In _data/v25.3/metrics/export, copied crdb_metrics.yaml, shared_metrics.yaml, and tenant_metrics.yaml from https://github.com/cockroachlabs/managed-service/tree/master/pkg/otel/assets.

In available-metrics-in-metrics-list.csv, added 6 rebalancing.* metrics from crdb_metrics.yaml.

In metric-names.md, modified to (a) use metrics.yaml for Description, Unit and Type, (b) use crdb_metrics.yaml, shared_metrics.yaml, and tenant_metrics.yaml for Supported Deployments.

Author: florence-crl

src/current/_data/v25.3/metrics/available-metrics-in-metrics-list.csv

@@ -476,4 +476,10 @@ auth.ldap.conn.latency
 auth.password.conn.latency
 auth.scram.conn.latency
 sql.exec.latency.detail
-sql.query.unique.count
+sql.query.unique.count
+rebalancing.cpunanospersecond
+rebalancing.lease.transfers
+rebalancing.range.rebalances
+rebalancing.replicas.cpunanospersecond
+rebalancing.replicas.queriespersecond
+rebalancing.state.imbalanced_overfull_options_exhausted

src/current/_data/v25.3/metrics/export/crdb_metrics.yaml

@@ -0,0 +1,220 @@
+# CrdbCluster-specific metrics. The metrics in this list will be combined with
+# the ones in shared_metrics.yaml.
+#
+# Each metric is mapped to its appropriate aggregation function (sum, max, or mean)
+# which is used to aggregate metrics by specific labels.
+metrics:
+  addsstable_applications: sum
+  addsstable_copies: sum
+  addsstable_proposals: sum
+  admission_wait_sum_kv: sum
+  admission_wait_sum_kv_stores: sum
+  admission_wait_sum_sql_kv_response: sum
+  admission_wait_sum_sql_sql_response: sum
+  capacity: max
+  capacity_available: max
+  capacity_reserved: max
+  capacity_used: max
+  exec_error: sum
+  exec_latency: sum
+  exec_success: sum
+  gcbytesage: mean
+  gossip_bytes_received: sum
+  gossip_bytes_sent: sum
+  gossip_connections_incoming: sum
+  gossip_connections_outgoing: sum
+  gossip_connections_refused: sum
+  gossip_infos_received: sum
+  gossip_infos_sent: sum
+  intentage: mean
+  intentbytes: sum
+  intentcount: sum
+  jobs_row_level_ttl_currently_paused: sum
+  jobs_row_level_ttl_currently_running: sum
+  jobs_row_level_ttl_delete_duration: sum
+  jobs_row_level_ttl_num_active_spans: sum
+  jobs_row_level_ttl_resume_completed: sum
+  jobs_row_level_ttl_resume_failed: sum
+  jobs_row_level_ttl_rows_deleted: sum
+  jobs_row_level_ttl_rows_selected: sum
+  jobs_row_level_ttl_select_duration: sum
+  jobs_row_level_ttl_span_total_duration: sum
+  jobs_row_level_ttl_total_expired_rows: sum
+  jobs_row_level_ttl_total_rows: sum
+  keybytes: sum
+  keycount: sum
+  leases_epoch: sum
+  leases_error: sum
+  leases_expiration: sum
+  leases_success: sum
+  leases_transfers_error: sum
+  leases_transfers_success: sum
+  livebytes: sum
+  livecount: sum
+  liveness_epochincrements: sum
+  liveness_heartbeatfailures: sum
+  liveness_heartbeatlatency: sum
+  liveness_heartbeatsuccesses: sum
+  liveness_livenodes: max
+  physical_replication_logical_bytes: sum
+  physical_replication_replicated_time_seconds: max
+  physical_replication_sst_bytes: sum
+  queue_consistency_pending: sum
+  queue_consistency_process_failure: sum
+  queue_consistency_process_success: sum
+  queue_consistency_processingnanos: sum
+  queue_gc_info_abortspanconsidered: sum
+  queue_gc_info_abortspangcnum: sum
+  queue_gc_info_abortspanscanned: sum
+  queue_gc_info_intentsconsidered: sum
+  queue_gc_info_intenttxns: sum
+  queue_gc_info_numkeysaffected: sum
+  queue_gc_info_pushtxn: sum
+  queue_gc_info_resolvesuccess: sum
+  queue_gc_info_resolvetotal: sum
+  queue_gc_info_transactionspangcaborted: sum
+  queue_gc_info_transactionspangccommitted: sum
+  queue_gc_info_transactionspangcpending: sum
+  queue_gc_info_transactionspanscanned: sum
+  queue_gc_pending: sum
+  queue_gc_process_failure: sum
+  queue_gc_process_success: sum
+  queue_gc_processingnanos: sum
+  queue_raftlog_pending: sum
+  queue_raftlog_process_failure: sum
+  queue_raftlog_process_success: sum
+  queue_raftlog_processingnanos: sum
+  queue_raftsnapshot_pending: sum
+  queue_raftsnapshot_process_failure: sum
+  queue_raftsnapshot_process_success: sum
+  queue_raftsnapshot_processingnanos: sum
+  queue_replicagc_pending: sum
+  queue_replicagc_process_failure: sum
+  queue_replicagc_process_success: sum
+  queue_replicagc_processingnanos: sum
+  queue_replicagc_removereplica: sum
+  queue_replicate_addreplica: sum
+  queue_replicate_pending: sum
+  queue_replicate_process_failure: sum
+  queue_replicate_process_success: sum
+  queue_replicate_processingnanos: sum
+  queue_replicate_purgatory: sum
+  queue_replicate_rebalancereplica: sum
+  queue_replicate_removedeadreplica: sum
+  queue_replicate_removereplica: sum
+  queue_replicate_transferlease: sum
+  queue_split_pending: sum
+  queue_split_process_failure: sum
+  queue_split_process_success: sum
+  queue_split_processingnanos: sum
+  queue_tsmaintenance_pending: sum
+  queue_tsmaintenance_process_failure: sum
+  queue_tsmaintenance_process_success: sum
+  queue_tsmaintenance_processingnanos: sum
+  raft_commandsapplied: sum
+  raft_enqueued_pending: sum
+  raft_heartbeats_pending: sum
+  raft_process_commandcommit_latency: sum
+  raft_process_logcommit_latency: sum
+  raft_process_tickingnanos: sum
+  raft_process_workingnanos: sum
+  raft_rcvd_app: sum
+  raft_rcvd_appresp: sum
+  raft_rcvd_dropped: sum
+  raft_rcvd_heartbeat: sum
+  raft_rcvd_heartbeatresp: sum
+  raft_rcvd_prevote: sum
+  raft_rcvd_prevoteresp: sum
+  raft_rcvd_prop: sum
+  raft_rcvd_snap: sum
+  raft_rcvd_timeoutnow: sum
+  raft_rcvd_transferleader: sum
+  raft_rcvd_vote: sum
+  raft_rcvd_voteresp: sum
+  raft_ticks: sum
+  raftlog_behind: max
+  raftlog_truncated: sum
+  range_adds: sum
+  range_raftleadertransfers: sum
+  range_removes: sum
+  range_snapshots_generated: sum
+  range_splits: sum
+  rangekeybytes: sum
+  rangekeycount: sum
+  ranges: sum
+  ranges_overreplicated: sum
+  ranges_unavailable: sum
+  ranges_underreplicated: sum
+  rangevalbytes: sum
+  rangevalcount: sum
+  rebalancing_cpunanospersecond: mean
+  rebalancing_lease_transfers: sum
+  rebalancing_queriespersecond: sum
+  rebalancing_range_rebalances: sum
+  rebalancing_readbytespersecond: sum
+  rebalancing_readspersecond: sum
+  rebalancing_replicas_cpunanospersecond: sum
+  rebalancing_replicas_queriespersecond: sum
+  rebalancing_requestspersecond: sum
+  rebalancing_state_imbalanced_overfull_options_exhausted: sum
+  rebalancing_writebytespersecond: sum
+  rebalancing_writespersecond: sum
+  replicas: sum
+  replicas_leaders: sum
+  replicas_leaders_not_leaseholders: sum
+  replicas_leaseholders: sum
+  replicas_quiescent: sum
+  replicas_reserved: sum
+  requests_backpressure_split: sum
+  requests_slow_lease: sum
+  requests_slow_raft: sum
+  rocksdb_block_cache_hits: sum
+  rocksdb_block_cache_misses: sum
+  rocksdb_block_cache_pinned_usage: sum
+  rocksdb_block_cache_usage: sum
+  rocksdb_bloom_filter_prefix_checked: sum
+  rocksdb_bloom_filter_prefix_useful: sum
+  rocksdb_compactions: sum
+  rocksdb_flushes: sum
+  rocksdb_memtable_total_size: sum
+  rocksdb_num_sstables: sum
+  rocksdb_read_amplification: mean
+  rocksdb_table_readers_mem_estimate: sum
+  schedules_scheduled_row_level_ttl_executor_failed: sum
+  security_certificate_expiration_ca: max
+  storage_l0_num_files: sum
+  storage_l0_sublevels: sum
+  sys_cgo_allocbytes: sum
+  sys_cgo_totalbytes: sum
+  sys_cgocalls: sum
+  sys_cpu_combined_percent_normalized: mean
+  sys_cpu_sys_ns: sum
+  sys_cpu_sys_percent: mean
+  sys_cpu_user_ns: sum
+  sys_cpu_user_percent: mean
+  sys_fd_open: sum
+  sys_fd_softlimit: max
+  sys_gc_count: sum
+  sys_gc_pause_ns: sum
+  sys_gc_pause_percent: mean
+  sys_go_allocbytes: sum
+  sys_go_totalbytes: sum
+  sys_goroutines: sum
+  sys_host_disk_iopsinprogress: sum
+  sys_host_disk_read_bytes: sum
+  sys_host_disk_read_count: sum
+  sys_host_disk_write_bytes: sum
+  sys_host_disk_write_count: sum
+  sys_host_net_recv_bytes: sum
+  sys_host_net_send_bytes: sum
+  sys_rss: sum
+  sys_runnable_goroutines_per_cpu: mean
+  sys_totalmem: max
+  sysbytes: sum
+  syscount: sum
+  timeseries_write_bytes: sum
+  timeseries_write_errors: sum
+  timeseries_write_samples: sum
+  totalbytes: sum
+  valbytes: sum
+  valcount: sum

src/current/_data/v25.3/metrics/export/shared_metrics.yaml

@@ -0,0 +1,87 @@
+# This shared list serves as a base list for all clusters that use metrics
+# export. It was originally taken from [1], and has been modified for CC's use
+# case over time.
+#
+# In this shared list, we have removed all system and storage related metrics,
+# with the exception of `sys_uptime`. For CRDB or tenant specific metrics, see
+# crdb_metrics.yaml and tenant_metrics.yaml.
+#
+# NOTE: When adding a new metric to this list, please ensure that it applies to
+# both CRDB and tenant clusters, or else they should go into the cluster
+# specific list.
+#
+# Each metric is mapped to its appropriate aggregation function (sum, max, or mean)
+# which is used to aggregate metrics by specific labels.
+#
+# [1]: https://www.cockroachlabs.com/docs/stable/essential-metrics-dedicated
+metrics:
+  changefeed_backfill_count: sum
+  changefeed_backfill_pending_ranges: sum
+  changefeed_commit_latency: sum
+  changefeed_emitted_messages: sum
+  changefeed_error_retries: sum
+  changefeed_failures: sum
+  changefeed_max_behind_nanos: max
+  changefeed_message_size_hist: sum
+  changefeed_running: sum
+  clock_offset_meannanos: mean
+  clock_offset_stddevnanos: mean
+  distsender_batches: sum
+  distsender_batches_partial: sum
+  distsender_errors_notleaseholder: sum
+  distsender_rpc_sent: sum
+  distsender_rpc_sent_local: sum
+  distsender_rpc_sent_nextreplicaerror: sum
+  jobs_changefeed_resume_retry_error: sum
+  requests_slow_distsender: sum
+  round_trip_latency: sum
+  schedules_BACKUP_failed: sum
+  schedules_BACKUP_last_completed_time: max
+  schedules_BACKUP_started: sum
+  schedules_BACKUP_succeeded: sum
+  sql_bytesin: sum
+  sql_bytesout: sum
+  sql_conn_latency: sum
+  sql_conns: sum
+  sql_ddl_count: sum
+  sql_delete_count: sum
+  sql_distsql_contended_queries_count: sum
+  sql_distsql_exec_latency: sum
+  sql_distsql_flows_active: sum
+  sql_distsql_flows_total: sum
+  sql_distsql_queries_active: sum
+  sql_distsql_queries_total: sum
+  sql_distsql_select_count: sum
+  sql_distsql_service_latency: sum
+  sql_exec_latency: sum
+  sql_failure_count: sum
+  sql_full_scan_count: sum
+  sql_insert_count: sum
+  sql_mem_distsql_current: sum
+  sql_mem_distsql_max: max
+  sql_mem_internal_session_current: sum
+  sql_mem_internal_session_max: max
+  sql_mem_internal_txn_current: sum
+  sql_mem_internal_txn_max: max
+  sql_mem_root_current: sum
+  sql_misc_count: sum
+  sql_new_conns: sum
+  sql_query_count: sum
+  sql_select_count: sum
+  sql_service_latency: sum
+  sql_statements_active: sum
+  sql_txn_abort_count: sum
+  sql_txn_begin_count: sum
+  sql_txn_commit_count: sum
+  sql_txn_latency: sum
+  sql_txn_rollback_count: sum
+  sql_txns_open: sum
+  sql_update_count: sum
+  sys_uptime: max
+  txn_aborts: sum
+  txn_commits: sum
+  txn_commits1PC: sum
+  txn_durations: sum
+  txn_restarts: sum
+  txn_restarts_serializable: sum
+  txn_restarts_writetooold: sum

src/current/_data/v25.3/metrics/export/tenant_metrics.yaml

@@ -0,0 +1,27 @@
+# CrdbTenant-specific metrics. The metrics in this list will be combined with
+# the ones in shared_metrics.yaml.
+#
+# Each metric is mapped to an aggregation function that indicates how the metric
+# should be aggregated when needed
+#
+# Note: The aggregation functions here are declared as noop because they are
+# only applied for UA-enabled clusters. For regular tenant clusters, these
+# values are not used but are included for consistency with the overall data
+# structure.
+metrics:
+  tenant_sql_usage_request_units: noop
+  tenant_sql_usage_kv_request_units: noop
+  tenant_sql_usage_read_batches: noop
+  tenant_sql_usage_read_requests: noop
+  tenant_sql_usage_read_bytes: noop
+  tenant_sql_usage_write_batches: noop
+  tenant_sql_usage_write_requests: noop
+  tenant_sql_usage_write_bytes: noop
+  tenant_sql_usage_sql_pods_cpu_seconds: noop
+  tenant_sql_usage_pgwire_egress_bytes: noop
+  tenant_sql_usage_external_io_ingress_bytes: noop
+  tenant_sql_usage_external_io_egress_bytes: noop
+  tenant_sql_usage_cross_region_network_ru: noop
+  tenant_sql_usage_estimated_cpu_seconds: noop
+  tenant_sql_usage_provisioned_vcpus: noop
+  sql_aggregated_livebytes: noop