CHANGELOG.md

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

Added

Changed

Fixed

Removed

[1.7.0] - 2026-03-27

This release adds to the graphite plugin ability to send metrics to the multiple servers and hide replication metrics of the deleted replica.

Added

• graphite: ability to send metrics to the multiple servers. Backward compatibility with previous plugin version is preserved. From now on init method assigns an unique name to the created fiber using incoming graphite server opts (if passed). Added new stop() method to stop all fibers started by the plugin (#540).

Fixed

• Deleting a replica via box.space._cluster:delete() doesn't delete information about this replica from the metrics (it's gone only after cluster is restarted) (#538).

[1.6.2] - 2025-12-17

This release fixes the No such file or directory error that occurs if the /proc/<pid>/stat file does not exist.

Fixed

• The error message level displayed when the /proc/<pid>/stat file is missing has been changed to verbose. (#536).

[1.6.1] - 2025-10-20

This release introduces fix of the box.schema.needs_upgrade metric.

Fixed

• Use box.schema.needs_upgrade instead of box.schema_needs_upgrade (#529).

[1.6.0] - 2025-10-14

This release adds new memory and schema metrics.

Added

• tnt_memory metric (#519).
• tnt_memory_virt metric (#521).
• tnt_schema_needs_upgrade metric (#524).

Fixed

• Fixed a potential process metrics error when fio.read returns an empty string or an error (#527).

[1.5.0] - 2025-08-13

Added

• tnt_cartridge_config_checksum metric.

Removed

• tnt_cartridge_config_applied metric that only reflected local node state.

[1.4.0] - 2025-07-07

Added

• tnt_cartridge_config_applied metric.
• New optional label_keys parameter for counter() and gauge() metrics.

[1.3.1] - 2025-02-24

[1.3.0] - 2025-02-24

Added

• New metrics category in config: cpu_extended.

Fixed

• Use box.info.ro instead of box.cfg.read_only in replication metrics.

[1.2.0] - 2024-07-09

Added

• New Tarantool 3 metrics:
  • tnt_config_alerts
  • tnt_config_status

[1.1.0] - 2024-05-17

Added

• metrics.cfg{} "all" metasection for array include and exclude (metrics.cfg{include={'all'}} can be used instead of metrics.cfg{include='all'}, metrics.cfg{exclude={'all'}} can be used instead of metrics.cfg{include='none'})
• tnt_election_leader_idle metric.
• Histogram now logs a warning if observe is called with cdata value.

Changed

• Inconsistent metrics descrtiptions for memtx metrics.
• New memory metrics:
  • tnt_memtx_tuples_data_total
  • tnt_memtx_tuples_data_read_view
  • tnt_memtx_tuples_data_garbage
  • tnt_memtx_index_total
  • tnt_memtx_index_read_view
  • tnt_vinyl_memory_tuple

Deprecated

• Using cdata values with histogram:observe.

[1.0.0] - 2023-05-22

Changed

• Don't export self replication status.

Removed

• cartridge.roles.metrics moved to cartridge-metrics-role repository (incompatible change).

[0.17.0] - 2023-03-23

Added

• metrics.cfg{} -- a single entrypoint to setup the module:
  • include and exclude options with the same effect as in enable_default_metrics(include, exclude) (but its deprecated features already disabled);
  • labels options with the same effect as set_global_labels(labels);
  • values and effect (like default metrics callbacks) are preserved between reloads;
  • does not deal with external features like cartridge HTTP setup
• Versioning support through require('metrics')._VERSION

Changed

• Setup cartridge hotreload inside the role
• Extend enable_default_metrics() API:
  • 'all' and 'none' options for include argument,
  • simultaneous include and exclude arguments (exclude has higher priority)
• Build rock with cmake
• Override built-in metrics, if installed

Deprecated

• Passing nonexistent metrics to enable_default_metrics()
• Using {} as include in enable_default_metrics() to enable all metrics
• Versioning support through require('metrics').VERSION

[0.16.0] - 2023-01-27

Added

• Handle to clear psutils metrics
• invoke_callbacks option for metrics.collect()
• Ability to set metainfo for collectors
• Set metainfo.default to true for all collectors from enable_default_metrics() and psutils collectors
• default_only option for metrics.collect()

Fixed

• Fix is_healthy function to rely on membership state
• Fix psutils time metrics
• Fix missing psutils cpu count after clear

Removed

• Set non-number values in gauge:set
• Deprecated metrics from previous releases

[0.15.1] - 2022-09-20

Added

• memtx MVCC memory monitoring

Fixed

• tnt_synchro_queue_len metric type
• Reset callbacks on hotreload
• Fix queries in quantile

[0.15.0] - 2022-08-09

Fixed

• Clean info about spaces and indexes after their drop
• Fixed error when indexing spaces

Added

• Label thread for per thread reporting net statistics metrics.
• tnt_cartridge_failover_trigger_total metric
• New synchro and election metrics:
  • tnt_synchro_queue_owner
  • tnt_synchro_queue_term
  • tnt_synchro_queue_len
  • tnt_synchro_queue_busy
  • tnt_election_state
  • tnt_election_vote
  • tnt_election_leader
  • tnt_election_term
• Renamed LuaJit metrics:
  • lj_gc_allocated_total
  • lj_gc_freed_total
  • lj_gc_steps_atomic_total
  • lj_gc_steps_finalize_total
  • lj_gc_steps_pause_total
  • lj_gc_steps_propagate_total
  • lj_gc_steps_sweep_total
  • lj_gc_steps_sweepstring_total
  • lj_jit_snap_restore_total
  • lj_jit_trace_abort_total
  • lj_strhash_hit_total
  • lj_strhash_miss_total

Deprecated

• Metrics:
  • lj_gc_allocated
  • lj_gc_freed
  • lj_gc_steps_atomic
  • lj_gc_steps_finalize
  • lj_gc_steps_pause
  • lj_gc_steps_propagate
  • lj_gc_steps_sweep
  • lj_gc_steps_sweepstring
  • lj_jit_snap_restore
  • lj_jit_trace_abort
  • lj_strhash_hit
  • lj_strhash_miss

Removed

• Deprecated metrics from previous releases

[0.14.0] - 2022-06-28

Fixed

• Float numbers in Graphite exporter
• Signed timestamp in Graphite exporter
• Increase Shared.make_key performance in observations with empty label
• Forbid observation of non-number value in collectors (except gauge:set)
• Clean dead threads from psutils.cpu metric

Added

• tnt_cartridge_cluster_issues metric

Deprecated

• Set non-number values in gauge:set

Removed

• HTTP middleware v2
• enable_cartridge_metrics function

[0.13.0] - 2022-03-23

Fixed

• Don't reset collectors when Cartridge roles hot reload
• pairs instead of ipairs in iterations over replication info

Changed

• Type changed from gauge to counter:
  • tnt_net_sent_total
  • tnt_net_received_total
  • tnt_net_connections_total
  • tnt_net_requests_total
  • tnt_stats_op_total

Added

• New metrics:
  • tnt_vinyl_tuples (same as tnt_space_count)
  • tnt_fiber_amount (same as tnt_fiber_count)
  • lj_gc_memory (same as lj_gc_total)
  • tnt_cpu_number (same as tnt_cpu_count)
  • tnt_cpu_time (same as tnt_cpu_total)
  • tnt_vinyl_scheduler_dump_total (same as tnt_vinyl_scheduler_dump_count)
  • tnt_replication_lag
  • tnt_vinyl_regulator_blocked_writers
  • tnt_net_requests_in_progress_total
  • tnt_net_requests_in_progress_current
  • tnt_net_requests_in_stream_total
  • tnt_net_requests_in_stream_current
  • tnt_replication_lsn
  • tnt_replication_status
  • tnt_ev_loop_time
  • tnt_ev_loop_prolog_time
  • tnt_ev_loop_epilog_time

Deprecated

• Metrics:
  • tnt_net_sent_rps
  • tnt_net_received_rps
  • tnt_net_connections_rps
  • tnt_net_requests_rps
  • tnt_stats_op_rps
  • tnt_space_count
  • tnt_fiber_count
  • lj_gc_total
  • tnt_cpu_count
  • tnt_cpu_total
  • tnt_vinyl_scheduler_dump_count
  • tnt_replication_<id>_lag
  • tnt_replication_master_<id>_lsn
  • tnt_replication_replica_<id>_lsn

[0.12.0] - 2021-11-18

Changed

• Update http dependency to 1.1.1

Fixed

• Cast number64 to json number in json export plugin #321

Deprecated

• HTTP middleware v2

[0.11.0] - 2021-09-23

Added

• collector's method remove to clear observations with given label pairs #263
• counter:reset() method #260
• tnt_read_only metric #275

Removed

• Average collector

Fixed

• Throw an error when http_middelware is processing a wrong handler #199
• cartridge issues metric fails before cartridge.cfg() call #298

Changed

• quantile metric is NAN if no samples provided for an age #303

[0.10.0] - 2021-08-03

Changed

• metrics registry refactoring to search with O(1) #188
• ipairs instead of pairs while iteration in histogram #196
• set_export function provide default metrics config to make role reloadable #248
• metrics registry refactoring to add and remove callbacks with O(1) #276

Fixed

• be gentle to http routes, don't leave gaps in the array #246
• allow to create summary without observations #265

Added

• tnt_clock_delta metric to compute clock difference on instances #232
• set custom global labels in config and with set_labels function #259
• allow to include and exclude default metrics in config and in enable_default_metrics function #222
• unregister_callback function #262

Deprecated

• enable_cartridge_metrics function

[0.9.0] - 2021-05-28

Fixed

• cartridge metrics role fails to start without http #225
• quantile overflow after fiber.yield() #235
• role and module hot reload #227, #228

Changed

• tnt_cartridge_issues gathers only local issues #211

Added

• Age buckets in summary

[0.8.0] - 2021-04-13

Added

• New default metrics: cpu_user_time, cpu_system_time
• Vinyl metrics

[0.7.1] - 2021-03-18

Added

• zone label support for Tarantool Cartridge >= '2.4.0'
• rpm packaging for CentOS 8, Fedora 30, 31, 32

[0.7.0] - 2021-02-09

Added

• instance health check plugin

[0.6.1] - 2021-01-20

Fixed

• package reloading works for metrics.quantile
• instance_name in alias label if no alias present

[0.6.0] - 2020-12-01

Fixed

• metrics.clear() disables default metrics
• cartridge role is permanent
• cartridge role configuration without clusterwide config
• graphite plugin kills previous workers on init
• graphite plugin format numbers without ULL/LL-suffixes
• graphite plugin time in seconds
• graphite plugin allows empty prefix

Added

• Luajit platform metrics
• enable_cartridge_metrics function
• Cartridge issues gauge

Changed

• CI on Github Actions

[0.5.0] - 2020-09-18

Added

• Summary collector

Deprecated

• Average collector

[0.4.0] - 2020-07-14

Added

• New default metrics: cpu_total, cpu_thread
• histogram:observe_latency for measure latency of function call with example

[0.3.0] - 2020-06-11

Added

• Role for tarantool/cartridge
• Documentaion and examples on tarantool/http server middleware

Fixed

• Throw exception when http_middleware.build_default_collector is called with same name
• Attempt to index non-existent master vclock on a replica after the death of the master

[0.2.0] - 2020-05-07

Added

• tarantool/http server middleware to collect http server metrics

Fixed

• collect failure for vinyl metrics in strict mode
• prometheus exporter: render 0 for 0ULL value instead of +Inf
• Travis CI build failures

Changed

• Renamed info_vclock_{ID} metric to info_vclock, moved {ID} to tags
• Renamed stat_op_{OP_TYPE}_total metric to stat_op_total, moved {OP_TYPE} to tags
• Renamed stat_op_{OP_TYPE}_rps metric to stat_op_rps, moved {OP_TYPE} to tags
• Renamed space_index_{IDX_NAME}_bsize to space_index_bsize, moved {IDX_NAME} to tags

[0.1.8] - 2020-01-15

Added

• Ability to set labels globally for each metric