WiredTiger transactions	The performance trends of read and write requests and available concurrent requests. Related metrics: mongodb.wiredTiger.write_concurrent_trans_out: the number of concurrent write requests. mongodb.wiredTiger.read_concurrent_trans_out: the number of concurrent read requests. mongodb.wiredTiger.write_concurrent_trans_available: the number of concurrent write requests that are available. mongodb.wiredTiger.read_concurrent_trans_available: the number of concurrent read requests that are available. Note These metrics are supported only for replica set instances of the standard edition.
	WiredTiger	The performance trends of the cache metrics of the WiredTiger engine used by the instance. Related metrics: mongodb.wiredTiger.cache.pages.read.into.cache: the amount of data that is read into the cache. mongodb.wiredTiger.cache.pages.written.from.cache: the amount of data that is written from the cache to the disk. mongodb.wiredTiger.cache.maximum.bytes.configured: the maximum available disk size that you can configure. Note These metrics are supported only for replica set instances of the standard edition.
Standalone instance Replica set instance Sharded cluster instance Serverless instance	Disk usage	mongodb.disk_usage: the disk usage of the instance.
	QPS	The performance trends of the QPS of the instance. Related metrics: mongodb.opcounters.insert: the number of insert operations. mongodb.opcounters.query: the number of query operations. mongodb.opcounters.update: the number of update operations. mongodb.opcounters.delete: the number of delete operations. mongodb.opcounters.getmore: the number of read operations. mongodb.opcounters.command: the number of command operations.
	Connections	mongodb.connections.current: the total number of connections to the instance.
	Network traffic	The performance trends of the network traffic of the instance. Related metrics: mongodb.network.bytesIn: the inbound network traffic. mongodb.network.bytesOut: the outbound network traffic. mongodb.network.numRequests: the number of requests that are processed.
Replica set instance Sharded cluster instance	IOPS usage	The performance trends of the IOPS usage of the instance. Related metrics: mongodb.iops.data_iops: the number of read and write operations performed on the data disk. mongodb.iops.log_iops: the number of read and write operations performed on the log disk.
Replica set instance Sharded cluster instance	IOPS utilization	mongodb.iops_usage: the IOPS utilization of the instance. Note The IOPS usage and IOPS utilization of a replica set or sharded cluster instance that runs MongoDB 4.2 and uses cloud disks cannot be queried.
Replica set instance	GlobalLock	Th performance trends of a global lock of the instance. Related metrics: mongodb.globalLock.currentQueue.total: the number of operations queued waiting for a global read and write lock. mongodb.globalLock.currentQueue.writers: the number of operations queued waiting for a global write lock. mongodb.globalLock.currentQueue.readers: the number of operations queued waiting for a global read lock.