Overview of metrics|V4.6.0|OceanBase Database| docs|Distributed Database

This topic describes the key metrics of OceanBase Database and its core components, including clusters, tenants, and hosts.

Monitor item query

You can query the definition information of all monitor items in the sys tenant by using an SQL statement. Here is an example:

obclient> SELECT STAT_ID, CLASS, NAME FROM V$STATNAME;

Many monitor items are associated with internal implementations. Therefore, they may be changed or deleted. For details about how to query monitor items of specific categories, see Monitor items.

Monitoring item categories

OceanBase Database V4.x supports 15 categories of monitoring items, covering more than 600 monitoring items, including network, request queues, transactions, SQL, cache, storage, resource quotas, and logs. The following table lists the monitoring item categories.

Monitoring item category ID	Category	Description
1	NETWORK	Network
2	QUEUE	Request queue
4	TRANS	Transaction
8	SQL	SQL
16	CACHE	Cache
32	STORAGE	Storage
64	RESOURCE	Resource
128	DEBUG	Log
256	CLOG	CLOG
512	ELECT	Election
1024	OBSERVER	System
2048	RS	Root Service
3072	Time Model	Time model
4096	TABLEAPI	TableAPI
8192	WR	WR

Network

Metric ID	Metric Type	Metric Name	Description
10000	1	rpc packet in	The total number of RPC packets received by the tenant. The count increases by 1 when the tenant receives an RPC request or receives a response after sending an RPC request.
10001	1	rpc packet in bytes	The total number of bytes of RPC packets received by the tenant, in Bytes. The count increases by the size of the RPC request or response when the tenant receives an RPC request or receives a response after sending an RPC request.
10002	1	rpc packet out	The total number of RPC packets sent by the tenant. The count increases by 1 when the tenant sends an RPC request or sends a response after receiving an RPC request.
10003	1	rpc packet out bytes	The total number of bytes of RPC packets sent by the tenant, in Bytes. The count increases by the size of the RPC request or response when the tenant sends an RPC request or sends a response after receiving an RPC request.
10004	1	rpc deliver fail	The total number of failures when an RPC request is forwarded to the tenant's work queue. The count increases by 1 when an RPC request fails to be forwarded to the tenant's work queue.
10005	1	rpc net delay	The total network transmission delay of RPC requests received by the tenant, in microseconds. When an RPC IO thread parses an RPC request, the current timestamp is obtained and subtracted from the timestamp obtained from the packet header to calculate the network transmission delay, which is then accumulated to RPC_NET_DELAY.
10006	1	rpc net frame delay	The total network frame processing time of RPC requests received by the tenant, in microseconds. When an RPC IO thread forwards an RPC request to the tenant's work queue, the current timestamp is obtained and subtracted from the timestamp recorded when the RPC packet is parsed to calculate the network frame processing time, which is then accumulated to RPC_NET_FRAME_DELAY.
10007	1	mysql packet in	The total number of SQL request packets received by the tenant. The count increases by 1 when a complete SQL request packet is decoded by the network thread.
10008	1	mysql packet in bytes	The total number of bytes of SQL request packets received by the tenant, in Bytes. The size of the original data packets received by the network thread is obtained based on the protocol rules.
10009	1	mysql packet out	The total number of SQL response packets sent by the tenant. The count increases by 1 when a complete response packet is encoded and sent by the network thread.
10010	1	mysql packet out bytes	The total number of bytes of SQL response packets sent by the tenant, in Bytes. The size of the encoded response result set is accumulated.
10011	1	mysql deliver fail	The total number of failures when SQL login requests are added to the login queue or other requests are added to the work thread queue. The count increases by 1 when the login request thread queue is full for login requests, or when the tenant's work thread queue is full for other requests.
10012	1	rpc compress original packet cnt	The total number of RPC requests sent by the tenant that specify the need for compression. The count increases by 1 when the compressor_type is set for an RPC request.
10013	1	rpc compress compressed packet cnt	The total number of compressed RPC requests sent by the tenant. The count increases by 1 when the compressor_type is set for an RPC request and the compression is successful (the compressed size is smaller than the original size).
10014	1	rpc compress original size	The total original size of RPC requests sent by the tenant that specify the need for compression, in Bytes. The original size of the RPC packet before compression is accumulated to the RPC_COMPRESS_ORIGINAL_SIZE metric when the compressor_type is set for an RPC request.
10015	1	rpc compress compressed size	The total compressed size of compressed RPC requests sent by the tenant, in Bytes. The compressed size of the RPC packet is accumulated to the RPC_COMPRESS_COMPRESSED_SIZE metric when the compressor_type is set for an RPC request and the compression is successful.
10016	1	rpc stream compress original packet cnt	Deprecated in V4.x. Retained only in the code, in times.
10017	1	rpc stream compress compressed packet cnt	Deprecated in V4.x. Retained only in the code, in times.
10018	1	rpc stream compress original size	Deprecated in V4.x. Retained only in the code, in Bytes.
10019	1	rpc stream compress compressed size	Deprecated in V4.x. Retained only in the code, in Bytes.
110000	1	standby fetch log bytes	The total size of data received by an OBServer node from a standby node during network-based synchronization. The value is in bytes. When an OBServer node receives data from a standby node during network-based synchronization, it accumulates the received data size to the STANDBY_FETCH_LOG_BYTES metric.
110001	1	standby fetch log bandwidth limit	The bandwidth limit allocated to an OBServer node during network-based synchronization. The value is in bytes. After network-based synchronization is enabled, the RS node allocates bandwidth to nodes in the cluster based on the set strategy. The OBServer node records the allocated bandwidth limit to the STANDBY_FETCH_LOG_BANDWIDTH_LIMIT metric.

Request queue

Metric ID	Metric type	Metric name	Description
20000	2	request enqueue count	The total number of requests that are enqueued. The count increases by 1 for each request that is enqueued.
20001	2	request dequeue count	The total number of requests that are dequeued. The count increases by 1 for each request that is dequeued.
20002	2	request queue time	The total time for which all requests are waiting in the queue, in microseconds. When a working thread of a tenant acquires a request, it calculates the waiting time by subtracting the time when the request was enqueued from the current time and accumulates the value in the REQUEST_QUEUE_TIME metric.

Transactions

Metric ID	Metric type	Metric name	Description
30000	4	trans commit log sync time	The cumulative time spent by the tenant on this server for a transaction to successfully commit to a majority of replicas, in microseconds. When a transaction is successfully committed to a majority of replicas, the current time is subtracted from the time when the commit was successful to calculate the time spent by the majority, which is then added to this metric.
30001	4	trans commit log sync count	The number of times a transaction has been successfully committed and completed by a majority of replicas. Each time a transaction is successfully committed by a majority of replicas, the count is incremented.
30002	4	trans commit log submit count	The number of times a transaction has been successfully committed. Each time a transaction is successfully committed, the count is incremented.
30005	4	trans start count	The number of transactions started by the tenant on this server. Each time a transaction is successfully started on the OBServer node where the session is created, the count is incremented.
30006	4	trans total used time	The cumulative time spent by the tenant on this server for transactions, in microseconds. Each time a transaction is completed on the OBServer node where the session is created, the time spent is added to this metric.
30007	4	trans commit count	The number of transactions successfully committed by the tenant on this server. Each time a transaction is successfully committed on the OBServer node where the session is connected, the count is incremented.
30008	4	trans commit time	The cumulative time spent by the tenant on this server for transaction commits, in microseconds. Each time a transaction is successfully committed on the OBServer node where the session is created, the commit time is added to this metric.
30009	4	trans rollback count	The number of transactions successfully rolled back by the tenant on this server. Each time a transaction is successfully rolled back on the OBServer node where the session is created, the count is incremented.
30010	4	trans rollback time	The cumulative time spent by the tenant on this server for transaction rollbacks, in microseconds. Each time a transaction is successfully rolled back on the OBServer node where the session is created, the rollback time is added to this metric.
30011	4	trans timeout count	The number of transactions that timed out by the tenant on this server. Each time a transaction times out on the OBServer node where the session is created, the count is incremented.
30012	4	trans local trans count	The number of single-node transactions completed by the tenant on this server (including both commit and rollback). Each time a transaction is completed on the OBServer node where the session is created, the number of participants is checked. If the number of participants is 1, the count is incremented.
30013	4	trans distribute trans count	The number of distributed transactions completed by the tenant on this server (including both commit and rollback). Each time a transaction is completed on the OBServer node where the session is created, the number of participants is checked. If the number of participants is greater than 1, the count is incremented.
30017	4	trans without participant count	The number of read-only transactions completed by the tenant on this server (including both commit and rollback). Each time a transaction is completed on the OBServer node where the session is created, the number of participants is checked. If the number of participants is 0, the count is incremented.
30018	4	redo log replay count	The number of times redo logs of transactions have been replayed within the tenant. Each time a redo log of a transaction is replayed, the count is incremented.
30019	4	redo log replay time	The cumulative time spent by the tenant on this server for replaying redo logs of transactions, in microseconds. Each time a redo log is replayed, the time spent for the single replay is added to this metric.
30020	4	prepare log replay count	The number of times prepare logs of transactions have been replayed within the tenant. Each time a prepare log of a transaction is replayed, the count is incremented.
30021	4	prepare log replay time	The cumulative time spent by the tenant on this server for replaying prepare logs of transactions, in microseconds. Each time a prepare log is replayed, the time spent for the single replay is added to this metric.
30022	4	commit log replay count	The number of times commit logs of transactions have been replayed within the tenant. Each time a commit log of a transaction is replayed, the count is incremented.
30023	4	commit log replay time	The cumulative time spent by the tenant on this server for replaying commit logs of transactions. Each time a commit log is replayed, the time spent for the single replay is added to this metric.
30024	4	abort log replay count	The number of times abort logs of transactions have been replayed within the tenant. Each time an abort log of a transaction is replayed, the count is incremented.
30025	4	abort log replay time	The cumulative time spent by the tenant on this server for replaying abort logs of transactions, in microseconds. Each time an abort log is replayed, the time spent for the single replay is added to this metric.
30026	4	clear log replay count	The number of Clear logs replayed in the current tenant. The count increases by 1 each time a Clear log is replayed.
30027	4	clear log replay time	The cumulative time spent replaying Clear logs in the current tenant, in microseconds. The time spent on each replay is recorded and added to this metric.
30053	4	gts request total count	The total number of GTS requests processed by the GTS server. This count increases by 1 each time the GTS server processes a GTS request.
30054	4	gts acquire total time	The total waiting time for asynchronous GTS acquisition, in microseconds. This metric is not yet implemented and therefore has no meaningful value.
30056	4	gts acquire total wait count	The total number of asynchronous GTS acquisitions. This count increases by 1 each time a GTS is acquired.
30063	4	gts wait elapse total time	The total waiting time for GTS pushes, in microseconds. This metric is not yet implemented and therefore has no meaningful value.
30065	4	gts wait elapse total wait count	The total number of GTS pushes. This metric is not yet implemented and therefore has no meaningful value.
30077	4	trans early lock release enable count	This metric is deprecated starting from V4.4.1.
30078	4	trans early lock release unable count	This metric is deprecated starting from V4.4.1.
30079	4	read elr row count	The total number of rows where early row lock release occurred during MemTable writes. This metric counts the number of rows successfully locked during early row lock release.
30080	4	local trans total used time	The cumulative time spent on local transactions completed on the current server, including both Commit and Rollback operations, in microseconds. This metric is incremented each time a transaction is completed on the server where the session was created, provided the number of participants is 1.
30081	4	distributed trans total used time	The cumulative time spent on distributed transactions completed on the current server, including both Commit and Rollback operations, in microseconds. This metric is incremented each time a transaction is completed on the server where the session was created, provided the number of participants is greater than 1.
30082	4	tx data hit mini cache count	The number of times TxData was read from the MINI Cache. The MINI Cache is the fastest cache for TxData, and a higher hit rate indicates better performance. (The MINI Cache is request-level, with one per Access Context.) This metric increments by 1 each time a TxData is read from the MINI Cache.
30083	4	tx data hit kv cache count	The number of times TxData was read from the KVCache. The KVCache performs slightly slower than the MINI Cache. Most read operations are expected to hit either the KVCache or the MINI Cache. This metric increments by 1 each time a TxData is read from the KVCache.
30084	4	tx data read tx ctx count	The number of times TxData was read from the transaction context table. This metric increments by 1 each time a TxData is read from the transaction context table.
30085	4	tx data read tx data memtable count	The number of times TxData was read from the Memtable of the transaction data table. This metric increments by 1 each time a TxData is read from the Memtable of the transaction data table.
30086	4	tx data read tx data sstable count	The number of times TxData was read from the SSTable of the transaction data table. (Reading TxData from the SSTable is the slowest read operation, so this metric is expected to be relatively low.) This metric increments by 1 each time a TxData is read from the SSTable of the transaction data table.
30200	4	xa start total count	The total number of XA_START statements executed for XA transactions, including both successful and failed executions. This count increases by 1 each time the XA_START logic is executed.
30201	4	xa start total used time	The total time spent executing XA_START statements for XA transactions, including both successful and failed executions, in microseconds. This metric increments by the local processing time after the XA_START logic is executed.
30202	4	xa start with rpc total count	The total number of successful remote XA_START statements executed for XA transactions. This count increases by 1 each time a remote XA_START statement is successfully executed.
30203	4	failed xa start total count	The total number of failed XA_START statements executed for XA transactions. This count increases by 1 each time an XA_START statement fails.
30204	4	xa end total count	This view displays the total number of times the XA_END statement is executed for XA transactions, including both successful and failed executions. The count is incremented after the XA_END logic is executed on the transaction side.
30205	4	xa end total used count	This view displays the total execution time of the XA_END statement for XA transactions, in microseconds. The local processing time is added after the XA_END logic is executed on the transaction side.
30206	4	xa end with rpc total count	This view displays the total number of successful remote executions of the XA_END statement for XA transactions. The count is incremented after the XA_END logic is executed on the transaction side, if the execution is remote and successful.
30207	4	failed xa end total count	This view displays the total number of failed executions of the XA_END statement for XA transactions. The count is incremented after the XA_END logic is executed on the transaction side, if an error occurs.
30208	4	xa prepare total count	This view displays the total number of times the XA_PREPARE statement is executed for XA transactions, including both successful and failed executions. The count is incremented after the XA_PREPARE logic is executed on the transaction side.
30209	4	xa prepare total used time	This view displays the total execution time of the XA_PREPARE statement for XA transactions, in microseconds. The local processing time is added after the XA_PREPARE logic is executed on the transaction side.
30210	4	xa prepare with rpc total count	This view displays the total number of successful remote executions of the XA_PREPARE statement for XA transactions. The count is incremented after the XA_START logic is executed on the transaction side, if the execution is remote and successful.
30211	4	failed xa prepare total count	This view displays the total number of failed executions of the XA_PREPARE statement for XA transactions. The count is incremented after the XA_PREPARE logic is executed on the transaction side, if an error occurs.
30212	4	xa commit total count	This view displays the total number of times the XA_COMMIT statement is executed for XA transactions, including both successful and failed executions. The count is incremented after the XA_COMMIT logic is executed on the transaction side.
30213	4	xa commit total used time	This view displays the total execution time of the XA_COMMIT statement for XA transactions, in microseconds. The local processing time is added after the XA_COMMIT logic is executed on the transaction side.
30214	4	xa commit with rpc total count	This view displays the total number of successful remote executions of the XA_COMMIT statement for XA transactions. The count is incremented after the XA_COMMIT logic is executed on the transaction side, if the execution is remote and successful.
30215	4	failed xa commit total count	This view displays the total number of failed executions of the XA_COMMIT statement for XA transactions. The count is incremented after the XA_COMMIT logic is executed on the transaction side, if an error occurs.
30216	4	xa rollback total count	This view displays the total number of times the XA_ROLLBACK statement is executed for XA transactions, including both successful and failed executions. The count is incremented after the XA_ROLLBACK logic is executed on the transaction side.
30217	4	xa rollback total used time	This view displays the total execution time of the XA_ROLLBACK statement for XA transactions, in microseconds. The local processing time is added after the XA_ROLLBACK logic is executed on the transaction side.
30218	4	xa rollback with rpc total count	This view displays the total number of successful remote executions of the XA_ROLLBACK statement for XA transactions. The count is incremented after the XA_START logic is executed on the transaction side, if the execution is remote and successful.
30219	4	failed xa rollback total count	This view displays the total number of failed executions of the XA_ROLLBACK statement for XA transactions. The count is incremented after the XA_ROLLBACK logic is executed on the transaction side, if an error occurs.
30220	4	started xa trans count	This view displays the total number of successfully started XA transactions. The count is incremented after a XA transaction is successfully started.
30221	4	read only xa trans total count	This view displays the total number of read-only XA transactions. The count is incremented when a read-only XA transaction is committed.
30222	4	xa trans with one phase commit total count	This view displays the total number of XA transactions that use one-phase commit. The count is incremented when a XA transaction is committed using one-phase commit.
30223	4	inner sql total count in xa statement	This view displays the total number of INNER SQL statements executed during the execution of XA logic. The count is incremented after each INNER SQL statement is executed.
30224	4	total count of inner sql (latency >= 10ms) in xa statement	This view displays the total number of INNER SQL statements executed during the execution of XA logic that take more than 10 ms. The count is incremented when an INNER SQL statement takes more than 10 ms.
30225	4	total count of inner sql (latency >= 20ms) in xa statement	This view displays the total number of INNER SQL statements executed during the execution of XA logic that take more than 20 ms. The count is incremented when an INNER SQL statement takes more than 20 ms.
30226	4	inner sql total used time in xa statement	The total time spent on INNER SQL statements executed during an XA transaction, measured in microseconds. When an INNER SQL statement is executed during an XA transaction, the time spent on that statement is added to this metric.
30227	4	inner rpc total count in xa statement	The total number of RPC calls made during an XA transaction. Each time an RPC call is made during an XA transaction, this count is incremented by 1.
30228	4	total count of inner rpc (latency >= 10ms) in xa statement	The total number of RPC calls with a latency of 10 ms or more during an XA transaction. Each time an RPC call with a latency of 10 ms or more is made during an XA transaction, this count is incremented by 1.
30229	4	total count of inner rpc (latency >= 20ms) in xa statement	The total number of RPC calls with a latency of 20 ms or more during an XA transaction. Each time an RPC call with a latency of 20 ms or more is made during an XA transaction, this count is incremented by 1.
30230	4	inner rpc total used time in xa statement	The total time spent on RPC calls during an XA transaction, measured in microseconds. When an RPC call is made during an XA transaction, the time spent on that call is added to this metric.
30231	4	dblink trans total count	The total number of DBLink transactions. This count is incremented by 1 each time a DBLink transaction is started.
30232	4	failed dblink trans total count	The total number of failed DBLink transactions. This count is incremented by 1 each time a DBLink transaction fails.
30233	4	dblink trans promotion total count	The total number of times a regular transaction is upgraded to a DBLink transaction. This count is incremented by 1 each time a regular transaction is upgraded to a DBLink transaction.
30234	4	dblink trans callback total count	The total number of times a reverse link transaction is called during a DBLink transaction. This count is incremented by 1 each time a reverse link transaction is called during a DBLink transaction.
30235	4	dblink trans commit total count	The total number of times a DBLink transaction is committed. This count is incremented by 1 each time a DBLink transaction is committed.
30236	4	dblink trans commit total used time	The total time spent on committing a DBLink transaction, measured in microseconds. When a DBLink transaction is committed, the time spent on the commit is added to this metric.
30237	4	failed dblink trans commit total count	The total number of failed DBLink transaction commits. This count is incremented by 1 each time a DBLink transaction commit fails.
30238	4	dblink trans rollback total count	The total number of times a DBLink transaction is rolled back. This count is incremented by 1 each time a DBLink transaction is rolled back.
30239	4	dblink trans rollback total used time	The total time spent on rolling back a DBLink transaction, measured in microseconds. When a DBLink transaction is rolled back, the time spent on the rollback is added to this metric.
30240	4	failed dblink trans rollback total count	The total number of failed DBLink transaction rollbacks. This count is incremented by 1 each time a DBLink transaction rollback fails.

SQL

Metric ID	Metric Type	Metric Name	Description
40000	8	sql select count	The number of SQL SELECT statements executed.
40001	8	sql select time	The total execution time of SQL SELECT statements, in microseconds.
40002	8	sql insert count	The number of INSERT statements issued by users.
40003	8	sql insert time	The total execution time of INSERT statements issued by users.
40004	8	sql replace count	The number of REPLACE statements issued by users.
40005	8	sql replace time	The total execution time of REPLACE statements issued by users.
40006	8	sql update count	The number of UPDATE statements issued by users.
40007	8	sql update time	The total execution time of UPDATE statements issued by users, in microseconds.
40008	8	sql delete count	The number of DELETE statements issued by users.
40009	8	sql delete time	The total execution time of DELETE statements issued by users, in microseconds.
40010	8	sql local count	The number of SQL LOCAL statements executed.
40011	8	sql remote count	The number of SQL REMOTE statements executed.
40012	8	sql distributed count	The number of SQL statements executed in a distributed manner.
40013	8	active sessions	The number of active sessions. The number of active sessions is incremented by 1 when a new connection is established and decremented by 1 when a connection is disconnected.
40014	8	single query count	The total number of statements sent using the non-multiple query protocol.
40015	8	multiple query count	The total number of statements sent using the multiple query protocol.
40016	8	multiple query with one stmt count	The number of statements sent using the multiple query protocol that contain only one SQL statement.
40018	8	sql other count	The number of executions of SQL statements other than SELECT, INSERT, REPLACE, UPDATE, DELETE, COMMIT, and ROLLBACK.
40019	8	sql other time	The total execution time of SQL statements other than SELECT, INSERT, REPLACE, UPDATE, DELETE, COMMIT, and ROLLBACK, in microseconds.
40020	8	ps prepare count	The number of times the PREPARE statement is executed.
40021	8	ps prepare time	The total execution time of the PREPARE statement, in microseconds.
40022	8	ps execute count	The number of times the EXECUTE statement is executed.
40023	8	ps close count	The number of times the CLOSE statement is executed.
40024	8	ps close time	The total execution time of the CLOSE statement, in microseconds.
40025	8	sql commit count	The number of times the COMMIT statement is executed.
40026	8	sql commit time	The total execution time of the COMMIT statement, in microseconds.
40027	8	sql rollback count	The number of times the ROLLBACK statement is executed.
40028	8	sql rollback time	The total execution time of the ROLLBACK statement, in microseconds.
40030	8	opened cursors current	The number of cursors that are currently opened. The number of cursors that are currently opened. When a cursor is opened, the count increases by 1. When a cursor is closed, the count decreases by 1.
40031	8	opened cursors cumulative	The number of cursors that are cumulatively opened. The number of cursors that are cumulatively opened. When a cursor is opened, the count increases by 1.
40102	8	sql inner insert count	The total number of INSERT statements that are issued in INNER SQL¹.
40103	8	sql inner insert time	The total time consumed by INSERT statements that are issued in INNER SQL, in microseconds¹.
40104	8	sql inner replace count	The total number of REPLACE statements that are issued in INNER SQL¹.
40105	8	sql inner replace time	The total time consumed by REPLACE statements that are issued in INNER SQL, in microseconds¹.
40106	8	sql inner update count	The total number of UPDATE statements that are issued in INNER SQL¹.
40107	8	sql inner update time	The total time consumed by UPDATE statements that are issued in INNER SQL, in microseconds¹.
40108	8	sql inner delete count	The total number of DELETE statements that are issued in INNER SQL¹.
40109	8	sql inner delete time	The total time consumed by DELETE statements that are issued in INNER SQL, in microseconds¹.
40112	8	user logons cumulative	The total number of user logons.
40113	8	user logouts cumulative	The total number of user logouts. When a session is disconnected, the count increases by 1.
40114	8	user logons failed cumulative	The total number of failed user logons. The process of establishing a session with an OBServer node fails.
40115	8	user logons time cumulative	The total time consumed by user logons, in microseconds. The time consumed by the process of establishing a session when a user logs on.
40117	8	sql remote execute time	The total time consumed by SQL REMOTE statements, in microseconds. The total time consumed by SQL REMOTE statements.
40118	8	sql distributed execute time	The total time consumed by distributed execution of SQL statements, in microseconds. The total time consumed by distributed execution, which is the sum of the time consumed by distributed execution.
40119	8	sql fail count	The number of failed SQL executions.
40120	8	inner sql local count	The number of INNER SQL LOCAL statements that are executed.
40121	8	inner sql remote count	The number of INNER SQL statements that are executed in a remote node.
40122	8	inner sql distributed count	The number of INNER SQL statements that are executed in a distributed manner.

Note

Note 1: An INNER SQL statement is an SQL statement that is not sent by a user but sent by OceanBase Database. It is usually used to query or modify internal tables.

Cache

Metric ID	Metric Type	Metric Name	Description
50000	16	row cache hit	The number of times data rows are fetched from the cache by the table get operator. If a data row is found in the cache, the value of this metric increases by 1.
50001	16	row cache miss	The number of times data rows are fetched from the cache by the table get operator. If a data row is not found in the cache, the value of this metric increases by 1.
50004	16	bloom filter cache hit	The number of times macroblock bloom filters are used in SQL statements. This metric indicates that the SQL statement uses a built-in macroblock bloom filter².
50005	16	bloom filter cache miss	The number of times macroblock bloom filters are not used in SQL statements. This metric indicates that the macroblock bloom filter is not built².
50006	16	bloom filter filts	The number of effective filter operations performed by macroblock bloom filters. This metric indicates that the corresponding row does not exist².
50007	16	bloom filter passes	The number of times macroblock bloom filters fail to filter rows. This metric indicates that the corresponding row may exist².
50008	16	block cache hit	The number of times microblocks are fetched from the block cache. If a microblock is found in the cache, the value of this metric increases by 1.
50009	16	block cache miss	The number of times microblocks are fetched from the block cache. If a microblock is not found in the cache, the value of this metric increases by 1.
50010	16	location cache hit	The number of times location cache is hit. This metric includes the number of hits in the ls_location_cache and tablet_ls_cache. When querying the mapping between tablets and logstreams or the location of a logstream, if effective data is successfully retrieved from the local cache, the value of this metric increases by 1. This metric is triggered in the get method of ObTabletLSService and ObLSLocationService.
50011	16	location cache miss	The number of times location cache is not hit. This metric includes the number of misses in the ls_location_cache and tablet_ls_cache. When querying the mapping between tablets and logstreams or the location of a logstream, if effective data cannot be retrieved from the local cache (the cache does not exist, has expired, or is invalid), the value of this metric increases by 1. This metric is triggered in the get method of ObTabletLSService and ObLSLocationService when is_cache_hit is false.
50012	16	location cache wait	The number of times threads are blocked while waiting for the location cache to acquire a location. When the number of concurrent threads reaches the maximum limit (max_count), threads need to wait on a semaphore. Each time a thread enters the waiting state in the ObLocationSem::acquire method, the value of this metric increases by 1. This metric is triggered when has_wait is true, to control the concurrency level of location information retrieval and prevent excessive concurrent requests from causing system pressure.
50013	16	location cache get hit from proxy virtual table	The number of times location cache is hit when querying tablet location information from the `__all_virtual_proxy_schema` view. When querying tablet location information from the ObAllVirtualProxySchema proxy virtual table, if the mapping between tablets and logstreams and the location of the logstream can be successfully retrieved from the cache, the value of this metric increases by 1.
50014	16	location cache get miss from proxy virtual table	The number of times location cache is not hit when querying tablet location information from the `__all_virtual_proxy_schema` view. When querying tablet location information from the ObAllVirtualProxySchema proxy virtual table, if the mapping between tablets and logstreams or the location of the logstream cannot be retrieved from the cache, the value of this metric increases by 1.
50017	16	location nonblock get hit	The number of non-blocking reads from the ls_location_cache and tablet_ls_cache. When using the nonblock_get method to query the mapping between tablets and logstreams or the location of a logstream, if effective data is successfully retrieved from the local cache, the value of this metric increases by 1. This metric is triggered in the `ObTabletLSService::nonblock_get` and `ObLSLocationService::nonblock_get` methods when OB_SUCC(ret) is true. Unlike the regular get method, nonblock_get does not wait for cache updates and immediately returns the result.
50028	16	kvcache sync wash time	The time consumed in synchronously flushing cache memory when memory is insufficient, in microseconds. This metric is incremented by the corresponding time when try_flush_washable_mb is called and cache flushing is performed.
50029	16	kvcache sync wash count	The number of times that cache is flushed to memory when memory is insufficient. This value increases by 1 when the `try_flush_washable_mb` function is called and cache is flushed.
50033	16	fuse row cache hit	The number of times that cache is hit when a table get operator reads a row that is not in the memtable. In the `get_fuse_row_cache` function, this value increases by 1 if the row corresponding to the specified rowkey is found in the cache.
50034	16	fuse row cache miss	The number of times that cache is not hit when a table get operator reads a row that is not in the memtable. In the `get_fuse_row_cache` function, this value increases by 1 if the row corresponding to the specified rowkey is not found in the cache.
50035	16	schema cache hit	The number of times that schema is successfully retrieved from the schema cache. When a query for schema is executed, this value increases by 1 if the corresponding schema information is successfully retrieved from the local cache. This value is incremented in the `ObSchemaCache::get_schema` method when the `cache_.get()` function call succeeds.
50036	16	schema cache miss	The number of times that schema is not successfully retrieved from the schema cache. When a query for schema is executed, this value increases by 1 if the corresponding schema information cannot be retrieved from the local cache (the cache does not exist, has expired, or is invalid). This value is incremented in the `ObSchemaCache::get_schema` method when the `cache_.get()` function call fails.
50037	16	tablet ls cache hit	The number of times that cache is hit when a tablet is read. In the `get_from_cache_` function, this value increases by 1 if the tablet is found in the cache.
50038	16	tablet ls cache miss	The number of times that cache is not hit when a tablet is read. In the `get_from_cache` function, this value increases by 1 if the tablet is not found in the cache.
50045	16	opt table stat cache hit	The hit rate of table-level or partition-level statistics cache. If the cache hit occurs, the statistics are directly read from memory. The unit is a percentage. KVCache records the hit rate of table-level or partition-level statistics cache.
50046	16	opt table stat cache miss	The miss rate of table-level or partition-level statistics cache. If the cache miss occurs, the statistics are loaded from the internal table to memory. The unit is a percentage. KVCache records the miss rate of table-level or partition-level statistics cache.
50047	16	opt column stat cache hit	The hit rate of column-level statistics cache. If the cache hit occurs, the statistics are directly read from memory. The unit is a percentage. KVCache records the hit rate of column-level statistics cache.
50048	16	opt column stat cache miss	The miss rate of column-level statistics cache. If the cache miss occurs, the statistics are loaded from the internal table to memory. The unit is a percentage. KVCache records the miss rate of column-level statistics cache.
50049	16	tmp page cache hit	The number of times that cache is hit when a temporary file is read. The temporary file is stored in a cache that is organized in 8 KB pages. Before data is read from the temporary file, the KVCache is queried. If the cache hit occurs, the data is read from the KVCache, and this value increases by 1.
50050	16	tmp page cache miss	The number of times that cache is not hit when a temporary file is read. The temporary file is stored in a cache that is organized in 8 KB pages. Before data is read from the temporary file, the KVCache is queried. If the cache miss occurs, an I/O operation is initiated, and this value increases by 1.
50051	16	tmp block cache hit	The number of times that cache is hit when a temporary file is read. Before data is read from the temporary file, the KVCache is queried to determine whether the data is in the cache. If the cache hit occurs, the data is read from the cache, and this value increases by 1.
50052	16	tmp block cache miss	The number of times that cache is not hit when a temporary file is read.
50055	16	opt ds stat cache hit	The hit rate of dynamic sampling statistics cache. If the cache hit occurs, the results are directly read from memory. The unit is a percentage. KVCache records the hit rate of dynamic sampling statistics cache.
50056	16	opt ds stat cache miss	The miss rate of dynamic sampling statistics cache. If the cache miss occurs, a sampling SQL statement is initiated to collect statistics. The unit is a percentage. KVCache records the miss rate of dynamic sampling statistics cache.
50057	16	storage meta cache hit	The number of times that cache is hit for secondary and tertiary storage metadata. The tablet stores some large metadata (more than 1 KB) such as the schema of a table and multiple sstables. These metadata can be accessed on demand and do not need to be stored in memory. Therefore, we use KVCache to manage the memory lifecycle of these structures. These two metrics record the hit statistics of the metadata KVCache.
50058	16	storage meta cache miss	The number of times that cache is not hit for secondary and tertiary storage metadata. The tablet stores some large metadata (more than 1 KB) such as the schema of a table and multiple sstables. These metadata can be accessed on demand and do not need to be stored in memory. Therefore, we use KVCache to manage the memory lifecycle of these structures. These two metrics record the hit statistics of the metadata KVCache.
50059	16	tablet cache hit	This view displays the number of tablet fragment cache hits. A tablet is a data fragment, which is a common structure in many processes. When a tenant supports a large number of partitions, limited memory cannot guarantee that all tablets are resident in memory. Therefore, we use KVCache to manage and maintain the memory lifecycle of tablets. These metrics indicate the cache hit rate of tablets.
50060	16	tablet cache miss	This view displays the number of tablet fragment cache misses. A tablet is a data fragment, which is a common structure in many processes. When a tenant supports a large number of partitions, limited memory cannot guarantee that all tablets are resident in memory. Therefore, we use KVCache to manage and maintain the memory lifecycle of tablets. These metrics indicate the cache hit rate of tablets.
50061	16	schema history cache hit	This view displays the number of schema history cache hits. The schema history cache stores the mapping between `schema_type, tenant_id, schema_id, schema_version` and precise_schema_version to avoid repeatedly querying system tables for the exact schema version. If the corresponding exact schema version is successfully retrieved from the history cache, the value is incremented by 1.
50062	16	schema history cache miss	This view displays the number of schema history cache misses. The schema history cache stores the mapping between `schema_type, tenant_id, schema_id, schema_version` and precise_schema_version to avoid repeatedly querying system tables for the exact schema version. If the corresponding exact schema version cannot be retrieved from the history cache (the cache does not exist, has expired, or is invalid), the value is incremented by 1.
50063	16	opt system stat cache hit	This view displays the cache hit rate of cost model system parameters. If the cache is hit, the result is directly read from the cache in memory. The unit is a percentage. KVCache statistics the cache hit rate of cost model parameters.
50064	16	opt system stat cache miss	This view displays the cache hit rate of cost model system parameters. If the cache is not hit, the cost model-related parameters are loaded from the internal table. The unit is a percentage. KVCache statistics the cache miss rate of cost model parameters.
50065	16	log kv cache hit	This view displays the number of hits in the cold cache. When reading logs from PALF, if the hot cache does not hit, the system attempts to read from the cold cache. If the log is found in the cold cache, the value is incremented by 1.
50066	16	log kv cache miss	This view displays the number of misses in the cold cache. When reading logs from PALF, if the hot cache does not hit, the system attempts to read from the cold cache. If the log is not found in the cold cache, the value is incremented by 1.
50067	16	data block cache miss	This view displays the number of times data microblocks are not found in the block cache. When reading data microblocks, the system checks if the microblock is in the cache. If not found, the value is incremented by 1.
50068	16	index block cache miss	This view displays the number of times intermediate index microblocks are not found in the block cache. When reading intermediate index microblocks, the system checks if the microblock is in the cache. If not found, the value is incremented by 1.
50069	16	multi version fuse row cache hit	This view displays the number of times rows are found in the cache when reading rows from a materialized view. In the `get_fuse_row_cache` function, if the table is a materialized view and the row corresponding to the rowkey is found in the cache, the value is incremented by 1.
50070	16	multi version fuse row cache miss	This view displays the number of times rows are not found in the cache when reading rows from a materialized view. In the `get_fuse_row_cache` function, if the table is a materialized view and the row corresponding to the rowkey is not found in the cache, the value is incremented by 1.
50071	16	backup index cache hit	This view displays the number of hits in the backup index cache. Backup indexes are stored in the cache. During recovery, the system first reads from the cache and, if not found, reads from the backup medium.
50072	16	backup index cache miss	This view displays the number of misses in the backup index cache. Backup indexes are stored in the cache. During recovery, the system first reads from the cache and, if not found, reads from the backup medium.
50073	16	backup meta cache hit	This view displays the number of hits in the backup metadata cache. Backup metadata is stored in the cache. During recovery, the system first reads from the cache and, if not found, reads from the backup medium.
50074	16	backup meta cache miss	This view displays the number of misses in the backup metadata cache. Backup metadata is stored in the cache. During recovery, the system first reads from the cache and, if not found, reads from the backup medium.
50075	16	truncate info cache hit	This view displays the number of times truncate info is found in the cache. In the `get_truncate_info_array` function, if the truncate info is found in the cache, the value is incremented by 1.
50076	16	truncate info cache miss	This view displays the number of times truncate info is not found in the cache. In the `get_truncate_info_array` function, if the truncate info is not found in the cache, the value is incremented by 1.
120001	16	tablet ls cache size	The size of the tablet cache, in bytes. The `get_from_cache` function uses the tablet cache.
120002	16	table stat cache size	The size of the memory space occupied by table-level or partition-level statistics cache, in bytes. The statistics are collected by the KVCache module.
120003	16	table col stat cache size	The size of the memory space occupied by column-level statistics cache, in bytes. The statistics are collected by the KVCache module.
120004	16	index block cache size	The size of the cache used for intermediate index microblocks, in bytes. The `get_micro_block_handle` function uses the `index_block_cache_` for intermediate index microblocks.
120006	16	user block cache size	The size of the cache used for data microblocks, in bytes. The `get_micro_block_handle` function uses the `index_block_cache_` for intermediate index microblocks.
120008	16	user row cache size	The size of the cache used for pre-fetched data rows by the table get operator, in bytes. The `lookup_in_cache` function uses this cache.
120009	16	bloom filter cache size	The total size of the Bloom filter cache, in bytes. When a user inserts data, the primary key conflict needs to be checked. To avoid I/O operations, a common performance optimization is to build a Bloom filter for the data range. These metrics are related to the Bloom filter cache.
120010	16	log kv cache size	The size of the logs read from the cold cache, in bytes. When reading logs from PALF, if the hot cache does not hit, it will attempt to read from the cold cache. If the log is found in the cold cache, the size of the logs read from the cold cache is accumulated.
240001	16	ss_micro_cache hit count	The total number of times all user requests hit the local microblock cache in shared storage mode. When a user request needs to access microblock data, if the data is obtained from the local microblock cache, it is considered a hit, and this metric is incremented by one.
240002	16	ss_micro_cache miss count	The total number of times all user requests missed the local microblock cache in shared storage mode. When a user request needs to access microblock data, if the data is not obtained from the local microblock cache, it is considered a miss, and this metric is incremented by one.
240003	16	ss_micro_cache failure count of adding cache	The total number of times attempts to add microblock data to the local microblock cache failed in shared storage mode. If the addition of new microblock data to the local microblock cache fails, this metric is incremented by one.
240004	16	ss_micro_cache failure count of getting cache	The total number of times attempts to retrieve specified microblocks from the local microblock cache failed in shared storage mode. If the retrieval of specified microblock data from the local microblock cache fails, this metric is incremented by one (if the cache miss is not considered a failure).
240005	16	ss_micro_cache major_compaction prewarm count	The number of times successfully added to the local microblock cache during the pre-warm phase of a major compaction in shared storage mode. During the pre-warm phase, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (even if it already exists), this metric is incremented by one.
240006	16	ss_micro_cache migrate prewarm count	The number of times successfully added to the local microblock cache during the pre-warm phase of a migration or cross-AZ disaster recovery in shared storage mode. During the pre-warm phase, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (even if it already exists), this metric is incremented by one.
240007	16	ss_micro_cache replica prewarm count	The number of times successfully added to the local microblock cache during the pre-warm phase between leader/follower replicas in shared storage mode. During the pre-warm phase, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (even if it already exists), this metric is incremented by one.
240008	16	ss_micro_cache ddl prewarm count	The number of times successfully added to the local microblock cache during the DDL pre-warm phase in shared storage mode. During the DDL pre-warm phase, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (even if it already exists), this metric is incremented by one.
240009	16	ss_micro_cache total hold micro_block count	The total number of micro_meta entries held in the local microblock cache in shared storage mode, including both valid and invalid metadata. For each microblock data added to the local microblock cache, a metadata entry (micro_meta) is created in memory. As data expires or is evicted, microblock data can transition from valid to invalid. This metric represents the total number of metadata entries in memory.
240010	16	ss_micro_cache total hold micro_block size	The total size of microblock data corresponding to micro_meta entries held in the local microblock cache in shared storage mode, including both valid and invalid metadata, in bytes. For each microblock data added to the local microblock cache, a metadata entry (micro_meta) is created in memory. As data expires or is evicted, microblock data can transition from valid to invalid. This metric represents the total size of microblock data corresponding to all metadata entries in memory.
240011	16	ss_micro_cache hit total bytes	The total size of data that is hit in the local microblock cache of all users in shared storage mode, in bytes. When a user request requires access to microblock data and the data is obtained from the local microblock cache, it is considered a hit. The value of this statistic increases by micro_size.
240012	16	ss_micro_cache miss total bytes	The total size of data that is not hit in the local microblock cache of all users in shared storage mode, in bytes. When a user request requires access to microblock data and the data is not obtained from the local microblock cache, it is considered a miss. The value of this statistic increases by micro_size.
240013	16	ss_micro_cache major_compaction prewarm bytes	The total size of microblocks successfully added to the local microblock cache during the major compaction prewarm scenario in shared storage mode, in bytes. During major compaction prewarm, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (including if it already exists), the value of this statistic increases by micro_size.
240014	16	ss_micro_cache migrate prewarm bytes	The total size of microblocks successfully added to the local microblock cache during the migration replication or cross-AZ disaster recovery prewarm scenario in shared storage mode, in bytes. During migration replication or cross-AZ disaster recovery prewarm, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (including if it already exists), the value of this statistic increases by micro_size.
240015	16	ss_micro_cache replica prewarm bytes	The total size of microblocks successfully added to the local microblock cache during the leader/follower replica prewarm scenario in shared storage mode, in bytes. During leader/follower replica prewarm, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (including if it already exists), the value of this statistic increases by micro_size.
240016	16	ss_micro_cache ddl prewarm bytes	The total size of microblocks successfully added to the local microblock cache during the DDL prewarm scenario in shared storage mode, in bytes. During DDL prewarm, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (including if it already exists), the value of this statistic increases by micro_size.
240019	16	ss_micro_cache total used disk size	The size of the disk space already used by the local microblock cache in shared storage mode, in bytes. The local microblock cache reserves a block of disk space by preallocating a large file. This disk space is logically divided into 2 MB data blocks. This statistic indicates the total size of these data blocks that are already used.
240024	16	ss_micro_cache add micro_block count of common_io	The total number of times microblocks are added to the local microblock cache based on user requests in shared storage mode. When a user request requires pulling a microblock from object storage to the local disk, this value increases by 1.
240025	16	ss_micro_cache get micro_block count of common_io	The total number of times microblocks are retrieved from the tenant's local microblock cache based on user requests in shared storage mode. When a user request requires reading microblock data from the local microblock cache, this value increases by 1 (regardless of whether the read is successful).
240026	16	ss_micro_cache total new added micro_block count	The total number of times microblocks are added to the local microblock cache based on user requests in shared storage mode. When a user request requires adding microblock data to the local microblock cache, if the microblock does not currently exist in the cache, it is considered the first addition, and this statistic increases by 1.
240027	16	ss_micro_cache new added micro_block count of common_io	The total number of times microblocks are successfully added to the tenant's local microblock cache based on user requests in shared storage mode. When a user request requires successfully pulling a microblock from object storage to the local disk, if the microblock does not exist in the cache before the addition, this value increases by 1.
240028	16	ss_micro_cache total evicted micro_block count	The total number of ghost micro_meta entries accumulated in the local microblock cache in shared storage mode. The local microblock cache uses the ARC cache replacement algorithm. When a microblock needs to be evicted, its data is deleted, and the micro_meta becomes a ghost micro_meta. This statistic indicates the total number of micro_meta entries that have been converted to ghost micro_meta.
240029	16	ss_micro_cache total deleted micro_block count	The total number of ghost micro_meta entries accumulated in the local microblock cache in shared storage mode. The local microblock cache uses the ARC cache replacement algorithm. If a microblock is evicted but its micro_meta is retained, the micro_meta is considered a ghost micro_meta. This statistic indicates the total number of ghost micro_meta entries that have been deleted.
240030	16	ss_micro_cache released phy_block count by reorgan_task	The total number of data blocks net released by microblock reorganization tasks in shared storage mode. The local microblock cache manages disk space in 2 MB data blocks. When the disk space is full, low-utilization data blocks are aggregated into new data blocks to release the space. This statistic indicates the total number of net-released data blocks.
240031	16	ss_micro_cache storage_cache_policy prewarm count	The number of times microblocks are successfully added to the local microblock cache during manual hot data prewarm scenarios in shared storage mode. During manual hot data prewarm, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (including if it already exists), this statistic increases by 1.
240032	16	ss_micro_cache storage_cache_policy prewarm bytes	The total size of microblocks successfully added to the local microblock cache during manual hot data prewarm scenarios in shared storage mode, in bytes. During manual hot data prewarm, hot microblock data is attempted to be added to the local microblock cache. If the microblock data is successfully added to the cache (including if it already exists), the value of this statistic increases by micro_size.
245001	16	ss_micro_cache micro_meta used memory size	The size of memory consumed by the local microblock cache in shared storage mode, in bytes. This includes the total size of memory consumed by microblock Meta data within the local microblock cache.
245002	16	ss_micro_cache total alloc disk size	In shared storage mode, the disk space allocated to the local microblock cache, in bytes. When initializing the local microblock cache module for a tenant, a disk space size is calculated and allocated to the local microblock cache.
245011	16	ss_micro_cache hot micro_block lack count	In shared storage mode, the number of microblock data entries missing in the local microblock cache for KVCache. A periodic task scans all microblocks in KVCache and compares them with the local microblock cache. Each time the task runs, the value of this statistic starts at 0 and increases by 1 for each microblock that exists in KVCache but not in the local microblock cache.
245012	16	ss_micro_cache read iops of common_io	In shared storage mode, the number of reads from the local microblock cache per second based on user requests. The total number of attempts to read microblock data from the local microblock cache based on user requests within 2 seconds is calculated, and the IOPS per second is derived from this value.
245013	16	ss_micro_cache read throughput of common_io	In shared storage mode, the amount of data read from the local microblock cache per second based on user requests, in bytes. The total size of microblock data attempted to be read from the local microblock cache based on user requests within 2 seconds is calculated, and the THROUGHPUT per second is derived from this value.
245014	16	ss_micro_cache write iops of common_io	In shared storage mode, the number of writes to the local microblock cache per second based on user requests. The total number of attempts to write microblock data to the local microblock cache based on user requests within 2 seconds is calculated, and the IOPS per second is derived from this value.
245015	16	ss_micro_cache write throughput of common_io	In shared storage mode, the amount of data written to the local microblock cache per second based on user requests, in bytes. The total size of microblock data attempted to be written to the local microblock cache based on user requests within 2 seconds is calculated, and the THROUGHPUT per second is derived from this value.
245016	16	ss_micro_cache read iops of prewarm_io	In shared storage mode, the number of reads from the local microblock cache per second based on preheating requests. The total number of attempts to read microblock data from the local microblock cache based on preheating requests within 2 seconds is calculated, and the IOPS per second is derived from this value.
245017	16	ss_micro_cache read throughput of prewarm_io	In shared storage mode, the amount of data read from the local microblock cache per second based on preheating requests, in bytes. The total size of microblock data attempted to be read from the local microblock cache based on preheating requests within 2 seconds is calculated, and the THROUGHPUT per second is derived from this value.
245018	16	ss_micro_cache write iops of prewarm_io	In shared storage mode, the number of writes to the local microblock cache per second based on preheating requests. The total number of attempts to write microblock data to the local microblock cache based on preheating requests within 2 seconds is calculated, and the IOPS per second is derived from this value.
245019	16	ss_micro_cache write throughput of prewarm_io	In shared storage mode, the amount of data written to the local microblock cache per second based on preheating requests, in bytes. The total size of microblock data attempted to be written to the local microblock cache based on preheating requests within 2 seconds is calculated, and the THROUGHPUT per second is derived from this value.
245020	16	ss_micro_cache valid micro_block count	In shared storage mode, the total number of valid microblocks in the local microblock cache. This value is updated each time microblock data is added to or evicted from the local microblock cache.
245021	16	ss_micro_cache valid micro_block size	In shared storage mode, the total size of valid microblock data in the local microblock cache, in bytes. This value is updated each time microblock data is added to or evicted from the local microblock cache.
245022	16	ss_macro_cache max capacity of the META_FILE type	In shared storage mode, the maximum disk cache capacity for Meta-type macroblocks in the macroblock cache, in bytes³.
245023	16	ss_macro_cache META_FILE used disk size	In shared storage mode, the disk size used by Meta-type macroblocks in the macroblock cache, in bytes. This is the total size of Meta-type macroblocks in the macroblock cache³.
245024	16	ss_macro_cache META_FILE hit count	In shared storage mode, the number of hits for Meta-type macroblocks in the macroblock cache. Each time a macroblock is attempted to be retrieved from the macroblock cache, the hit or miss count is incremented based on the macroblock type³.
245025	16	ss_macro_cache META_FILE miss count	In shared storage mode, the number of misses for Meta-type macroblocks in the macroblock cache. Each time a macroblock is attempted to be retrieved from the macroblock cache, the hit or miss count is incremented based on the macroblock type³.
245026	16	ss_macro_cache META_FILE hit bytes	In shared storage mode, the total size of hits for Meta-type macroblocks in the macroblock cache, in bytes. Each time a macroblock is attempted to be read from the macroblock cache, if the macroblock is found and it is of the Meta type, the size of the read data is added to this value³.
245027	16	ss_macro_cache META_FILE miss bytes	In shared storage mode, the total size of misses for Meta-type macroblocks in the macroblock cache, in bytes. Each time a macroblock is attempted to be read from the macroblock cache, if the macroblock is not found and it is of the Meta type, the size of the read data is added to this value³.
245028	16	ss_macro_cache max capacity of the TMP_FILE type	In shared storage mode, the maximum disk cache capacity for TMP_FILE-type macroblocks in the macroblock cache, in bytes. The macroblock cache supports cache eviction among different types, so this value is equal to the tenant's allocated macroblock cache disk size minus a reserved portion³.
245029	16	ss_macro_cache TMP_FILE used disk size	The size of the temporary file-type macroblocks in the macroblock cache in shared storage mode, in bytes. This value indicates the total size of the temporary file-type macroblocks in the macroblock cache³.
245030	16	ss_macro_cache TMP_FILE hit count	The number of hits for the temporary file-type macroblocks in the macroblock cache in shared storage mode. Each time a macroblock is requested from the macroblock cache, the hit or miss count is incremented based on the macroblock type³.
245031	16	ss_macro_cache TMP_FILE miss count	The number of misses for the temporary file-type macroblocks in the macroblock cache in shared storage mode. Each time a macroblock is requested from the macroblock cache, the hit or miss count is incremented based on the macroblock type³.
245032	16	ss_macro_cache TMP_FILE hit bytes	The total size of the temporary file-type macroblocks in the macroblock cache in shared storage mode, in bytes. Each time a macroblock is requested from the macroblock cache, if the macroblock is found in the cache and the macroblock type is temporary file-type, the size of the macroblock is added to this value³.
245033	16	ss_macro_cache TMP_FILE miss bytes	The total size of the temporary file-type macroblocks in the macroblock cache in shared storage mode, in bytes. Each time a macroblock is requested from the macroblock cache, if the macroblock is not found in the cache and the macroblock type is temporary file-type, the size of the macroblock is added to this value³.
245034	16	ss_macro_cache max capacity of the MACRO_BLOCK type	The maximum disk cache capacity for the normal macroblock type in the macroblock cache in shared storage mode, in bytes. The cache capacities of different macroblock types can be shared, so this value is the tenant's allocated macroblock cache disk size minus the reserved size³.
245035	16	ss_macro_cache MACRO_BLOCK used disk size	The size of the normal macroblocks in the macroblock cache in shared storage mode, in bytes. This value indicates the total size of the normal macroblocks in the macroblock cache³.
245036	16	ss_macro_cache MACRO_BLOCK hit count	The number of hits for the normal macroblocks in the macroblock cache in shared storage mode. Each time a macroblock is requested from the macroblock cache, the hit or miss count is incremented based on the macroblock type³.
245037	16	ss_macro_cache MACRO_BLOCK miss count	The number of misses for the normal macroblocks in the macroblock cache in shared storage mode. Each time a macroblock is requested from the macroblock cache, the hit or miss count is incremented based on the macroblock type³.
245038	16	ss_macro_cache MACRO_BLOCK hit bytes	The total size of the normal macroblocks in the macroblock cache in shared storage mode, in bytes. Each time a macroblock is requested from the macroblock cache, if the macroblock is found in the cache and the macroblock type is normal, the size of the macroblock is added to this value³.
245039	16	ss_macro_cache MACRO_BLOCK miss bytes	The total size of the normal macroblocks in the macroblock cache in shared storage mode, in bytes. Each time a macroblock is requested from the macroblock cache, if the macroblock is not found in the cache and the macroblock type is normal, the size of the macroblock is added to this value³.
245040	16	ss_macro_cache max capacity of the HOT_TABLET_MACRO_BLOCK type	The maximum disk cache capacity for the hot tablet macroblock type in the macroblock cache in shared storage mode, in bytes. The cache capacities of different macroblock types can be shared, so this value is the tenant's allocated macroblock cache disk size minus the reserved size³.
245041	16	ss_macro_cache HOT_TABLET_MACRO_BLOCK used disk size	The size of the hot tablet macroblocks in the macroblock cache in shared storage mode, in bytes. This value indicates the total size of the hot tablet macroblocks in the macroblock cache³.
245042	16	ss_macro_cache HOT_TABLET_MACRO_BLOCK hit count	The number of hits for the hot tablet macroblocks in the macroblock cache in shared storage mode. Each time a macroblock is requested from the macroblock cache, the hit or miss count is incremented based on the macroblock type³.
245043	16	ss_macro_cache HOT_TABLET_MACRO_BLOCK miss count	The number of misses for the hot tablet macroblocks in the macroblock cache in shared storage mode. Each time a macroblock is requested from the macroblock cache, the hit or miss count is incremented based on the macroblock type³.
245044	16	ss_macro_cache HOT_TABLET_MACRO_BLOCK hit bytes	The total size of the hot tablet macroblocks in the macroblock cache in shared storage mode, in bytes. Each time a macroblock is requested from the macroblock cache, if the macroblock is found in the cache and the macroblock type is hot tablet, the size of the macroblock is added to this value³.
245045	16	ss_macro_cache HOT_TABLET_MACRO_BLOCK miss bytes	The total size of the hot tablet macroblocks in the macroblock cache in shared storage mode, in bytes. Each time a macroblock is requested from the macroblock cache, if the macroblock is not found in the cache and the macroblock type is hot tablet, the size of the macroblock is added to this value³.
245046	16	ss_macro_cache private macro hit count	The number of hits for the private macroblocks in the macroblock cache in shared storage mode³.
245047	16	ss_macro_cache private macro miss count	The number of misses of private macros in the macro cache in shared storage mode. Each time a macro is attempted to be retrieved from the macro cache, the hit or miss count is incremented based on the macro type³.
245048	16	ss_macro_cache private macro hit bytes	The total size of hits of private macros in the macro cache in shared storage mode, in bytes. Each time a macro is attempted to be read from the macro cache, if the macro cache hit occurs and the macro type is private, the value is incremented by the size of the read macro³.
245049	16	ss_macro_cache private macro miss bytes	The total size of misses of private macros in the macro cache in shared storage mode, in bytes. Each time a macro is attempted to be read from the macro cache, if the macro cache miss occurs and the macro type is private, the value is incremented by the size of the read macro³.
245050	16	ss_macro_cache shared macro hit count	The number of hits of shared macros in the macro cache in shared storage mode. Each time a macro is attempted to be retrieved from the macro cache, the hit or miss count is incremented based on the macro type³.
245051	16	ss_macro_cache shared macro miss count	The number of misses of shared macros in the macro cache in shared storage mode. Each time a macro is attempted to be retrieved from the macro cache, the hit or miss count is incremented based on the macro type³.
245052	16	ss_macro_cache shared macro hit bytes	The total size of hits of shared macros in the macro cache in shared storage mode, in bytes. Each time a macro is attempted to be read from the macro cache, if the macro cache hit occurs and the macro type is shared, the value is incremented by the size of the read macro³.
245053	16	ss_macro_cache shared macro miss bytes	The total size of misses of shared macros in the macro cache in shared storage mode, in bytes. Each time a macro is attempted to be read from the macro cache, if the macro cache miss occurs and the macro type is shared, the value is incremented by the size of the read macro³.
245054	16	external table disk cache hit count	The number of hits of the disk cache for external tables in SN/SS mode. Each time an external table query is executed and the disk cache (enabled by default) is hit, the value is incremented by 1. After the disk cache for external tables is enabled, the data of the external table is split into macros of a specified size and stored on the disk.
245055	16	external table disk cache miss count	The number of misses of the disk cache for external tables in SN/SS mode. Each time an external table query is executed and the disk cache (enabled by default) is missed, the value is incremented by 1.
245056	16	external table disk cache hit bytes	The total size of hits of the disk cache for external tables in SN/SS mode, in bytes. Each time an external table query is executed and the disk cache (enabled by default) is hit, the value is incremented by the size of the read macro.
245057	16	external table disk cache miss bytes	The total size of misses of the disk cache for external tables in SN/SS mode, in bytes. Each time an external table query is executed and the disk cache (enabled by default) is missed, the value is incremented by the size of the read macro.
245058	16	ss_macro_cache total alloc disk size	The disk space allocated to the local macro cache in shared storage mode, in bytes. When the local macro cache module of a tenant is initialized, a disk space size is calculated and allocated to the local macro cache.

Note

Note 2: When you insert data, the system checks for primary key conflicts. To avoid sending I/O requests, a common performance optimization is to build a Bloom filter for the data range. The following metrics are related to Bloom filter cache.
Note 3: Macroblock cache is categorized into META, TMP_FILE, MACRO, and HOT_MACRO types. META includes all Meta-type macroblocks, and TMP_FILE includes temporary file-type macroblocks. If a tablet is marked as a hot tablet, all its macroblocks are of the HOT_MACRO type. Other macroblocks are of the MACRO type. Different types of macroblocks in the cache can occupy each other's space. Therefore, the value is the tenant's allocated macroblock cache disk size minus the reserved space.

Storage

Metric ID	Metric Type	Metric Name	Description
60000	32	io read count	The total IOPS of read I/O operations. This metric increases by 1 for each completed read I/O operation.
60001	32	io read delay	The total delay of read I/O operations, in microseconds. This metric accumulates the hardware service delay of each read I/O operation. The delay starts when the request is dispatched to the disk and ends when the disk completes processing and returns the result.
60002	32	io read bytes	The total bandwidth of read I/O operations. This metric accumulates the total read I/O bandwidth. The unit is Bytes. You can calculate the average read I/O bandwidth per second based on the change in this metric over a period of time. This metric increases by the read I/O bandwidth for each completed read I/O operation.
60003	32	io write count	The total IOPS of write I/O operations. This metric increases by 1 for each completed write I/O operation.
60004	32	io write delay	The total delay of write I/O operations, in microseconds. This metric accumulates the hardware service delay of each write I/O operation. The delay starts when the request is dispatched to the disk and ends when the disk completes processing and returns the result.
60005	32	io write bytes	The total bandwidth of write I/O operations. This metric accumulates the total write I/O bandwidth. The unit is Bytes. You can calculate the average write I/O bandwidth per second based on the change in this metric over a period of time. This metric increases by the write I/O bandwidth for each completed write I/O operation.
60006	32	memstore scan count	The number of range queries on the MemTable. This metric increases by 1 when a range query on the MemTable starts.
60007	32	memstore scan succ count	The number of successful range queries on the MemTable. This metric increases by 1 when a range query on the MemTable succeeds.
60008	32	memstore scan fail count	The number of failed range queries on the MemTable. This metric increases by 1 when a range query on the MemTable fails for any reason.
60009	32	memstore get count	The number of point queries on the MemTable. This metric increases by 1 when a point query on the MemTable starts.
60010	32	memstore get succ count	The number of successful point queries on the MemTable. This metric increases by 1 when a point query on the MemTable succeeds.
60011	32	memstore get fail count	The number of failed point queries on the MemTable. This metric increases by 1 when a point query on the MemTable fails for any reason.
60012	32	memstore apply count	The number of writes to the MemTable. This metric increases by 1 when a write to the MemTable starts.
60013	32	memstore apply succ count	The number of successful writes to the MemTable. This metric increases by 1 when a write to the MemTable succeeds.
60014	32	memstore apply fail count	The number of failed writes to the MemTable. This metric increases by 1 when a write to the MemTable fails for any reason.
60016	32	memstore get time	The time consumed by point queries on the MemTable, in microseconds. This metric increases by the time consumed by each point query on the MemTable, regardless of whether the query succeeds.
60017	32	memstore scan time	The time consumed by range queries on the MemTable, in microseconds. This metric increases by the time consumed by each range query on the MemTable, regardless of whether the query succeeds.
60018	32	memstore apply time	The time consumed by writes to the MemTable, in microseconds. This metric increases by the time consumed by each write to the MemTable, regardless of whether the write succeeds.
60019	32	memstore read lock succ count	The number of rows read successfully from the MemTable. This metric increases by 1 for each row read successfully from the MemTable.
60020	32	memstore read lock fail count	The number of rows read from the MemTable and failed due to read lock conflicts. During a read operation on the MemTable, if the data being read is in the process of being committed but the commit status and version number are not yet determined (for example, the data is in the process of being written to the commit log or a distributed transaction is in progress), the read operation will wait for the read lock. If the operation times out due to this conflict, this metric increases by 1.
60021	32	memstore write lock succ count	The number of rows written successfully to the MemTable. During a write operation on the MemTable, lock conflicts, lost updates, and primary key conflicts are checked. If no conflicts are found, this metric increases by 1. Note A write operation on the MemTable may still fail on the SSTable.
60022	32	memstore write lock fail count	The number of rows that fail to be written to the MemTable due to conflicts. During MemTable writes, lock conflicts, lost updates, and primary key conflicts are checked. If any of these conflicts occur, the value increases by 1. Note This metric does not include checks on SSTables.
60023	32	memstore wait write lock time	The time spent waiting for the MemTable write lock to be released, in microseconds. During MemTable writes, if a lock conflict occurs, the request is placed in the lock conflict queue, and this value is incremented by the time from when the request was placed in the queue until it is woken up.
60024	32	memstore wait read lock time	The time spent waiting for the MemTable read lock to be released, in microseconds. During MemTable reads, if the data is in a state where it has been committed but the commit status and version number are not yet determined (such as during a pending commit log or distributed transaction), the read will wait on the read lock, and this value is incremented by the waiting time.
60027	32	io prefetch micro block count	The total number of microblock cache I/O operations. During query path optimization, microblocks on the disk that need to be accessed are prefetched and loaded into memory, forming the microblock cache. This metric counts the total number of times the system loads microblocks from the disk. Combined with the microblock cache hit rate on internal tables, it can be used to measure the effectiveness of query performance optimization.
60028	32	io prefetch micro block bytes	The total number of bytes in microblock cache I/O operations, in bytes. During query path optimization, microblocks on the disk that need to be accessed are prefetched and loaded into memory, forming the microblock cache. This metric counts the total number of bytes loaded from the disk. Combined with the microblock cache hit rate on internal tables, it can be used to measure the effectiveness of query performance optimization.
60031	32	storage read row count	The number of rows processed after fusing in the non-pushed path of the table scan operator, in rows. This value is incremented by 1 each time the `process_fuse_row` function is called.
60032	32	storage delete row count	The number of effective rows actually deleted by DML statements. This value is incremented by 1 each time a row is deleted in the storage engine.
60033	32	storage insert row count	The number of effective rows actually inserted by DML statements. This value is incremented by 1 each time a row is inserted in the storage engine.
60034	32	storage update row count	The number of rows updated by update operations. This value is incremented by the number of rows updated each time the `update_rows` function is called.
60037	32	memstore row purge count	This metric is deprecated starting from V4.4.1.
60038	32	memstore row compaction count	The number of times sparse multi-version rows in the MemTable are compacted into a single version. Sparse multi-version rows in the MemTable only record changed columns. When the number of multi-versions increases, the MemTable generates a single compact version to improve read performance. This value is incremented by 1 each time a compact version is generated.
60056	32	memstore read row count	The number of rows read from the MemTable before filtering. This value is incremented by 1 each time a row is read from the MemTable.
60057	32	ssstore read row count	The number of rows read from major, minor, and mini SSTables in the query path, including both pushed and non-pushed paths. In the pushed path, it includes filtered rows, equivalent to the total number of rows in the SSTable range. In the rowstore, it includes rows skipped by the skip index during prefetching, while in the columnstore, it does not. In the non-pushed path, it includes filtered rows, excluding uncommitted, rolled back, and multi-version rows within microblocks.
60058	32	memstore write bytes	The amount of data successfully written to the MemTable, in bytes. After a successful MemTable write, the primary key and data size are recorded.
60068	32	memstore write lock wakenup count in lock_wait_mgr	The number of times the MemTable write lock conflict is resolved. During MemTable writes, if a lock conflict occurs, the request is placed in the lock conflict queue. If the request is woken up, this value is incremented by 1.
60077	32	get row effect read	The number of effective rows read from microblocks by the table get operator. In the `ObMicroBlockRowGetter::get_block_row` function, if the row flag is not DF_NOT_EXIST, this value is incremented by 1.
60078	32	get row empty read	The number of invalid rows read from microblocks by the table get operator, in rows. In the `ObMicroBlockRowGetter::get_block_row` function, if the row flag is DF_NOT_EXIST, this value is incremented by 1.
60083	32	memstore write lock wait timeout count	The number of times the MemTable write lock conflict timeout occurs. During MemTable writes, if a lock conflict occurs, the request is placed in the lock conflict queue. If the request times out before being woken up, this value is incremented by 1.
60084	32	accessed data micro block count by io	The number of microblocks that need to be accessed and require I/O because they were not found in the cache. When accessing a data microblock, if it is not found in the cache and an I/O operation is performed, this value is incremented by 1.
60085	32	data micro block cache hit	The number of data microblocks hit in the block cache. When accessing a data microblock, if it is found in the cache, this value is incremented by 1.
60086	32	accessed index micro block count by io	The number of microblocks accessed from the intermediate index layer without cache hit. This counter increments by 1 for each microblock accessed from the intermediate index layer without cache hit.
60087	32	index micro block cache hit	The number of microblocks accessed from the intermediate index layer with cache hit. This counter increments by 1 for each microblock accessed from the intermediate index layer with cache hit.
60088	32	blockscaned data micro block count	The number of microblocks opened in the push-down path. This counter includes the number of microblocks skipped by the skip index for both rowstore and columnstore. For columnstore, the number of microblocks in each CG is summed. The unit is the number of microblocks. For rowstore, this counter increments by 1 when a microblock is opened in the push-down path. For columnstore, this counter increments by 1 when a microblock is opened in the push-down path.
60089	32	blockscaned row count	The total number of rows in the range covered by the microblocks opened in the push-down path. This counter includes filtered rows (rows that cannot be pushed down, such as uncommitted rows, rolled-back rows, and multi-version rows within a microblock). For rowstore, this counter includes rows skipped by the skip index. For columnstore, this counter does not include rows skipped by the skip index. The unit is the number of rows. For rowstore, this counter increments by the total number of rows in the microblock when the microblock is opened in the push-down path. For columnstore, this counter increments by the total number of rows processed after the filter or projection is completed.
60090	32	storage filtered row count	The total number of rows filtered by the push-down or non-push-down filter. This counter increments by the number of filtered rows during the filter operation. For rowstore, this counter includes rows skipped by the skip index. For columnstore, this counter does not include rows skipped by the skip index.
60091	32	minor ssstore read row count	The number of rows read from minor sstables and mini sstables in the push-down or non-push-down path. The unit is the number of rows. For the push-down path, this counter includes filtered rows, equivalent to the total number of rows in the range covered by the minor/mini sstables accessed in the push-down path. For the non-push-down path, this counter includes filtered rows, excluding uncommitted rows, rolled-back rows, and multi-version rows within a microblock.
60092	32	major ssstore read row count	The number of rows read from major sstables in the push-down or non-push-down path. The unit is the number of rows. For the push-down path, this counter includes filtered rows, equivalent to the total number of rows in the range covered by the major sstables accessed in the push-down path. For rowstore, this counter includes rows skipped by the skip index. For columnstore, this counter does not include rows skipped by the skip index. For the non-push-down path, this counter includes filtered rows, excluding uncommitted rows, rolled-back rows, and multi-version rows within a microblock.
60093	32	storage waiting throttle time	The total time spent throttling memory allocation to dynamically match the system's compaction speed. When the compaction speed cannot keep up with the memory write speed, some modules' memory allocations are throttled. This counter tracks the total throttling time for these memory allocations, in microseconds. After the memory usage of modules like MemTable (TxData, Mds) exceeds a certain threshold, a thread sleeps for a period after each memory allocation, and the total sleep time is accumulated in this counter.
69000	32	backup io read count	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69001	32	backup io read bytes	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69002	32	backup io write count	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69003	32	backup io write bytes	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69010	32	backup delete count	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69012	32	backup io read delay	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69013	32	backup io write delay	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69017	32	backup delete delay	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69019	32	backup io list count	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69020	32	backup io read failed count	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69021	32	backup io write failed count	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69022	32	backup io delete failed count	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69023	32	backup io list failed count	This metric is deprecated starting from V4.4.1. Please check this metric in the I/O manager.
69024	32	backup io tagging count	This metric is deprecated starting with V4.4.1. Please check this metric in the I/O manager.
69025	32	backup io tagging failed count	This metric is deprecated starting with V4.4.1. Please check this metric in the I/O manager.
130000	32	active memstore used	The size of the active memstore on the node, in bytes. The memstore is used during node replay or write operations. When the memstore usage reaches a certain level, a freeze is triggered. All unfrozen memstores are considered active.
130001	32	total memstore used	The total amount of memory used by tenant memstores on the node, in bytes. This includes both active and frozen memstores, representing all tenant memstore memory usage on the node.
130002	32	major freeze trigger	The size of the memstore that triggers a freeze, in bytes. This is calculated by adding memstore_freeze_trigger and max_cached_memstore_size in the get_tenant_memstore_cond function.
130004	32	memstore limit	The maximum memory available for tenant memstores on the node, in bytes. Users can specify the memory limit for user tenants using the `memstore_limit_percentage` and `_memstore_limit_percentage` parameters. `memstore_limit_percentage` is a cluster-level parameter, while `_memstore_limit_percentage` is tenant-level. `_memstore_limit_percentage` takes precedence over `memstore_limit_percentage`. If neither is set, tenants with less than 8 GB of memory have a 40% memstore availability, and those with more than 8 GB have a 50% availability.
240022	32	object storage io head count	The count of object storage HEAD operations. This count increases by 1 when an object storage HEAD request is initiated.
240023	32	object storage io head fail count	The count of failed object storage HEAD operations. This count increases by 1 when an object storage HEAD request fails.

Resources

Metric ID	Metric Type	Metric Name	Description
140001	64	min memory size	The memory specification of the tenant, which determines the memory upper limit of the tenant. The OmtNodeBalancer background thread periodically retrieves the latest min_memory value of the tenant from the `DBA_OB_UNIT_CONFIGS` table and updates it to the Limit of ObTenantMemoryMgr.
140002	64	max memory size	The memory specification of the tenant, which determines the memory upper limit of the tenant, in bytes. The OmtNodeBalancer background thread periodically retrieves the latest min_memory value of the tenant from the `DBA_OB_UNIT_CONFIGS` table and updates it to the Limit of ObTenantMemoryMgr.
140003	64	memory usage	The memory occupied by the tenant, in bytes. ObTenantMemoryMgr maintains a real-time memory value to record the actual memory occupied by the tenant.
140004	64	min cpus	The CPU specification of the tenant. It determines the number of work threads that remain in the tenant and the number of threads in some background modules. The OmtNodeBalancer background thread periodically retrieves the latest max_cpu value of the tenant from the `DBA_OB_UNIT_CONFIGS` table and synchronizes it to the multi-tenant system. The OmtNodeBalancer background thread periodically adjusts the number of work threads based on the synchronized max_cpu value.
140005	64	max cpus	The CPU specification of the tenant. It determines the number of work threads that remain in the tenant and the number of threads in some background modules. When cgroups are enabled, it determines the upper limit of the CPU usage rate of the tenant threads. The OmtNodeBalancer background thread periodically retrieves the latest max_cpu value of the tenant from the `DBA_OB_UNIT_CONFIGS` table and synchronizes it to the multi-tenant system. The OmtNodeBalancer background thread periodically adjusts the number of work threads based on the synchronized max_cpu value.
140006	64	cpu usage	The thread usage rate of the tenant's work threads, which reflects the workload of the tenant's work threads, in percentage. The OmtNodeBalancer background thread statistics the ratio of the time spent by the tenant's work threads processing requests to the total time every 10 seconds, normalizes the ratio, and multiplies it by the tenant's min_cpu value to obtain the cpu_usage value.
140008	64	observer memory used size	The actual memory used by the OBServer node, in bytes. AChunkMgr maintains a real-time memory value to record the actual memory used by the OBServer node (which is less than or equal to the actual memory occupied).
140009	64	observer memory free size	The chunk cache of the OBServer node. AChunkMgr maintains a real-time memory value to record the cache size of the OBServer node. The cache represents the size of memory allocated to the OBServer node but not actually used.
140010	64	is mini mode	Indicates whether the OBServer node process is in mini mode. If the memory of the OBServer node is less than 12 GB, it is considered to be in mini mode.
140011	64	observer memory hold size	The memory occupied by the OBServer node, including the cache and all user-actually-used memory, in bytes. AChunkMgr maintains a real-time memory value to record the memory occupied by the OBServer node.
140012	64	worker time	The total busy time of the tenant's work threads, in microseconds. The OmtNodeBalancer background thread statistics the total time spent by the tenant's work threads processing requests every 10 seconds and accumulates it into the worker_time value.
140013	64	cpu time	The total CPU time used by the tenant's work threads and background threads, in microseconds. When cgroups are enabled: The total CPU time of the tenant threads can be obtained by reading the cpuacct.usage file in the cgroup directory of the tenant. When cgroups are not enabled: The OmtNodeBalancer background thread traverses all tenant threads every 10 seconds, calculates the difference between the values of `/proc/[pid]/task/[tid]/stat` before and after the traversal, and accumulates the CPU time increment.
140014	64	effective observer memory limit	The memory upper limit of the OBServer node, in bytes. Only the sys tenant can query this value, which is node-level. The default value of the `memory_limit_percentage` configuration item sets the memory upper limit of the OBServer node to `memory_limit_percentage * physical memory of the machine`. Users can set the `memory_limit` configuration item to determine the memory upper limit of the OBServer node.
140015	64	effective system memory	The reserved memory of the 500 tenant and the hidden sys memory, in bytes. By default, it adapts to a memory value based on the `memory_limit` configuration item; it can also be explicitly specified using the `system_memory` configuration item.
140016	64	effective hidden sys memory	The memory upper limit of the hidden sys tenant, in bytes. Each OBServer node allocates a portion of its memory to the sys tenant. By default, it adapts to a memory value based on the `system_memory` configuration item; users can also explicitly specify it using the `_hidden_sys_tenant_memory` configuration item.
140017	64	max session num	The maximum number of sessions allowed for a tenant. The maximum number of sessions allowed for a tenant is calculated based on the memory size of the tenant.
140018	64	kvcache hold	The size of memory occupied by the KVCache, in bytes. The memory manager tracks the memory occupied by the KVCache.
140019	64	unmanaged memory size	The size of memory directly read from mmap by the library function, in bytes. This is a global value that records the memory allocated by the library function calls to mmap/unmap in real time. It refers to the memory allocated by the mmap calls of the third-party library functions called on the OBServer node.
140020	64	hidden sys data disk size	The size of the disk configured for the sys tenant in shared storage mode, in bytes. For non-sys tenants, this value is 0. For the sys tenant, the value depends on the following scenarios: If the value of the hidden parameter `_ss_hidden_sys_tenant_data_disk_size` is 0, the value of HIDDEN_SYS_DATA_DISK_SIZE is twice the value of the hidden parameter `_hidden_sys_tenant_memory` when the OBServer node starts. If the value of the hidden parameter `_ss_hidden_sys_tenant_data_disk_size` is not 0, the value of HIDDEN_SYS_DATA_DISK_SIZE is equal to the value of the hidden parameter.

Logs

Metric ID	Metric Type	Metric Name	Description
70000	128	refresh schema count	The number of times the schema is successfully refreshed. When the OceanBase Database needs to refresh the schema, the value is incremented each time the schema refresh operation is successfully completed. This is implemented in the `ObServerSchemaService::refresh_schema` method. The operation includes two modes: full refresh `refresh_full_schema` and incremental refresh `refresh_increment_schema`, which are used to maintain consistency between the local metadata cache and the system tables.
70001	128	refresh schema time	The total time spent on successful schema refreshes, in microseconds. When the OceanBase Database refreshes the schema, the total time for the entire refresh process is recorded. In the `ObServerSchemaService::refresh_schema` method, when OB_SUCC(ret) is True, the time difference from the start to the end of the method is accumulated. This includes the overhead of both full refresh `refresh_full_schema` and incremental refresh `refresh_increment_schema`.
70002	128	inner sql connection execute count	The number of INNER SQL executions. The count is incremented by 1 for each INNER SQL executed.
70003	128	inner sql connection execute time	The total time spent on INNER SQL executions, in microseconds. The total time for INNER SQL executions is accumulated.
70006	128	log stream table operator get count	The number of successful reads of loaded LS information from the `__all_ls_meta_table` table. When `ObLSTableOperator` successfully retrieves `ls_info` from the `__all_ls_meta_table` table, the value is incremented by 1. This is implemented in the `ObLSTableOperator::get` method. The operator supports retrieving log stream information from memory, RPC, or persistent tables, used for location caching and metadata management.
70007	128	log stream table operator get time	The total time spent on successful reads of loaded LS information from the `__all_ls_meta_table` table, in microseconds. When `ObLSTableOperator` retrieves log stream information from the `__all_ls_meta_table` table, the total time for the entire retrieval process is recorded. In the `ObLSTableOperator::get` method, the time difference from the start to the end of the method is accumulated. This includes the overhead of retrieving log stream information from memory, RPC, or persistent tables.
160001	128	oblogger log bytes	The total size of logs printed, in bytes. This refers to the output printing process of system logs such as observer.log.
160002	128	election log bytes	The total size of election logs printed, in bytes. This refers to the accumulated size of election.log.
160004	128	oblogger total log count	The number of logs printed. This refers to the output printing process of system logs such as observer.log.
160019	128	async error log dropped count	The number of times ERROR logs failed to be printed (due to throttling, etc.). This refers to the output printing process of system logs such as observer.log.
160020	128	async warn log dropped count	The number of times WARN logs failed to be printed. This refers to the output printing process of system logs such as observer.log.
160021	128	async info log dropped count	The number of times INFO logs failed to be printed (due to throttling, etc.). This refers to the output printing process of system logs such as observer.log.
160022	128	async trace log dropped count	The number of times TRACE logs failed to be printed. This refers to the output printing process of system logs such as observer.log.
160023	128	async debug log dropped count	The number of times DEBUG logs failed to be printed (due to throttling, etc.). This refers to the output printing process of system logs such as observer.log.
160024	128	async log flush speed	The average number of logs written per second. This refers to the output printing process of system logs such as observer.log.
160025	128	async generic log write count	The number of times default logs were successfully printed. This refers to the output printing process of system logs such as observer.log.
160026	128	async user request log write count	Deprecated, retained only in code. This refers to the output printing process of system logs such as observer.log.
160027	128	async data maintain log write count	The number of times dump and merge logs were successfully printed. This refers to the output printing process of system logs such as observer.log.
160028	128	async root service log write count	The number of times RS logs were successfully printed. This refers to the output printing process of system logs such as observer.log.
160029	128	async schema log write count	The number of times schema logs failed to be printed (due to throttling, etc.). This refers to the output printing process of system logs such as observer.log.
160030	128	async force allow log write count	The number of times forced log writing is successful. This refers to the process of outputting system logs such as observer.log.
160031	128	async generic log dropped count	The number of times default log writing is failed (due to throttling, for example). This refers to the process of outputting system logs such as observer.log.
160032	128	async user request log dropped count	This view is deprecated and retained only in the code. This refers to the process of outputting system logs such as observer.log.
160033	128	async data maintain log dropped count	The number of times data maintenance log writing is failed (due to throttling, for example). This refers to the process of outputting system logs such as observer.log.
160034	128	async root service log dropped count	The number of times root service log writing is failed (due to throttling, for example). This refers to the process of outputting system logs such as observer.log.
160035	128	async schema log dropped count	The number of times schema log writing is failed (due to throttling, for example). This refers to the process of outputting system logs such as observer.log.
160036	128	async force allow log dropped count	The number of times forced log writing is failed (due to throttling, for example). This refers to the process of outputting system logs such as observer.log.

CLOG

Monitoring Item ID	Monitoring Item Type	Monitoring Item Name	Description
80001	256	palf write io count to disk	The number of times clog is written to disk. This value increases by 1 each time clog is successfully written to disk.
80002	256	palf write size to disk	The size of clog written to disk, in bytes. This value increases by the size of clog written to disk each time clog is successfully written to disk.
80003	256	palf write total time to disk	The total time taken to write clog to disk, in microseconds. This value increases by the time taken to write clog to disk each time clog is successfully written to disk.
80004	256	palf read count from hot cache	The number of times clog is read from the hot cache. If clog is read from the hot cache, this value increases by 1.
80005	256	palf read size from hot cache	The size of clog read from the hot cache, in bytes. If clog is read from the hot cache, this value increases by the size of clog read from the hot cache.
80006	256	palf read total time from hot cache	The total time taken to read clog from the hot cache, in microseconds. If clog is read from the hot cache, this value increases by the time taken to read clog from the hot cache.
80007	256	palf read io count from disk	The number of times clog is read from disk. This value increases by 1 each time clog is successfully read from disk.
80008	256	palf read size from disk	The size of clog read from disk, in bytes. This value increases by the size of clog read from disk each time clog is successfully read from disk.
80009	256	palf read total time from disk	The total time taken to read clog from disk, in microseconds. This value increases by the time taken to read clog from disk each time clog is successfully read from disk.
80010	256	palf handle rpc request count	The number of RPCs processed by the PALF layer. This value increases by 1 each time an RPC is processed.
80011	256	archive read log size	The total size of clog read by the log archiving module, in bytes. This value is periodically updated by the log archiving module to accumulate the total size of clog read in a unit of time. The update period is 10 seconds.
80012	256	archive write log size	The total size of clog written to the archiving medium, in bytes. This value is periodically updated by the log archiving module to accumulate the total size of clog written to the archiving medium in a unit of time. The update period is 10 seconds.
80013	256	restore read log size	The total size of clog read by the log restore module, in bytes. This value is periodically updated by the log restore module to accumulate the total size of clog read in a unit of time. The update period is 10 seconds.
80014	256	restore write log size	The total size of clog written by the log restore module, in bytes. This value is periodically updated by the log restore module to accumulate the total size of clog written in a unit of time. The update period is 10 seconds.
80057	256	clog trans log total size	The total size of data committed to the PALF layer, in bytes. This value increases by the size of data committed to the PALF layer.
80058	256	log storage compress original size	The total size of logs submitted to the PALF layer before compression, in bytes. This value is periodically updated by the log compression module to accumulate the total size of logs submitted to the PALF layer before compression in a unit of time. The update period is 10 seconds.
80059	256	log storage compress compressed size	The total size of logs submitted to the PALF layer after compression, in bytes. This value is periodically updated by the log compression module to accumulate the total size of logs submitted to the PALF layer after compression in a unit of time. The update period is 10 seconds.
81001	256	external log service fetch log size	The size of logs fetched from this node by the log synchronization link (such as OBCDC or a network standby database), in bytes. This value is periodically updated by the log synchronization link to accumulate the size of logs returned by each RPC processed. The update period is 10 seconds.
81002	256	external log service fetch log count	The number of logs fetched from this node by the log synchronization link (such as OBCDC or a network standby database). This value is periodically updated by the log synchronization link to accumulate the number of logs returned by each RPC processed. The update period is 10 seconds.
81003	256	external log service fetch rpc count	The number of RPCs sent by the log synchronization link (such as OBCDC or a network standby database) to fetch logs from this node. This value is periodically updated by the log synchronization link to accumulate the number of RPCs processed. The update period is 10 seconds.

Root Service

Metric ID	Metric Type	Metric Name	Description
110015	2048	success rpc process	The number of RPC requests successfully processed by the Root Service. This value is incremented by 1 each time the Root Service successfully processes an RPC request. This metric includes all RPC requests sent to the Root Service and executed successfully.
110016	2048	failed rpc process	The number of RPC requests failed to be processed by the Root Service. This value is incremented by 1 each time the Root Service fails to process an RPC request. This metric includes all RPC requests sent to the Root Service but failed to be executed.
110017	2048	balancer succ execute count	The number of times the Root Service load balancer successfully executes unit balancing. This value is incremented by 1 each time the Root Service load balancer successfully executes a unit balancing operation.
110018	2048	balancer failed execute count	The number of times the Root Service load balancer failed to execute unit balancing. This value is incremented by 1 each time the Root Service load balancer fails to execute a unit balancing operation.

ASH

Column	Type	Nullable	Description
200001	3072	DB time	The total time spent by foreground processes on database operations, including CPU and non-idle wait time. The unit is microseconds. OceanBase Database periodically updates the Time Model statistics. When a thread is active, it is either in the ON_CPU state or waiting for an event. The Time Model statistics are accumulated based on the duration of each state.
200002	3072	DB CPU	The total CPU time consumed by foreground processes. The unit is microseconds. OceanBase Database periodically updates the Time Model statistics. When a thread is active, it is either in the ON_CPU state or waiting for an event. The Time Model statistics are accumulated based on the duration of each state.
200005	3072	background elapsed time	The total time spent by background processes. The unit is microseconds. OceanBase Database periodically updates the Time Model statistics. When a thread is active, it is either in the ON_CPU state or waiting for an event. The Time Model statistics are accumulated based on the duration of each state.
200006	3072	background cpu time	The total CPU time consumed by background processes. The unit is microseconds. OceanBase Database periodically updates the Time Model statistics. When a thread is active, it is either in the ON_CPU state or waiting for an event. The Time Model statistics are accumulated based on the duration of each state.
200010	3072	non idle wait time	The total time spent by foreground processes on non-idle waits. This is the total time when a session is in a waiting state but not in an idle wait. The unit is microseconds. OceanBase Database periodically updates the Time Model statistics. When a thread is active, it is either in the ON_CPU state or waiting for an event. The Time Model statistics are accumulated based on the duration of each state.
200011	3072	idle wait time	The total time spent by foreground processes on idle waits. This is the total time when a session is in an idle wait state. The unit is microseconds. OceanBase Database periodically updates the Time Model statistics. When a thread is active, it is either in the ON_CPU state or waiting for an event. The Time Model statistics are accumulated based on the duration of each state.
200012	3072	background database time	The total time spent by background processes on database operations. The unit is microseconds. OceanBase Database periodically updates the Time Model statistics. When a thread is active, it is either in the ON_CPU state or waiting for an event. The Time Model statistics are accumulated based on the duration of each state.
200013	3072	background database non-idle wait time	The total time spent by background processes on non-idle waits during database operations. The unit is microseconds. OceanBase Database periodically updates the Time Model statistics. When a thread is active, it is either in the ON_CPU state or waiting for an event. The Time Model statistics are accumulated based on the duration of each state.
200014	3072	background database idle wait time	The total time spent by background processes on idle waits during database operations. The unit is microseconds. OceanBase Database periodically updates the Time Model statistics. When a thread is active, it is either in the ON_CPU state or waiting for an event. The Time Model statistics are accumulated based on the duration of each state.
200015	3072	diagnostic info object allocated count	The total number of diagnostic objects allocated. Diagnostic objects on OBServer nodes are used to store diagnostic statistics for both foreground and background tasks (including SQL and RPC tasks). These objects are typically allocated at the start of a task and recycled at the end.
200016	3072	diagnostic info object allocate failure count	The total number of diagnostic object allocation failures. The unit is the number of failures. Diagnostic objects on OBServer nodes are used to store diagnostic statistics for both foreground and background tasks (including SQL and RPC tasks). These objects are typically allocated at the start of a task and recycled at the end.
200017	3072	diagnostic info object returned count	The total number of diagnostic objects recycled. The unit is the number of objects. Diagnostic objects on OBServer nodes are used to store diagnostic statistics for both foreground and background tasks (including SQL and RPC tasks). These objects are typically allocated at the start of a task and recycled at the end.
200018	3072	diagnostic info object return failure count	The total number of diagnostic object recycling failures. Diagnostic objects on OBServer nodes are used to store diagnostic statistics for both foreground and background tasks (including SQL and RPC tasks). These objects are typically allocated at the start of a task and recycled at the end.
220001	3072	concurrency wait total time	The total wait time caused by concurrency-related wait events, such as resource locking. The unit is microseconds. This is accumulated from the corresponding wait events in OceanBase Database.
220002	3072	user io wait total time	The total time spent by user processes waiting for I/O operations (such as reading or writing data to disk) to complete. The unit is microseconds. This is accumulated from the corresponding wait events in OceanBase Database.
220003	3072	application wait total time	The total wait time caused by user application code, such as lock waits due to row-level locks or explicit lock commands. The unit is microseconds. This is accumulated from the corresponding wait events in OceanBase Database.
220004	3072	schedule wait total time	The total time consumed by wait events in the SCHEDULER category, in microseconds. This value is obtained by aggregating the corresponding wait events in OceanBase Database.
220005	3072	network wait total time	The total time consumed by network wait events, in microseconds. This value is obtained by aggregating the corresponding wait events in OceanBase Database.

TableAPI

WR

Metric ID	Metric Type	Metric Name	Description
210001	8192	wr snapshot task elapse time	The total time consumed by the snapshot collection task of the Workload Repository (WR), in microseconds. The WR periodically collects data from the internal views of OceanBase Database, records the system status, and provides relevant views for users to display and manage the DBMS_WORKLOAD_REPOSITORY package.
210002	8192	wr snapshot task cpu time	The total CPU time consumed by the snapshot collection task of the Workload Repository (WR), in microseconds. The WR periodically collects data from the internal views of OceanBase Database, records the system status, and provides relevant views for users to display and manage the DBMS_WORKLOAD_REPOSITORY package.
210003	8192	wr purge task elapse time	The total time consumed by the WR snapshot deletion task, in microseconds. The WR periodically collects data from the internal views of OceanBase Database, records the system status, and provides relevant views for users to display and manage the DBMS_WORKLOAD_REPOSITORY package.
210004	8192	wr purge task cpu time	The total CPU time consumed by the WR snapshot deletion task, in microseconds. The WR periodically collects data from the internal views of OceanBase Database, records the system status, and provides relevant views for users to display and manage the DBMS_WORKLOAD_REPOSITORY package.
210005	8192	wr schedular elapse time	The total time consumed by the WR task scheduling, in microseconds. The WR periodically collects data from the internal views of OceanBase Database, records the system status, and provides relevant views for users to display and manage the DBMS_WORKLOAD_REPOSITORY package.
210006	8192	wr schedular cpu time	The total CPU time consumed by the WR task scheduling, in microseconds. The WR periodically collects data from the internal views of OceanBase Database, records the system status, and provides relevant views for users to display and manage the DBMS_WORKLOAD_REPOSITORY package.
210007	8192	wr user submit snapshot elapse time	The total time consumed by the snapshot collection task initiated by a user, in microseconds. The WR periodically collects data from the internal views of OceanBase Database, records the system status, and provides relevant views for users to display and manage the DBMS_WORKLOAD_REPOSITORY package.
210008	8192	wr user submit snapshot cpu time	The total CPU time consumed by the snapshot collection task initiated by a user, in microseconds. The WR periodically collects data from the internal views of OceanBase Database, records the system status, and provides relevant views for users to display and manage the DBMS_WORKLOAD_REPOSITORY package.
210009	8192	wr collected active session history row count	The total number of rows collected in the ob_active_session_history view by the Workload Repository (WR). The WR periodically collects data from the internal views of OceanBase Database, records the system status, and provides relevant views for users to display and manage the DBMS_WORKLOAD_REPOSITORY package.
210010	8192	ash schedular elapse time	The total time consumed by the ASH sampling thread, in microseconds. ASH is a diagnostic mechanism that samples and records the status of all active sessions in OceanBase Database. A background thread samples the status of active sessions every second.
210011	8192	ash schedular cpu time	The total CPU time consumed by the ASH sampling thread, in microseconds. ASH is a diagnostic mechanism that samples and records the status of all active sessions in OceanBase Database. A background thread samples the status of active sessions every second.

References

For more information about the specific monitoring items, see Overview of system monitoring metrics.

This topic describes the key metrics of OceanBase Database and its core components, including clusters, tenants, and hosts.

Monitor item query

You can query the definition information of all monitor items in the sys tenant by using an SQL statement. Here is an example:

obclient> SELECT STAT_ID, CLASS, NAME FROM V$STATNAME;

Many monitor items are associated with internal implementations. Therefore, they may be changed or deleted. For details about how to query monitor items of specific categories, see Monitor items.