Home  /  Manage Materialize  /  Monitoring and alerting

Appendix: Metrics

View as Markdown

This page lists the Prometheus metrics exposed by Materialize, along with their descriptions. A * in a metric name denotes a family of metrics whose name is completed at runtime (for example, mz_persist_*_bytes).

Metric Description Labels
*_postgres_connpool_acquire_seconds time spent acquiring connections from pool
*_postgres_connpool_acquires times a connection has been acquired from pool
*_postgres_connpool_available available connections in the pool
*_postgres_connpool_connection_errors number of errors when establishing a new connection
*_postgres_connpool_connections_created times a connection was created
*_postgres_connpool_size number of connections currently in pool
*_postgres_connpool_ttl_reconnections times a connection was recycled due to ttl
*_request_duration_seconds How long it takes for a request to complete in seconds. path, source
*_requests_active Number of currently active/open http requests. path, source
*_requests_total Total number of http requests received since process start. path, source, status
environmentd_needs_update Count of organizations in this cluster which are running outdated pod templates
jemalloc_active Total number of bytes in active pages allocated by the application
jemalloc_allocated Total number of bytes allocated by the application
jemalloc_metadata Total number of bytes dedicated to metadata.
jemalloc_resident Maximum number of bytes in physically resident data pages mapped
jemalloc_retained Total number of bytes in virtual memory mappings
mz_active_copy_tos The number of active COPY TO queries. session_type
mz_active_sessions The number of active coordinator sessions. session_type
mz_active_subscribes The number of active SUBSCRIBE queries. session_type
mz_adapter_commands The total number of adapter commands issued of the given type since process start. application_name, command_type, status
mz_append_table_duration_seconds Latency for appending to any (user or system) table.
mz_apply_catalog_implications_seconds The time it takes to apply catalog implications.
mz_arrangement_maintenance_active_info Whether maintenance is currently occuring. worker_id
mz_arrangement_maintenance_seconds_total The total time spent maintaining arrangements. worker_id
mz_arrangement_sizes_collection_time_seconds Seconds to read mz_object_arrangement_sizes and prepare history-table updates for one snapshot.
mz_arrangement_sizes_rows_written_total Total rows appended to mz_object_arrangement_size_history since process start.
mz_auth_refresh_tasks_active The number of active refresh tasks we have running.
mz_auth_request_count Total number of HTTP requests made to Frontegg for authentication path, status
mz_auth_request_duration_seconds How long it takes for a request to Frontegg to complete in seconds. path
mz_auth_session_refresh_count Total number of authentication sessions that get refreshed. outstanding_receivers, recent_drop
mz_auth_session_request_count Total number of session start requests the Authenticator has received. existing_session
mz_balancer_connection_active Count of currently open network connections. source
mz_balancer_connection_status Count of completed network connections, by status source, status
mz_balancer_metadata_seconds server uptime, labels are build metadata build_type, version
mz_balancer_tenant_connection_active Count of opened network connections by tenant. source, tenant
mz_balancer_tenant_connection_rx Number of bytes received from a client for a tenant. source, tenant
mz_balancer_tenant_connection_tx Number of bytes sent to a client for a tenant. source, tenant
mz_balancer_tenant_pgwire_sni_count Count of pgwire connections that have and do not have SNI available per tenant. has_sni, tenant
mz_bytes_read_total Count of bytes read from sources
mz_canceled_peeks_total The total number of canceled peeks since process start.
mz_catalog_allocate_id_seconds The time it takes to allocate IDs in the durable catalog.
mz_catalog_collection_entries Total number of entries, after consolidation, per catalog collection. collection
mz_catalog_info_metrics_reconcile_seconds Time taken to rebuild the catalog info metrics from a catalog snapshot.
mz_catalog_snapshot_consolidations Count of snapshot consolidation passes.
mz_catalog_snapshot_latency_seconds Total latency for fetching a snapshot of the durable catalog.
mz_catalog_snapshot_max_entries High-water mark of entries in the unconsolidated in-memory snapshot since process start.
mz_catalog_snapshot_seconds The time it takes to run `catalog_snapshot` when fetching the catalog. context
mz_catalog_snapshots_taken Count of snapshots taken.
mz_catalog_sync_latency_seconds Total latency for syncing the in-memory state of the durable catalog with the persisted contents.
mz_catalog_syncs Count of catalog syncs.
mz_catalog_transact_seconds The time it takes to run various catalog transact methods. method
mz_catalog_transaction_commit_latency_seconds Total latency for committing a durable catalog transactions.
mz_catalog_transaction_commits Count of transaction commits.
mz_catalog_transactions_started Total number of started transactions.
mz_check_scheduling_policies_seconds The time each policy in `check_scheduling_policies` takes. policy, thread
mz_cluster_handle_command_duration_seconds Time spent in handling commands. cluster, command_type, worker_id
mz_cluster_info Maps cluster IDs to the cluster's name and size. Constant 1. cluster_id, name, size
mz_cluster_server_last_command_received The time (in seconds since the Unix epoch) at which the server last received data from the controller, including CTP keepalives. Used to detect controller connections that are no longer reachable. server_name
mz_column_pager_budget_configured_bytes Most-recently-configured total budget for the column-pager tiered policy.
mz_column_pager_budget_remaining_bytes Bytes the column-pager tiered policy currently has available for resident columns.
mz_column_pager_paged_bytes_in_total Total uncompressed bytes handed to the pager for pageout, before any codec is applied.
mz_column_pager_paged_bytes_out_total Total on-storage bytes after codec / padding.
mz_column_pager_pagein_bytes_total Total uncompressed bytes delivered by page-in.
mz_column_pager_pageins_total Successful page-ins from `ColumnPager::take`.
mz_column_pager_pageouts_total Pager decisions that paged the chunk out.
mz_column_pager_resident_released_bytes_total Total bytes returned to the budget by ticket drops.
mz_column_pager_resident_released_total Resident-ticket drops returning budget.
mz_column_pager_skip_bytes_total Total bytes kept resident by skip decisions.
mz_column_pager_skip_decisions_total Pager decisions that kept the chunk resident.
mz_compute_collection_count The number and hydration status of maintained compute collections. hydrated, type, worker_id
mz_compute_command_message_bytes_total The total number of bytes sent in compute command messages. instance_id, replica_id
mz_compute_commands_total The total number of compute commands sent. command_type, instance_id, replica_id
mz_compute_controller_collection_count The number of installed compute collections. instance_id
mz_compute_controller_collection_unscheduled_count The number of installed but unscheduled compute collections. instance_id
mz_compute_controller_command_queue_size The size of the compute command queue. instance_id, replica_id
mz_compute_controller_connected_replica_count The number of replicas successfully connected to the compute controller. instance_id
mz_compute_controller_copy_to_count The number of active copy tos. instance_id
mz_compute_controller_history_command_count The number of commands in the controller's command history. command_type, instance_id
mz_compute_controller_history_dataflow_count The number of dataflows in the controller's command history. instance_id
mz_compute_controller_hydration_queue_size The size of the compute hydration queue. instance_id, replica_id
mz_compute_controller_peek_count The number of pending peeks. instance_id
mz_compute_controller_replica_connect_wait_time_seconds_total The total time the compute controller spent waiting for replica (re-)connection. instance_id, replica_id
mz_compute_controller_replica_connects_total The total number of replica (re-)connections made by the compute controller. instance_id, replica_id
mz_compute_controller_replica_count The number of replicas. instance_id
mz_compute_controller_response_recv_count The number of receives on the compute response queue. instance_id
mz_compute_controller_response_send_count The number of sends on the compute response queue. instance_id
mz_compute_controller_subscribe_count The number of active subscribes. instance_id
mz_compute_peek_duration_seconds A histogram of peek durations since restart. instance_id, result
mz_compute_peeks_total The total number of peeks served. instance_id, result
mz_compute_reconciliation_replaced_dataflows_count_total The total number of dataflows that were replaced during compute reconciliation. reason, worker_id
mz_compute_reconciliation_reused_dataflows_count_total The total number of dataflows that were reused during compute reconciliation. worker_id
mz_compute_replica_history_command_count The number of commands in the replica's command history. command_type, worker_id
mz_compute_replica_history_dataflow_count The number of dataflows in the replica's command history. worker_id
mz_compute_response_message_bytes_total The total number of bytes sent in compute response messages. instance_id, replica_id
mz_compute_responses_total The total number of compute responses sent. instance_id, replica_id, response_type
mz_connection_status Count of completed network connections, by status source, status
mz_coord_queue_busy_seconds The number of seconds the coord queue was processing before it was empty. This is a sampled metric and does not measure the full coord queue wait/idle times.
mz_coordinator_message_batch_size Message batch size handled by the coordinator.
mz_dataflow_events_read_total Count of events we have read from the wire format, status
mz_dataflow_replica_expiration_remaining_seconds The remaining seconds until replica expiration. Can go negative, can lag behind. worker_id
mz_dataflow_replica_expiration_timestamp_seconds The replica expiration timestamp in seconds since epoch. worker_id
mz_dataflow_shared_row_heap_capacity_bytes The heap capacity of the shared row. worker_id
mz_dataflow_wallclock_lag_seconds A summary of the second-by-second lag of the dataflow frontier relative to wallclock time, aggregated over the last minute. collection_id, instance_id, quantile, replica_id
mz_dataflow_wallclock_lag_seconds_count The total count of dataflow wallclock lag measurements. collection_id, instance_id, replica_id
mz_dataflow_wallclock_lag_seconds_sum The total sum of dataflow wallclock lag measurements. collection_id, instance_id, replica_id
mz_determine_timestamp The total number of calls to determine_timestamp. compute_instance, isolation_level, respond_immediately
mz_group_commit_catalog_upper_seconds The time it takes to advance the catalog shard upper during group commit.
mz_handle_scheduling_decisions_seconds The time `handle_scheduling_decisions` takes. altered_a_cluster
mz_index_peek_cursor_setup_seconds Time setting up cursor and literal constraints.
mz_index_peek_error_scan_seconds Time scanning the error trace for errors.
mz_index_peek_frontier_check_seconds Time checking trace frontiers.
mz_index_peek_result_sort_seconds Time sorting intermediate results during peek collection.
mz_index_peek_row_collection_seconds Time constructing RowCollection from peek results.
mz_index_peek_row_iteration_seconds Time iterating rows and evaluating MFP.
mz_index_peek_seek_fulfillment_seconds Time in seek_fulfillment method including frontier checks and data collection.
mz_index_peek_total_seconds Total time processing index peeks, from process_peek entry to response. Excluding peeks that use the peek response stash.
mz_kafka_partition_offset_max High watermark offset on broker for partition partition_id, source_id, topic
mz_linearize_message_seconds The number of seconds it takes to linearize strict serializable messages immediately_handled, type
mz_mcp_requests_total Total number of MCP requests received. endpoint_type, method, status
mz_mcp_tool_call_duration_seconds Duration of MCP tools/call invocations in seconds. endpoint_type, tool_name
mz_mcp_tool_calls_total Total number of MCP tools/call invocations. endpoint_type, status, tool_name
mz_memory_limiter_burst_budget_byteseconds The remaining memory burst budget.
mz_memory_limiter_duration_seconds A histogram of the time it took to run the memory limiter.
mz_memory_limiter_memory_limit_bytes The configured memory limit.
mz_memory_limiter_memory_usage_bytes The current memory usage.
mz_memory_limiter_vm_rss_bytes The current VmRSS metric.
mz_memory_limiter_vm_swap_bytes The current VmSwap metric.
mz_metrics_lgalloc_allocations_total Number of region allocations in size class size_class
mz_metrics_lgalloc_area_total_bytes Number of bytes in all areas in size class size_class
mz_metrics_lgalloc_areas_total Number of areas backing size class size_class
mz_metrics_lgalloc_clean_regions_bytes_total Number of clean regions in size class size_class
mz_metrics_lgalloc_clean_regions_total Number of clean regions in size class size_class
mz_metrics_lgalloc_clear_eager_bytes_total Number of thread regions in size class size_class
mz_metrics_lgalloc_clear_eager_total Number of thread regions in size class size_class
mz_metrics_lgalloc_clear_slow_bytes_total Number of thread regions in size class size_class
mz_metrics_lgalloc_clear_slow_total Number of thread regions in size class size_class
mz_metrics_lgalloc_deallocations_total Number of region deallocations for size class size_class
mz_metrics_lgalloc_file_allocated_size_bytes Sum of allocated sizes in size class size_class
mz_metrics_lgalloc_file_size_bytes Sum of file sizes in size class size_class
mz_metrics_lgalloc_free_regions_bytes_total Number of free regions in size class size_class
mz_metrics_lgalloc_free_regions_total Number of free regions in size class size_class
mz_metrics_lgalloc_global_regions_bytes_total Number of global regions in size class size_class
mz_metrics_lgalloc_global_regions_total Number of global regions in size class size_class
mz_metrics_lgalloc_refill_total Number of area refills for size class size_class
mz_metrics_lgalloc_slow_path_total Number of slow path region allocations for size class size_class
mz_metrics_lgalloc_thread_regions_bytes_total Number of thread regions in size class size_class
mz_metrics_lgalloc_thread_regions_total Number of thread regions in size class size_class
mz_metrics_lgalloc_vm_active_bytes Sum of active sizes in size class size_class
mz_metrics_lgalloc_vm_dirty_bytes Sum of dirty sizes in size class size_class
mz_metrics_lgalloc_vm_mapped_bytes Sum of mapped sizes in size class size_class
mz_metrics_libc_ru_inblock_total block input operations
mz_metrics_libc_ru_majflt_total page faults (hard page faults)
mz_metrics_libc_ru_maxrss_bytes maximum resident set size
mz_metrics_libc_ru_minflt_total page reclaims (soft page faults)
mz_metrics_libc_ru_nivcsw_total involuntary context switches
mz_metrics_libc_ru_nvcsw_total voluntary context switches
mz_metrics_libc_ru_oublock_total block output operations
mz_metrics_libc_ru_stime_seconds_total system CPU time used
mz_metrics_libc_ru_utime_seconds_total user CPU time used
mz_metrics_update_duration The time it took to update lgalloc stats name
mz_mysql_per_source_deletes The number of deletes for all tables in this source source_id
mz_mysql_per_source_ignored_messages The number of messages ignored because of an irrelevant type or relation_id source_id
mz_mysql_per_source_inserts The number of inserts for all tables in this source source_id
mz_mysql_per_source_messages_total The total number of replication messages for this source, not expected to be the sum of the other values. source_id
mz_mysql_per_source_tables_count The number of upstream tables for this source source_id
mz_mysql_per_source_updates The number of updates for all tables in this source source_id
mz_mysql_snapshot_count_latency The wall time used to obtain snapshot sizes. schema, source_id, table_name
mz_mysql_sum_gtid_txns The sum of all transaction-ids committed for each GTID Source-ID UUID seen for this source source_id
mz_oauth_protected_resource_metadata_requests_total Total number of requests to the OAuth Protected Resource Metadata endpoint. status
mz_object_info Maps catalog object IDs to the object's name, schema, database, and type. Constant 1. database_name, global_id, name, object_id, schema_name, type
mz_optimization_notices Number of optimization notices per notice type. notice_type
mz_optimizer_e2e_optimization_time_seconds A histogram of end-to-end optimization times since restart. object_type
mz_otel_on_close count of on_close events sent to otel
mz_parameter_frontend_last_cse_time_seconds The last known time when the LaunchDarkly client sent an event to the LaunchDarkly server (as unix timestamp).
mz_parameter_frontend_last_sse_time_seconds The last known time when the LaunchDarkly client received an event from the LaunchDarkly server (as unix timestamp).
mz_parameter_frontend_params_changed The number of parameter changes pulled from the LaunchDarkly frontend.
mz_parse_seconds The time it takes to parse a SQL statement. (Works for both Simple Queries and the Extended Query protocol.)
mz_persist_*_bytes total encoded size of * batches written
mz_persist_*_goodbytes total logical size of * batches written
mz_persist_*_key_lower_too_big count of * writes that were unable to write a key lower, because the size threshold was too low
mz_persist_*_step_inline time spent encoding * inline batches
mz_persist_*_step_part_writing blocking time spent writing parts for * updates
mz_persist_*_step_stats time spent computing * update stats
mz_persist_*_write_batch_order count of batches by the data ordering order
mz_persist_*_write_batch_part_seconds time spent writing * batches
mz_persist_*_write_stall_count count of * writes stalling to await max outstanding reqs
mz_persist_audit_blob_batch_part_bytes total size of batch parts in blob
mz_persist_audit_blob_batch_part_count count of batch parts in blob
mz_persist_audit_blob_bytes total size of blob
mz_persist_audit_blob_count count of all blobs
mz_persist_audit_blob_rollup_bytes total size of state rollups stored in blob
mz_persist_audit_blob_rollup_count count of all state rollups in blob
mz_persist_audit_step_seconds time spent on individual steps of audit step
mz_persist_backpressure_emitted_bytes A counter with the number of emitted bytes. name, shard
mz_persist_backpressure_last_backpressured_bytes The last count of bytes we are waiting to be retired in the operator. This cannot be directly compared to `retired_bytes`, but CAN indicate that backpressure is happening. name, shard
mz_persist_backpressure_retired_bytes A counter with the number of bytes retired by downstream processing. name, shard
mz_persist_blob_cache_evictions count of capacity-based cache evictions cache
mz_persist_blob_cache_hits_blobs count of blobs served via cache instead of s3 cache
mz_persist_blob_cache_hits_bytes total size of blobs served via cache instead of s3 cache
mz_persist_blob_cache_size_blobs count of blobs in the cache cache
mz_persist_blob_cache_size_bytes total size of blobs in the cache cache
mz_persist_blob_failures count of all blob operation failures honeycomb
mz_persist_cmd_cas_mismatch_count count of command retries from CaS mismatch cmd
mz_persist_cmd_compare_and_append_noop count of compare_and_append retries that were discoverd to have already committed
mz_persist_cmd_failed_count count of commands failed cmd
mz_persist_cmd_fetch_upper_count count of fetch_upper calls
mz_persist_cmd_seconds time spent applying commands cmd
mz_persist_cmd_started_count count of commands started cmd
mz_persist_cmd_succeeded_count count of commands succeeded cmd
mz_persist_columnar_op_count number of rows we've run the specified op on in our structured columnar format column, op, result
mz_persist_columnar_part_build_count number of times we've encoded our structured columnar format
mz_persist_columnar_part_build_seconds number of seconds we've spent encoding our structured columnar format
mz_persist_columnar_part_concat_bytes number of bytes we've copied when concatenating updates
mz_persist_compaction_admin_count count of compaction requests that were performed by admin tooling
mz_persist_compaction_applied count of compactions applied to state
mz_persist_compaction_applied_exact_match count of merge results that exactly replaced a SpineBatch
mz_persist_compaction_applied_subset_match count of merge results that replaced a subset of a SpineBatch
mz_persist_compaction_chunks_compacted count of run chunks compacted
mz_persist_compaction_concurrency_waits count of compaction requests that ever blocked due to concurrency limit
mz_persist_compaction_disabled count of total compaction requests dropped because compaction was disabled
mz_persist_compaction_dropped count of total compaction requests dropped due to a full queue
mz_persist_compaction_failed count of compactions failed
mz_persist_compaction_fast_path_eligible count of compaction requests that could have used the fast-path optimization
mz_persist_compaction_memory_violations count of compaction memory requirement violations
mz_persist_compaction_noop count of compactions discarded (obsolete)
mz_persist_compaction_not_all_prefetched count of compactions where not all inputs were prefetched
mz_persist_compaction_not_applied_too_many_updates count of merge results that did not apply due to too many updates
mz_persist_compaction_parts_prefetched count of compaction parts completely prefetched by the time they're needed
mz_persist_compaction_parts_waited count of compaction parts that had to be waited on
mz_persist_compaction_queued_seconds time that compaction requests spent queued
mz_persist_compaction_requested count of total compaction requests
mz_persist_compaction_runs_compacted count of runs compacted
mz_persist_compaction_schema_selection count of compactions and how we did schema selection selection
mz_persist_compaction_seconds time spent in compaction
mz_persist_compaction_skipped count of compactions skipped due to heuristics
mz_persist_compaction_started count of compactions started
mz_persist_compaction_step_seconds time spent on individual steps of compaction step
mz_persist_compaction_timed_out count of compactions that timed out
mz_persist_consensus_failures count of determinate consensus operation failures honeycomb
mz_persist_consolidation_parts_fetched_count count of parts that were fetched and used during consolidation
mz_persist_consolidation_parts_skipped_count count of parts that were never needed during consolidation
mz_persist_consolidation_parts_wasted_count count of parts that were fetched but not needed during consolidation
mz_persist_consolidation_wrong_sort_count count of runs that were sorted using the wrong ordering for the current consolidation
mz_persist_decode_count count of op decodes op
mz_persist_decode_seconds time spent in op decodes op
mz_persist_encode_count count of op encodes op
mz_persist_encode_seconds time spent in op encodes op
mz_persist_external_blob_delete_noop_count count of blob delete calls that deleted a non-existent key
mz_persist_external_blob_sizes histogram of blob sizes at put time
mz_persist_external_bytes_count total size represented by external service calls op
mz_persist_external_consensus_truncated_count count of versions deleted by consensus truncate calls
mz_persist_external_failed_count count of external service calls failed op
mz_persist_external_op_latency rountrip latency observed by individual performance-critical operations op
mz_persist_external_rtt_latency roundtrip-time to external service as seen by this process external
mz_persist_external_seconds time spent in external service calls op
mz_persist_external_started_count count of external service calls started op
mz_persist_external_succeeded_count count of external service calls succeeded op
mz_persist_gc_finished count of garbage collections finished
mz_persist_gc_merged_reqs count of garbage collection requests merged
mz_persist_gc_noop count of garbage collections skipped because they were already done
mz_persist_gc_seconds time spent in garbage collections
mz_persist_gc_started count of garbage collections started
mz_persist_gc_step_seconds time spent on individual steps of gc step
mz_persist_inline_part_commit_bytes total size of of inline parts committed to state
mz_persist_inline_part_commit_count count of inline parts committed to state
mz_persist_lease_dropped_part count of LeasedBatchParts that were dropped without being politely returned
mz_persist_lease_timeout_read count of readers whose lease timed out
mz_persist_lock_acquire_count count of locks acquired op
mz_persist_lock_blocking_acquire_count count of locks acquired that required blocking op
mz_persist_lock_blocking_seconds time spent blocked for a lock op
mz_persist_metadata_seconds server uptime, labels are build metadata build_type, version
mz_persist_parquet_column_size size in bytes of a column within a parquet file col, compressed
mz_persist_parquet_elided_null_buffer_count times we dropped an unnecessary null buffer returned by parquet decoding
mz_persist_parquet_encoded_size encoded size of a parquet file that we write to S3 format
mz_persist_parquet_row_group_count count of row groups in a parquet file format
mz_persist_peek_seconds Time spent in (experimental) Persist fast-path peeks. worker_id
mz_persist_pubsub_client_approx_diff_apply_latency_seconds histogram of (approximate) latency between sending a diff and applying it
mz_persist_pubsub_client_call_bytes_sent number of bytes sent for a given pubsub client call call
mz_persist_pubsub_client_call_failed times a pubsub client call failed call
mz_persist_pubsub_client_call_received times a pubsub client call was received call
mz_persist_pubsub_client_call_succeeded times a pubsub client call succeeded call
mz_persist_pubsub_client_grpc_broadcast_recv_lagged_count times a message was missed by broadcast receiver due to lag
mz_persist_pubsub_client_grpc_connect_call_attempt_count count of connection call attempts (including retries) to pubsub server
mz_persist_pubsub_client_grpc_connected whether the grpc client is currently connected
mz_persist_pubsub_client_grpc_connection_established_count count of grpc connection establishments to pubsub server
mz_persist_pubsub_client_grpc_error_count count of grpc errors received
mz_persist_pubsub_client_receiver_state_push_diff_fast_path count fast-path state push_diff calls
mz_persist_pubsub_client_receiver_state_push_diff_slow_path_failed count of unsuccessful slow-path state push_diff calls
mz_persist_pubsub_client_receiver_state_push_diff_slow_path_succeeded count of successful slow-path state push_diff calls
mz_persist_pubsub_server_active_connections number of active connections to server
mz_persist_pubsub_server_broadcasted_diff_bytes count of total broadcast diff bytes sent
mz_persist_pubsub_server_broadcasted_diff_count count of total broadcast diff messages sent
mz_persist_pubsub_server_broadcasted_diff_dropped_channel_full count of diffs dropped due to full connection channel
mz_persist_pubsub_server_call_count count of each pubsub server message received call
mz_persist_pubsub_server_operation_seconds time spent in pubsub server performing each operation op
mz_persist_pushdown_parts_audited_bytes total size of parts fetched only for pushdown audit
mz_persist_pushdown_parts_audited_count count of parts fetched only for pushdown audit
mz_persist_pushdown_parts_faked_bytes total size of parts replaced with fakes by aggressive projection pushdown
mz_persist_pushdown_parts_faked_count count of parts faked because of aggressive projection pushdown
mz_persist_pushdown_parts_fetched_bytes total size of parts not filtered by pushdown in bytes
mz_persist_pushdown_parts_fetched_count count of parts not filtered by pushdown
mz_persist_pushdown_parts_filtered_bytes total size of parts filtered by pushdown in bytes
mz_persist_pushdown_parts_filtered_count count of parts filtered by pushdown
mz_persist_pushdown_parts_inline_bytes total size of parts not fetched because they were inline
mz_persist_pushdown_parts_inline_count count of parts not fetched because they were inline
mz_persist_pushdown_parts_mismatched_stats_count number of parts read with unexpectedly the incorrect type of stats
mz_persist_pushdown_parts_projection_trimmed_bytes total bytes trimmed from columnar data because of projection pushdown
mz_persist_pushdown_parts_stats_trimmed_bytes total bytes trimmed from part stats
mz_persist_pushdown_parts_stats_trimmed_count count of trimmed part stats
mz_persist_read_batch_part_bytes total encoded size of batch parts read op
mz_persist_read_batch_part_count count of batch parts read op
mz_persist_read_batch_part_goodbytes total logical size of batch parts read op
mz_persist_read_batch_part_seconds time spent reading batch parts op
mz_persist_read_ts_rewite count of updates read with rewritten ts op
mz_persist_retry_finished_count count of retry loops finished op
mz_persist_retry_retries_count count of total attempts by retry loops op
mz_persist_retry_sleep_seconds time spent in retry loop backoff op
mz_persist_retry_started_count count of retry loops started op
mz_persist_s3_connect_timeouts number of timeouts establishing a connection to S3
mz_persist_s3_errors errors code, op
mz_persist_s3_operation_attempt_timeouts number of operation attempt timeouts (within a single retry)
mz_persist_s3_operation_timeouts number of operation timeouts (including retries)
mz_persist_s3_operations number of raw s3 calls on behalf of Blob interface methods op
mz_persist_s3_read_timeouts number of timeouts waiting on first response byte from S3
mz_persist_schema_cache_added_count count of schema cache entries added op
mz_persist_schema_cache_cached_count count of schema cache entries served from cache op
mz_persist_schema_cache_computed_count count of schema cache entries computed op
mz_persist_schema_cache_dropped_count count of schema cache entries dropped op
mz_persist_schema_cache_fetch_state_count count of state fetches by the schema cache
mz_persist_schema_cache_unavailable_count count of schema cache entries unavailable at current state op
mz_persist_schema_migration_count count of fetch part migrations op
mz_persist_schema_migration_len count of migrated update records op
mz_persist_schema_migration_migrate_seconds seconds spent applying migration logic
mz_persist_schema_migration_new_count count of migrations constructed
mz_persist_schema_migration_new_seconds seconds spent constructing migration logic
mz_persist_semaphore_acquire_count count of acquire calls (not acquired permits count) name
mz_persist_semaphore_acquired_permits total sum of acquired permits name
mz_persist_semaphore_available_permits currently available permits according to the semaphore
mz_persist_semaphore_blocking_count count of acquire calls that had to block name
mz_persist_semaphore_blocking_seconds total time spent blocking on permit acquisition name
mz_persist_semaphore_released_permits total sum of released permits name
mz_persist_shard_batch_part_count count of batch parts by shard name, shard
mz_persist_shard_batch_part_version_bytes total bytes in batch parts by shard and version name, shard, version
mz_persist_shard_batch_part_version_count count of batch parts by shard and version name, shard, version
mz_persist_shard_blob_gets number of Blob::get calls for this shard name, shard
mz_persist_shard_blob_sets number of Blob::set calls for this shard name, shard
mz_persist_shard_cmd_succeeded count of commands succeeded by shard name, shard
mz_persist_shard_compact_batches number of fully compact batches in the shard name, shard
mz_persist_shard_compacting_batches number of batches in the shard with compactions in progress name, shard
mz_persist_shard_compaction_applied count of compactions applied to state by shard name, shard
mz_persist_shard_count count of all active shards on this process
mz_persist_shard_diff_size_bytes total encoded diff size by shard name, shard
mz_persist_shard_gc_finished count of garbage collections finished by shard name, shard
mz_persist_shard_gc_live_diffs the number of diffs (or, alternatively, the number of seqnos) present in consensus state at GC time name, shard
mz_persist_shard_gc_seqno_held_parts count of parts referenced by some live state but not the current state (ie. parts kept only to satisfy seqno holds) at GC time name, shard
mz_persist_shard_hollow_batch_count count of hollow batches by shard name, shard
mz_persist_shard_inline_backpressure_count count of CaA attempts retried because of inline backpressure name, shard
mz_persist_shard_inline_part_bytes total size of parts inline in shard metadata name, shard
mz_persist_shard_inline_part_count count of parts inline in shard metadata name, shard
mz_persist_shard_largest_batch_size largest encoded batch size by shard name, shard
mz_persist_shard_live_writers number of writers that have recently appended updates to this shard name, shard
mz_persist_shard_noncompact_batches number of batches in the shard that aren't compact and have no ongoing compaction name, shard
mz_persist_shard_pubsub_diff_applied number of diffs received via pubsub that applied name, shard
mz_persist_shard_pubsub_diff_not_applied_out_of_order number of diffs received via pubsub that did not apply due to out-of-order delivery name, shard
mz_persist_shard_pubsub_diff_not_applied_stale number of diffs received via pubsub that did not apply due to staleness name, shard
mz_persist_shard_rewrite_part_count count of batch parts with rewrites by shard name, shard
mz_persist_shard_rollup_count count of rollups by shard name, shard
mz_persist_shard_rollup_size_bytes total encoded rollup size by shard name, shard
mz_persist_shard_schema_registry_version_count count of versions in the schema registry name, shard
mz_persist_shard_seqnos_held maximum count of gc-ineligible states by shard name, shard
mz_persist_shard_seqnos_since_last_rollup count of seqnos since last rollup name, shard
mz_persist_shard_since since by shard name, shard
mz_persist_shard_spine_batch_count count of spine batches by shard name, shard
mz_persist_shard_stale_version indicates whether the current version of the shard is less than the current version of the code name, shard
mz_persist_shard_unconsolidated_snapshot in snapshot_and_read, the number of times consolidating the raw data wasn't enough to produce consolidated output name, shard
mz_persist_shard_update_count count of updates by shard name, shard
mz_persist_shard_upper upper by shard name, shard
mz_persist_shard_usage_current_state_batches_bytes data in batches/parts referenced by current version of state name, shard
mz_persist_shard_usage_current_state_rollups_bytes data in rollups referenced by current version of state name, shard
mz_persist_shard_usage_leaked_bytes data reclaimable by a leaked blob detector name, shard
mz_persist_shard_usage_not_leaked_not_referenced_bytes data written by an active writer but not referenced by any version of state name, shard
mz_persist_shard_usage_referenced_not_current_state_bytes data referenced only by a previous version of state name, shard
mz_persist_sink_correction_capacity_decreases_total The cumulative capacity decreases observed on the correction buffer across workers and persist sinks.
mz_persist_sink_correction_capacity_increases_total The cumulative capacity increases observed on the correction buffer across workers and persist sinks.
mz_persist_sink_correction_deletions_total The cumulative deletions observed on the correction buffer across workers and persist sinks.
mz_persist_sink_correction_insertions_total The cumulative insertions observed on the correction buffer across workers and persist sinks.
mz_persist_sink_correction_max_per_sink_worker_capacity_updates The maximum capacity observed for the correction buffer of any single persist sink per worker. worker_id
mz_persist_sink_correction_max_per_sink_worker_len_updates The maximum length observed for the correction buffer of any single persist sink per worker. worker_id
mz_persist_state_apply_spine_fast_path count of spine diff applications that hit the fast path
mz_persist_state_apply_spine_flattened count of spine diff applications that flatten the trace
mz_persist_state_apply_spine_slow_path count of spine diff applications that hit the slow path
mz_persist_state_apply_spine_slow_path_lenient count of spine diff applications that hit the lenient compaction apply path
mz_persist_state_apply_spine_slow_path_lenient_adjustment count of adjustments made by the lenient compaction apply path
mz_persist_state_apply_spine_slow_path_with_reconstruction count of spine diff applications that hit the slow path with extra spine reconstruction step
mz_persist_state_fetch_recent_live_diffs_fast_path count of fetch_recent_live_diffs that hit the fast path
mz_persist_state_fetch_recent_live_diffs_slow_path count of fetch_recent_live_diffs that hit the slow path
mz_persist_state_force_applied_hostname count of when hostname diffs needed to be force applied
mz_persist_state_rollup_at_seqno_migration count of fetch_rollup_at_seqno calls that only worked because of the migration
mz_persist_state_rollup_write_noop count of no-op rollup writes reason
mz_persist_state_rollup_write_success count of rollups written successful (may not be linked in to state)
mz_persist_state_update_state_empty_path count of state update applications that found no new updates
mz_persist_state_update_state_fast_path count of state update applications that hit the fast path
mz_persist_state_update_state_noop_path count of state update applications that no-oped due to shared state
mz_persist_state_update_state_slow_path count of state update applications that hit the slow path
mz_persist_state_writer_added count of writers added to the state
mz_persist_state_writer_removed count of writers removed from the state
mz_persist_task_total_idle_duration Seconds of time spent idling, ie. waiting for a task to be woken up. name
mz_persist_task_total_idled_count The total number of task idles. Useful for computing the average idle time. name
mz_persist_task_total_scheduled_count The total number of task schedules. Useful for computing the average scheduled time. name
mz_persist_task_total_scheduled_duration Seconds of time spent scheduled, ie. ready to poll but not yet polled. name
mz_persist_wait_resolved_via_sleep count of wait-for-uppers resolved via sleep
mz_persist_wait_resolved_via_watch count of wait-for-uppers resolved via watch notify
mz_persist_wait_woken_via_sleep count of wait-for-uppers wakes via sleep
mz_persist_wait_woken_via_watch count of wait-for-uppers wakes via watch notify
mz_persist_watch_notify_lagged count of lagged events in the watch notification broadcast channel
mz_persist_watch_notify_noop count of watch notifications sent to an broadcast channel
mz_persist_watch_notify_recv count of watch notifications received from the broadcast channel
mz_persist_watch_notify_sent count of watch notifications sent to a non-empty broadcast channel
mz_persist_watch_notify_wait_finished count of watch wait calls resolved
mz_persist_watch_notify_wait_started count of watch wait calls started
mz_pgwire_ensure_transaction_seconds The time it takes to run `ensure_transactions` when processing pgwire messages. message_type
mz_pgwire_message_processing_seconds The time it takes to process each of the pgwire message types, measured in the Adapter frontend message_type
mz_pgwire_recv_scheduling_delay_ms The time between a pgwire connection's receiver task being woken up by incoming data and getting polled. message_type
mz_postgres_per_source_deletes The number of deletes for all tables in this source source_id
mz_postgres_per_source_ignored_messages The number of messages ignored because of an irrelevant type or relation_id source_id
mz_postgres_per_source_inserts The number of inserts for all tables in this source source_id
mz_postgres_per_source_messages_total The total number of replication messages for this source, not expected to be the sum of the other values. source_id
mz_postgres_per_source_tables_count The number of upstream tables for this source source_id
mz_postgres_per_source_transactions_total The number of committed transactions for all tables in this source source_id
mz_postgres_per_source_updates The number of updates for all tables in this source source_id
mz_postgres_per_source_wal_lsn LSN of the latest transaction committed for this source, see Postgres Replication docs for more details on LSN source_id
mz_postgres_snapshot_count_latency The wall time used to obtain snapshot sizes. source_id, table_name
mz_query_total The total number of queries issued of the given type since process start. session_type, statement_type
mz_replica_info Maps cluster replica IDs to the replica's name and size. Constant 1. cluster_id, name, replica_id, size
mz_result_rows_first_to_last_byte_seconds The time from just before sending the first result row to sending a final response message after having successfully flushed the last result row to the connection. (This can span multiple FETCH statements.) (This is never observed for unbounded SUBSCRIBEs, i.e., which have no last result row.) statement_type
mz_resume_upper The timestamp-domain resumption frontier chosen for a source's ingestion source_id
mz_row_set_finishing_seconds The time it takes to run RowSetFinishing::finish.
mz_session_startup_table_writes_seconds If we had to wait for builtin table writes before processing a query, how long did we wait for.
mz_shard_finalization_op_failed count of shard finalization operations that failed
mz_shard_finalization_op_started count of shard finalization operations that have started
mz_shard_finalization_op_succeeded count of shard finalization operations that succeeded
mz_shard_finalization_outstanding count of shards in need of finalization
mz_shard_finalization_pending_commit count of shards for which finalization has completed but has not yet been durably recorded
mz_sink_bytes_committed The number of bytes committed to the sink. sink_id, worker_id
mz_sink_bytes_staged The number of bytes staged but possibly not committed to the sink. sink_id, worker_id
mz_sink_consumed_progress_records The number of progress records consumed by the sink. sink_id
mz_sink_iceberg_commit_conflicts Number of commit conflicts in the iceberg sink sink_id, worker_id
mz_sink_iceberg_commit_duration_seconds Time spent committing batches to Iceberg in seconds sink_id, worker_id
mz_sink_iceberg_commit_failures Number of commit failures in the iceberg sink sink_id, worker_id
mz_sink_iceberg_data_files_written Number of data files written by the iceberg sink sink_id, worker_id
mz_sink_iceberg_delete_files_written Number of delete files written by the iceberg sink sink_id, worker_id
mz_sink_iceberg_snapshots_committed Number of snapshots committed by the iceberg sink sink_id, worker_id
mz_sink_iceberg_stashed_rows Number of stashed rows in the iceberg sink sink_id, worker_id
mz_sink_iceberg_writer_close_duration_seconds Time spent closing Iceberg DeltaWriters in seconds sink_id, worker_id
mz_sink_info Maps user sink IDs to the sink's type, envelope type, and cluster. Constant 1. cluster_id, envelope_type, sink_id, type
mz_sink_messages_committed The number of messages committed to the sink. sink_id, worker_id
mz_sink_messages_staged The number of messages staged but possibly not committed to the sink. sink_id, worker_id
mz_sink_oustanding_progress_records The number of outstanding progress records that need to be read before the sink can resume. sink_id
mz_sink_partition_count The number of partitions this sink is publishing to. sink_id
mz_sink_rdkafka_connects The number of connection attempts, including successful and failed attempts, and name resolution failures across all brokers. sink_id
mz_sink_rdkafka_disconnects The number of disconnections, whether triggered by the broker, the network, the load balancer, or something else across all brokers. sink_id
mz_sink_rdkafka_msg_cnt The current number of messages in producer queues. sink_id
mz_sink_rdkafka_msg_size The current total size of messages in producer queues. sink_id
mz_sink_rdkafka_outbuf_cnt The number of requests awaiting transmission across all brokers. sink_id
mz_sink_rdkafka_outbuf_msg_cnt The number of messages awaiting transmission across all brokers. sink_id
mz_sink_rdkafka_req_timeouts The total number of requests that timed out across all brokers. sink_id
mz_sink_rdkafka_tx The total number of requests sent to brokers. sink_id
mz_sink_rdkafka_tx_bytes The total number of bytes transmitted to brokers. sink_id
mz_sink_rdkafka_txerrs The total number of transmission errors across all brokers. sink_id
mz_sink_rdkafka_txmsg_bytes The total number of bytes transmitted (produced) to brokers. sink_id
mz_sink_rdkafka_txmsgs The total number of messages transmitted (produced) to brokers. sink_id
mz_sink_rdkafka_txretries The total number of request retries across all brokers. sink_id
mz_sink_rdkafka_waitresp_cnt The number of requests in-flight across all brokers that are awaiting a response. sink_id
mz_sink_rdkafka_waitresp_msg_cnt The number of messages in-flight across all brokers that are awaiting a response. sink_id
mz_slow_message_handling Latency for ALL coordinator messages. 'slow' is in the name for legacy reasons, but is not accurate. message_kind
mz_source_bytes_indexed The number of bytes of the source envelope state kept. This will be specific to the envelope in use. parent_source_id, shard_id, source_id, worker_id
mz_source_bytes_received The number of bytes worth of messages the worker has received from upstream. The way the bytes are counted is source-specific. parent_source_id, source_id, worker_id
mz_source_commit_upper_accepted_times The number of accepted remap bindings that are held in the reclock commit upper operator. source_id, worker_id
mz_source_commit_upper_ready_times The number of ready remap bindings that are held in the reclock commit upper operator. source_id, worker_id
mz_source_envelope_state_tombstones The number of outstanding tombstones in the source envelope state. This will be specific to the envelope in use parent_source_id, shard_id, source_id, worker_id
mz_source_error_inserts A counter representing the actual number of errors being inserted to the data shard shard, source_id, worker_id
mz_source_error_retractions A counter representing the actual number of errors being retracted from the data shard shard, source_id, worker_id
mz_source_info Maps user source IDs to the source's type, envelope type, and cluster. Constant 1. cluster_id, envelope_type, source_id, type
mz_source_messages_received The number of raw messages the worker has received from upstream. parent_source_id, source_id, worker_id
mz_source_offset_commit_failures A counter representing how many times we have failed to commit offsets for a source source_id
mz_source_offset_committed The total number of _values_ (source-defined unit) we have fully processed, and storage and committed. shard_id, source_id, worker_id
mz_source_offset_known The total number of _values_ (source-defined unit) present in upstream. shard_id, source_id, worker_id
mz_source_processed_batches A counter representing the number of persist sink batches with actual data we have successfully processed. shard, source_id, worker_id
mz_source_progress A timestamp gauge representing forward progess in the data shard shard, source_id, worker_id
mz_source_records_indexed The number of records in the source envelope state. This will be specific to the envelope in use parent_source_id, shard_id, source_id, worker_id
mz_source_rehydration_latency_ms The amount of time in milliseconds it took for the worker to rehydrate the source envelope state. This will be specific to the envelope in use. envelope, parent_source_id, shard_id, source_id, worker_id
mz_source_row_inserts A counter representing the actual number of rows being inserted to the data shard shard, source_id, worker_id
mz_source_row_retractions A counter representing the actual number of rows being retracted from the data shard shard, source_id, worker_id
mz_source_snapshot_committed Whether or not the worker has committed the initial snapshot for a source. parent_source_id, shard_id, source_id, worker_id
mz_source_snapshot_records_known The total number of records in the source's snapshot parent_source_id, shard_id, source_id, worker_id
mz_source_snapshot_records_staged The total number of records read from the source's snapshot parent_source_id, shard_id, source_id, worker_id
mz_source_updates_committed The number of updates (inserts + deletes) the worker has committed into the storage layer. parent_source_id, shard_id, source_id, worker_id
mz_source_updates_staged The number of updates (inserts + deletes) the worker has written but not yet committed to the storage layer. parent_source_id, shard_id, source_id, worker_id
mz_sql_server_per_source_deletes The number of deletes for all tables in this source source_id, worker_id
mz_sql_server_per_source_ignored_messages The number of messages ignored because of an irrelevant type or relation_id source_id, worker_id
mz_sql_server_per_source_inserts The number of inserts for all tables in this source source_id, worker_id
mz_sql_server_per_source_updates The number of updates for all tables in this source source_id, worker_id
mz_sql_server_snapshot_count_latency The wall time used to obtain snapshot sizes. source_id, table_name, worker_id
mz_sql_server_snapshot_table_count The number of tables that SQL Server still needs to snapshot source_id, worker_id
mz_sql_server_snapshot_table_lock The upstream tables locked for snapshot. source_id, table_name, worker_id
mz_start_time_environmentd Time in milliseconds from environmentd start until the adapter is ready. build_type, version
mz_stashed_peek_seconds Time spent reading a peek result and stashing it in the peek result stash (aka. persist blob). worker_id
mz_statement_logging_actual_bytes The total amount of SQL text that was logged by statement logging.
mz_statement_logging_record_count The total number of SQL statements tagged with whether or not they were recorded. sample
mz_statement_logging_unsampled_bytes The total amount of SQL text that would have been logged if statement logging were unsampled.
mz_storage_command_message_bytes_total The total number of bytes sent in storage command messages. instance_id, replica_id
mz_storage_commands_total The total number of storage commands sent. command_type, instance_id, replica_id
mz_storage_controller_connected_replica_count The number of replicas successfully connected to the storage controller. instance_id
mz_storage_controller_history_command_count The number of commands in the controller's command history. command_type, instance_id
mz_storage_controller_replica_connect_wait_time_seconds_total The total time the storage controller spent waiting for replica (re-)connection. instance_id, replica_id
mz_storage_controller_replica_connects_total The total number of replica (re-)connections made by the storage controller. instance_id, replica_id
mz_storage_regressed_offset_known number of regressed offset_known stats for this id id
mz_storage_response_message_bytes_total The total number of bytes received in storage response messages. instance_id, replica_id
mz_storage_responses_total The total number of storage responses received. instance_id, replica_id, response_type
mz_storage_rocksdb_multi_get_count_total The number of calls to rocksdb multi_get. source_id, worker_id
mz_storage_rocksdb_multi_get_latency The latencies, in fractional seconds, of getting batches of values from RocksDB for this source. source_id
mz_storage_rocksdb_multi_get_result_bytes_total The total size of records returned, when getting batches of values from RocksDB for this source. source_id, worker_id
mz_storage_rocksdb_multi_get_result_count_total The number of non-empty records returned, when getting batches of values from RocksDB for this source. source_id, worker_id
mz_storage_rocksdb_multi_get_size_total The batch size, of getting batches of values from RocksDB for this source. source_id, worker_id
mz_storage_rocksdb_multi_put_count_total The number of calls to rocksdb multi_put. source_id, worker_id
mz_storage_rocksdb_multi_put_latency The latencies, in fractional seconds, of putting batches of values into RocksDB for this source. source_id
mz_storage_rocksdb_multi_put_size_total The batch size, of putting batches of values into RocksDB for this source. source_id, worker_id
mz_storage_upsert_backpressure_emitted_bytes A counter with the number of emitted bytes. source_id, worker_id
mz_storage_upsert_backpressure_last_backpressured_bytes The last count of bytes we are waiting to be retired in the operator. This cannot be directly compared to `retired_bytes`, but CAN indicate that backpressure is happening. source_id, worker_id
mz_storage_upsert_backpressure_retired_bytes A counter with the number of bytes retired by downstream processing. source_id, worker_id
mz_storage_upsert_deletes_total The number of deletes done by the upsert operator. source_id, worker_id
mz_storage_upsert_inserts_total The number of inserts done by the upsert operator source_id, worker_id
mz_storage_upsert_merge_snapshot_deletes_total The number of deletes in a batch for merging snapshot updates for this source. source_id, worker_id
mz_storage_upsert_merge_snapshot_inserts_total The number of inserts in a batch for merging snapshot updates for this source. source_id, worker_id
mz_storage_upsert_merge_snapshot_latency The latencies, in fractional seconds, of merging snapshot updates into upsert state for this source. Specific implementations of upsert state may have more detailed metrics about sub-batches. source_id
mz_storage_upsert_merge_snapshot_updates_total The batch size, of merging snapshot updates into upsert state for this source. Specific implementations of upsert state may have more detailed metrics about sub-batches. source_id, worker_id
mz_storage_upsert_multi_get_latency The latencies, in fractional seconds, of getting values from the upsert state for this source. Specific implementations of upsert state may have more detailed metrics about sub-batches. source_id
mz_storage_upsert_multi_get_result_bytes_total The total size of records returned in a multi_get batch. Specific implementations of upsert state may have more detailed metrics about sub-batches. source_id, worker_id
mz_storage_upsert_multi_get_result_count_total The number of non-empty records returned in a multi_get batch. Specific implementations of upsert state may have more detailed metrics about sub-batches. source_id, worker_id
mz_storage_upsert_multi_get_size_total The batch size, of getting values from the upsert state for this source. Specific implementations of upsert state may have more detailed metrics about sub-batches. source_id, worker_id
mz_storage_upsert_multi_put_latency The latencies, in fractional seconds, of getting values into the upsert state for this source. Specific implementations of upsert state may have more detailed metrics about sub-batches. source_id
mz_storage_upsert_multi_put_size_total The batch size, of getting values into the upsert state for this source. Specific implementations of upsert state may have more detailed metrics about sub-batches. source_id, worker_id
mz_storage_upsert_state_rehydration_latency The latency, per-worker, in fractional seconds, of rehydrating the upsert state for this source source_id, worker_id
mz_storage_upsert_state_rehydration_total The number of values per-worker, rehydrated into the upsert state for this source source_id, worker_id
mz_storage_upsert_state_rehydration_updates The number of updates (both negative and positive), per-worker, rehydrated into the upsert state for this source source_id, worker_id
mz_storage_upsert_updates_total The number of updates done by the upsert operator source_id, worker_id
mz_storage_usage_collection_time_seconds The number of seconds the coord spends collecting usage metrics from storage.
mz_subscribe_outputs The total number of different subscribe outputs used session_type, subscribe_output
mz_subscribe_snapshots_skipped_total The number of collection snapshots that were skipped by the subscribe snapshot optimization.
mz_time_to_first_row_seconds Latency of an execute for a successful query from pgwire's perspective application_name, instance_id, isolation_level, strategy
mz_timely_step_duration_seconds The time spent in each compute step_or_park call cluster, worker_id
mz_timestamp_difference_for_bounded_staleness_ms How much older bounded-staleness timestamps are compared to serializable, in milliseconds. Measures the actual staleness incurred. compute_instance
mz_timestamp_difference_for_strict_serializable_ms Difference in timestamp in milliseconds for running in strict serializable vs serializable isolation level. compute_instance
mz_tokio_blocking_queue_depth The number of tasks currently scheduled in the blocking thread pool, spawned using spawn_blocking. runtime
mz_tokio_budget_forced_yield_count The number of times that tasks have been forced to yield back to the scheduler after exhausting their task budgets. runtime
mz_tokio_global_queue_depth The number of tasks currently scheduled in the runtime's global queue. runtime
mz_tokio_num_alive_tasks The current number of alive tasks in the runtime. runtime
mz_tokio_num_blocking_threads The number of additional threads spawned by the runtime. runtime
mz_tokio_num_idle_blocking_threads The number of idle threads which have spawned by the runtime for spawn_blocking calls. runtime
mz_tokio_num_workers The number of worker threads used by the runtime. runtime
mz_tokio_remote_schedule_count The number of tasks scheduled from outside of the runtime. runtime
mz_tokio_spawned_tasks_count The number of tasks spawned in this runtime since it was created. runtime
mz_tokio_worker_local_queue_depth The number of tasks currently scheduled in the workers' local queues. runtime
mz_tokio_worker_local_schedule_count The number of tasks scheduled from within the runtime on the given worker's local queue. runtime
mz_tokio_worker_mean_poll_time The mean duration of task polls in seconds. runtime
mz_tokio_worker_noop_count The number of times the given worker thread unparked but performed no work before parking again. runtime
mz_tokio_worker_overflow_count The number of times the given worker thread saturated its local queue. runtime
mz_tokio_worker_park_count The total number of times the worker threads have parked. runtime
mz_tokio_worker_park_unpark_count The total number of times the worker threads have parked and unparked. runtime
mz_tokio_worker_poll_count The number of tasks the given worker thread has polled. runtime
mz_tokio_worker_steal_count The number of tasks the given worker thread stole from another worker thread. runtime
mz_tokio_worker_steal_operations The number of times the given worker thread stole tasks from another worker thread. runtime
mz_tokio_worker_total_busy_duration The amount of time the worker threads have been busy, in seconds. runtime
mz_ts_oracle_batched_op_count count of batched operations op
mz_ts_oracle_batches_count count of batches of operations op
mz_ts_oracle_failed_count count of oracle operations failed op
mz_ts_oracle_retry_finished_count count of retry loops finished op
mz_ts_oracle_retry_retries_count count of total attempts by retry loops op
mz_ts_oracle_retry_sleep_seconds time spent in retry loop backoff op
mz_ts_oracle_retry_started_count count of retry loops started op
mz_ts_oracle_seconds time spent in oracle operations op
mz_ts_oracle_started_count count of oracle operations started op
mz_ts_oracle_succeeded_count count of oracle operations succeeded op
mz_txn_batch_commit_bytes total bytes committed via txn
mz_txn_batch_commit_count count of batches committed via txn
mz_txn_batch_unapplied_bytes total bytes committed via txn but not yet applied
mz_txn_batch_unapplied_count count of batches committed via txn but not yet applied
mz_txn_batch_unapplied_min_ts minimum ts of txn committed via txn but not yet applied
mz_txn_data_shard_count count of data shards registered to the txn set
mz_txn_op_duration_seconds time spent running a txn operation op
mz_txn_op_errored_count count of times a txn operation errored op
mz_txn_op_retry_count count of times a txn operation retried op
mz_txn_op_started_count count of times a txn operation started op
mz_txn_op_succeeded_count count of times a txn operation succeeded op
mz_webhook_get_appender_count Count of getting a webhook appender from the Coordinator.
mz_webhook_validation_reduce_failures Count of how many times we've failed to reduce a webhook source's CHECK statement. reason
outer_join_lowering_cases How many times the different outer join lowering cases happened. case
transform_hits How many times a given transform changed the plan. transform
transform_total How many times a given transform was applied. transform
Back to top ↑