Metrics reference

tnt_info_uptime

Time in seconds since the instance has started

tnt_read_only

Indicates if the instance is in read-only mode (1 if true, 0 if false)

tnt_info_memory_data

Number of bytes used to store user data (tuples) with the memtx engine and with level 0 of the vinyl engine, without regard for memory fragmentation.

tnt_info_memory_index

Number of bytes used for indexing user data. Includes memtx and vinyl memory tree extents, the vinyl page index, and the vinyl bloom filters.

tnt_info_memory_lua

Number of bytes used for the Lua runtime. Monitoring this metric can prevent memory overflow.

tnt_info_memory_net

Number of bytes used for network input/output buffers.

tnt_info_memory_tx

Number of bytes in use by active transactions. For the vinyl storage engine, this is the total size of all allocated objects (struct txv, struct vy_tx, struct vy_read_interval) and tuples pinned for those objects.

tnt_slab_arena_size

Total memory available to store both tuples and indexes. Includes allocated but currently free slabs.

tnt_slab_items_size

Total amount of memory available to store only tuples and not indexes. Includes allocated but currently free slabs.

tnt_slab_arena_used

The effective memory used to store both tuples and indexes. Disregards allocated but currently free slabs.

tnt_slab_items_used

The effective memory used to store only tuples and not indexes. Disregards allocated but currently free slabs.

tnt_slab_arena_used_ratio

tnt_slab_arena_used / tnt_slab_arena_size

tnt_slab_items_used_ratio

tnt_slab_items_used / tnt_slab_items_size

tnt_space_bsize

Total number of bytes in all tuples. This metric always has 2 labels: {name="test", engine="memtx"}, where name is the name of the space and engine is the engine of the space.

tnt_space_index_bsize

Total number of bytes taken by the index. This metric always has 2 labels: {name="test", index_name="pk"}, where name is the name of the space and index_name is the name of the index.

tnt_space_total_bsize

Total size of tuples and all indexes in the space. This metric always has 2 labels: {name="test", engine="memtx"}, where name is the name of the space and engine is the engine of the space.

tnt_vinyl_tuples

Total tuple count for vinyl. This metric always has 2 labels: {name="test", engine="vinyl"}, where name is the name of the space and engine is the engine of the space. For vinyl this metric is disabled by default and can be enabled only with global variable setup: rawset(_G, 'include_vinyl_count', true).

tnt_net_connections_current

Number of active network connections

tnt_net_requests_current

Number of pending network requests

tnt_net_requests_in_progress_current

Count of requests currently being processed in the tx thread

tnt_net_requests_in_stream_current

Count of requests currently waiting in queues of streams

tnt_fiber_csw

Overall number of fiber context switches

tnt_fiber_memalloc

Amount of memory reserved for fibers

tnt_fiber_memused

Amount of memory used by fibers

call

Requests to execute stored procedures

delete

Delete calls

error

Requests resulted in an error

eval

Calls to evaluate Lua code

execute

Execute SQL calls

insert

Insert calls

prepare

SQL prepare calls

replace

Replace calls

select

Select calls

update

Update calls

upsert

Upsert calls

tnt_info_vclock

LSN number in vclock. This metric always has the label {id="id"}, where id is the instance's number in the replica set.

tnt_replication_lsn

LSN of the tarantool instance. This metric always has labels {id="id", type="type"}, where id is the instance's number in the replica set, type is master or replica.

tnt_replication_lag

Replication lag value in seconds. This metric always has labels {id="id", stream="stream"}, where id is the instance's number in the replica set, stream is downstream or upstream.

tnt_replication_status

This metrics equals 1 when replication status is "follow" and 0 otherwise. This metric always has labels {id="id", stream="stream"}, where id is the instance's number in the replica set, stream is downstream or upstream.

tnt_runtime_used

Number of bytes used for the Lua runtime

tnt_runtime_tuple

Number of bytes used for the tuples (except tuples owned by memtx and vinyl)

lj_jit_trace_num

Number of JIT traces

lj_jit_trace_abort_total

Overall number of abort traces

lj_jit_mcode_size

Total size of allocated machine code areas

lj_strhash_miss_total

Total number of string allocations

lj_gc_steps_sweepstring_total

Count of incremental GC steps (sweepstring state)

lj_gc_steps_finalize_total

Count of incremental GC steps (finalize state)

lj_gc_steps_sweep_total

Count of incremental GC steps (sweep state)

lj_gc_steps_propagate_total

Count of incremental GC steps (propagate state)

lj_gc_steps_pause_total

Count of incremental GC steps (pause state)

lj_gc_tabnum

Number of allocated table objects

lj_gc_cdatanum

Number of allocated cdata objects

lj_gc_udatanum

Number of allocated udata objects

lj_gc_freed_total

Total amount of freed memory

lj_gc_memory

Current allocated Lua memory

lj_gc_allocated_total

Total amount of allocated memory

tnt_cpu_number

Total number of processors configured by the operating system

tnt_cpu_time

Host CPU time

tnt_cpu_thread

Tarantool thread CPU time. This metric always has the labels {kind="user", thread_name="tarantool", thread_pid="pid", file_name="init.lua"}, where: - kind can be either user or system - thread_name is tarantool, wal, iproto, or coio - file_name is the entrypoint file name, for example, init.lua.

tnt_cpu_system_time

Tarantool CPU system time

tnt_vinyl_disk_index_size

Amount of data in bytes stored in the .index files located in vinyl_dir

tnt_vinyl_regulator_write_rate

Actual average rate of performing write operations, bytes per second. The rate is calculated as a 5-second moving average. If the metric value is gradually going down, this can indicate disk issues.

tnt_vinyl_regulator_rate_limit

Write rate limit, bytes per second. The regulator imposes the limit on transactions based on the observed dump/compaction performance. If the metric value is down to approximately 10^5, this indicates issues with the disk or the scheduler.

tnt_vinyl_regulator_dump_watermark

Maximum amount of memory in bytes used for in-memory storing of a vinyl LSM tree. When this maximum is accessed, a dump must occur. For details, see engines-algorithm_filling_lsm. The value is slightly smaller than the amount of memory allocated for vinyl trees, reflected in the vinyl_memory parameter.

tnt_vinyl_regulator_blocked_writers

The number of fibers that are blocked waiting for Vinyl level0 memory quota.

tnt_vinyl_tx_rollback

Сounter of rollbacks (unsuccessful transaction ends). This is not merely a count of explicit box.rollback() requests – it includes requests that ended with errors.

tnt_vinyl_tx_conflict

Counter of conflicts that caused transactions to roll back. The ratio tnt_vinyl_tx_conflict / tnt_vinyl_tx_commit above 5% indicates that vinyl is not healthy. At that moment, you'll probably see a lot of other problems with vinyl.

tnt_vinyl_tx_read_views

Current number of read views – that is, transactions that entered the read-only state to avoid conflict temporarily. Usually the value is 0. If it stays non-zero for a long time, it is indicative of a memory leak.

tnt_vinyl_memory_level0

"Level 0" (L0) memory area, bytes. L0 is the area that vinyl can use for in-memory storage of an LSM tree. By monitoring this metric, you can see when L0 is getting close to its maximum (tnt_vinyl_regulator_dump_watermark), at which time a dump will occur. You can expect L0 = 0 immediately after the dump operation is completed.

tnt_vinyl_memory_page_index

Amount of memory in bytes currently used to store indexes. If the metric value is close to vinyl_memory, this indicates that vinyl_page_size was chosen incorrectly.

tnt_vinyl_memory_bloom_filter

Amount of memory in bytes used by bloom filters.

tnt_vinyl_memory_tuple

Total size of memory in bytes occupied by Vinyl tuples. It includes cached tuples and tuples pinned by the Lua world.

tnt_vinyl_scheduler_tasks

Number of scheduler dump/compaction tasks. The metric always has label {status = <status_value>}, where <status_value> can be one of the following: - inprogress for currently running tasks - completed for successfully completed tasks - failed for tasks aborted due to errors.

tnt_vinyl_scheduler_dump_time

Total time in seconds spent by all worker threads performing dumps.

tnt_vinyl_scheduler_dump_total

Counter of dumps completed.

tnt_ev_loop_prolog_time

Event loop prolog time (ms)

tnt_ev_loop_epilog_time

Event loop epilog time (ms)

tnt_synchro_queue_term

Current queue term.

tnt_synchro_queue_len

How many transactions are collecting confirmations now.

tnt_synchro_queue_busy

Whether the queue is processing any system entry (CONFIRM/ROLLBACK/PROMOTE/DEMOTE).

tnt_election_state

Election state (mode) of the node. When election is enabled, the node is writable only in the leader state. Possible values: - 0 (follower): all the non-leader nodes are called followers - 1 (candidate): the nodes that start a new election round are called candidates. - 2 (leader): the node that collected a quorum of votes becomes the leader

tnt_election_vote

ID of a node the current node votes for. If the value is 0, it means the node hasn't voted in the current term yet.

tnt_election_leader

Leader node ID in the current term. If the value is 0, it means the node doesn't know which node is the leader in the current term.

tnt_election_term

Current election term.

tnt_election_leader_idle

Time in seconds since the last interaction with the known leader.

tnt_memtx_tnx_statements are the transaction statements.

For example, the user started a transaction and made an action in it space:replace{0, 1}. Under the hood, this operation will turn into statement for the current transaction. This metric always has the label {kind="..."}, which has the following possible values: - total: the number of bytes that are allocated for the statements of all current transactions. - average: average bytes used by transactions for statements (txn.statements.total bytes / number of open transactions). - max: the maximum number of bytes used by one the current transaction for statements.

tnt_memtx_tnx_user

In Tarantool C API there is a function box_txn_alloc(). By using this function user can allocate memory for the current transaction. This metric always has the label {kind="..."}, which has the following possible values: - total: memory allocated by the box_txn_alloc() function on all current transactions. - average: transaction average (total allocated bytes / number of all current transactions). - max: the maximum number of bytes allocated by box_txn_alloc() function per transaction.

tnt_memtx_mvcc_trackers

Trackers that keep track of transaction reads. This metric always has the label {kind="..."}, which has the following possible values: - total: trackers of all current transactions are allocated in total (in bytes). - average: average for all current transactions (total memory bytes / number of transactions). - max: maximum trackers allocated per transaction (in bytes).

tnt_memtx_mvcc_conflicts

Allocated in case of transaction conflicts. This metric always has the label {kind="..."}, which has the following possible values: - total: bytes allocated for conflicts in total. - average: average for all current transactions (total memory bytes / number of transactions). - max: maximum bytes allocated for conflicts per transaction.

tnt_memtx_mvcc_tuples_used_stories

Tuples that are used by active read-write transactions. This metric always has the label {kind="..."}, which has the following possible values: - count: number of used tuples / number of stories. - total: amount of bytes used by stories used tuples.

tnt_memtx_mvcc_tuples_used_retained

Tuples that are used by active read-write transactions. But they are no longer in the index, but MVCC does not allow them to be removed. This metric always has the label {kind="..."}, which has the following possible values: - count: number of retained used tuples / number of stories. - total: amount of bytes used by retained used tuples.

tnt_memtx_mvcc_tuples_read_view_stories

Tuples that are not used by active read-write transactions, but are used by read-only transactions (i.e. in read view). This metric always has the label {kind="..."}, which has the following possible values: - count: number of read_view tuples / number of stories. - total: amount of bytes used by stories read_view tuples.

tnt_memtx_mvcc_tuples_read_view_retained

Tuples that are not used by active read-write transactions, but are used by read-only transactions (i.e. in read view). This tuples are no longer in the index, but MVCC does not allow them to be removed. This metric always has the label {kind="..."}, which has the following possible values: - count: number of retained read_view tuples / number of stories. - total: amount of bytes used by retained read_view tuples.

tnt_memtx_mvcc_tuples_tracking_stories

Tuples that are not directly used by any transactions, but are used by MVCC to track reads. This metric always has the label {kind="..."}, which has the following possible values: - count: number of tracking tuples / number of tracking stories. - total: amount of bytes used by stories tracking tuples.

tnt_memtx_mvcc_tuples_tracking_retained

Tuples that are not directly used by any transactions, but are used by MVCC to track reads. This tuples are no longer in the index, but MVCC does not allow them to be removed. This metric always has the label {kind="..."}, which has the following possible values: - count: number of retained tracking tuples / number of stories. - total: amount of bytes used by retained tracking tuples.

tnt_memtx_tuples_data_read_view

Memory (in bytes) held for read views.

tnt_memtx_tuples_data_garbage

Memory (in bytes) that is unused and scheduled to be freed (freed lazily on memory allocation).

tnt_memtx_index_total

Total amount of memory (in bytes) allocated for indexing data. This includes tnt_memtx_index_read_view metric value plus memory used for indexing tuples that are actually stored in memtx spaces.

tnt_memtx_index_read_view

Memory (in bytes) held for read views.

tnt_config_alerts

Count of current instance configuration apply alerts. {level="warn"} label covers warnings and {level="error"} covers errors.

tnt_config_status

The status of current instance configuration apply. status label contains possible status name. Current status has metric value 1, inactive statuses have metric value 0. none # HELP tnt_config_status Tarantool 3 configuration status # TYPE tnt_config_status gauge tnt_config_status{status="reload_in_progress",alias="router-001-a"} 0 tnt_config_status{status="uninitialized",alias="router-001-a"} 0 tnt_config_status{status="check_warnings",alias="router-001-a"} 0 tnt_config_status{status="ready",alias="router-001-a"} 1 tnt_config_status{status="check_errors",alias="router-001-a"} 0 tnt_config_status{status="startup_in_progress",alias="router-001-a"} 0 For example, this set of metrics means that current configuration for router-001-a status is ready.