Metrics t through z

tablet_base_max_compaction_score

• Unit: -
• Description: Highest base compaction score of tablets in this BE.

tablet_cumulative_max_compaction_score

• Unit: -
• Description: Highest cumulative compaction score of tablets in this BE.

tablet_merge_sstable_fallback_cohort_mismatch_total

• Unit: Count
• Description: Cumulative number of tablet merges that selected lazy index rebuild because source SST cohorts differed in count, order, or semantic metadata.

tablet_merge_sstable_fallback_duplicate_physical_file_total

• Unit: Count
• Description: Cumulative number of tablet merges that selected lazy index rebuild because a candidate cohort contained a duplicate physical SST filename.

tablet_merge_sstable_fallback_embedded_delvec_total

• Unit: Count
• Description: Cumulative number of tablet merges that selected lazy index rebuild because a required embedded delete vector could not be resolved after SST projection.

tablet_merge_sstable_fallback_nonuniform_mapping_total

• Unit: Count
• Description: Cumulative number of tablet merges that selected lazy index rebuild because source-to-target RSSID mapping or ownership was not uniform enough for metadata reuse.

tablet_merge_sstable_fallback_projected_domain_total

• Unit: Count
• Description: Cumulative number of tablet merges that selected lazy index rebuild because a projected SST owner, RSSID offset, or watermark fell outside the reusable live or supported domain.

tablet_merge_sstable_fallback_rowset_layout_mismatch_total

• Unit: Count
• Description: Cumulative number of tablet merges that selected lazy index rebuild because source rowset physical layouts differed.

tablet_merge_sstable_fallback_shared_or_mixed_total

• Unit: Count
• Description: Cumulative number of tablet merges that selected lazy index rebuild because shared or mixed source SST ownership prevented safe metadata reuse.

tablet_merge_sstable_fallback_unsupported_sst_form_total

• Unit: Count
• Description: Cumulative number of tablet merges that selected lazy index rebuild because source ranges or SST metadata did not meet the supported reuse form.

tablet_merge_sstable_meta_identical_total

• Unit: Count
• Description: Cumulative number of tablet merges that reused one complete, identical inherited SST cohort in the merged tablet metadata.

tablet_merge_sstable_meta_lazy_rebuild_total

• Unit: Count
• Description: Cumulative number of tablet merges that omitted source SST metadata so the next operation or user that loads the Primary Key index rebuilds it.

tablet_merge_sstable_meta_private_total

• Unit: Count
• Description: Cumulative number of tablet merges that projected and reused a complete cohort of private source SST metadata.

tablet_merge_sstable_omitted_bytes_total

• Unit: Bytes
• Description: Cumulative size of unique source SST files omitted from merged index metadata and recorded as orphan files during lazy rebuild fallback.

tablet_merge_sstable_omitted_file_total

• Unit: Count
• Description: Cumulative number of unique source SST files omitted from merged index metadata and recorded as orphan files during lazy rebuild fallback.

tablet_metadata_mem_bytes

• Unit: Bytes
• Description: Memory used by tablet metadata.

tablet_pre_split_data_tier_file_selection

• Unit: Count
• Type: Cumulative
• Labels: mode — subset (the sample scanned a subset of the files, sized by tablet_pre_split_data_tier_scan_byte_limit), under_limit (the input was within the limit, so every file was scanned), disabled (the limit is 0, so every file was scanned), all_selected (the input exceeded the limit, but taking whole files, at least one per partition and at least tablet_pre_split_data_tier_min_scan_files in all, still selected every file), partition_from_file_data (the input exceeded the limit, but a partition column is read from the file data rather than from the path, or path and literal partition columns are mixed, so every file was scanned: a subset could miss whole partitions), path_not_expressible (INSERT INTO ... SELECT FROM FILES() only: a selected file's path contains a character FILES cannot take as an exact path — ,, *, ?, [, {, \, a : in the path after the host, or leading or trailing whitespace — so the sample scanned every file through the statement's own path).
• Description: Total data-tier samples of Broker Load and INSERT INTO ... SELECT FROM FILES() loads, by how the scanned files were chosen.

tablet_pre_split_data_tier_scanned_bytes_percent

• Unit: -
• Type: Histogram
• Description: Share of the input bytes that each data-tier sample of a Broker Load or INSERT INTO ... SELECT FROM FILES() load scanned, as a percentage rounded down to an integer, so 100 means every file was scanned.

tablet_reshard_merge_candidate_blocked

• Unit: Count
• Description: Cumulative count of tablet-exclusion events during merge planning, not a live count of currently blocked tablets: it increases every planning pass that skips an otherwise-eligible tablet, and never decreases except across an FE restart. Watch its rate of increase, not its raw value. A tablet is skipped either because it still holds merge-blocking shared data files left over from a tablet split, or because it has not yet been proven free of them -- which usually means compaction has not caught up yet, but can also mean the tablet is on a BE that predates this check, or has simply not been observed yet. Clearing the shared files a split leaves behind takes a compaction that rewrites them: on a non-primary-key table ALTER TABLE ... COMPACT forces that on demand. On a primary-key table the same statement forces base compaction only for a tablet that carries outstanding deletes; a delete-free tablet, which is the usual shape right after a split, is unaffected by it and waits for ordinary compaction to reach those rowsets. Only the Leader FE increments this counter.

tablet_schema_mem_bytes

• Unit: Bytes
• Description: Memory used by tablet schema.

tablet_update_max_compaction_score

• Unit: -
• Description: Highest compaction score of tablets in Primary Key tables in the current BE.

threadpool_task_exception_total

• Unit: Count
• Description: Cumulative number of task exceptions caught and swallowed by ThreadPool worker threads across the BE process. Increments only when enable_threadpool_catch_task_exception is true. When that item is false (default), this metric stays unchanged because there is no enclosing catch clause. Use it to alert on swallowed failures while catch mode is enabled; pool name and exception detail remain in the BE ERROR logs.

thrift_connections_total

• Unit: Count
• Description: Total number of thrift connections (including finished connections).

thrift_current_connections (Deprecated)

thrift_opened_clients

• Unit: Count
• Description: Number of currently opened thrift clients.

thrift_server_acceptor_stall_ms

• Unit: ms
• Type: Instantaneous
• Description: Milliseconds since the FE Thrift accept loop last returned a connection. A wedged acceptor makes this climb, but so does an FE with no Thrift traffic, so read it together with the connection arrival rate rather than alerting on it alone. 0 also covers the case where no acceptor is running at all, before the Thrift server starts and after it stops, so a threshold alert on this metric cannot tell a dead Thrift service from a healthy one.

thrift_server_expired_connections_total

• Unit: Count
• Type: Cumulative
• Description: Total number of connections that the FE Thrift server closed without serving because they had waited in the pending queue longer than thrift_server_queue_timeout_ms. That check is disabled by default, so this counter stays at 0 until an operator arms the timeout. A rising value means the FE was working through a backlog of connections whose callers had almost certainly given up already; read it together with thrift_server_queue_wait_ms to see how far behind the queue had fallen.

thrift_server_queue_wait_ms

• Unit: ms
• Type: Instantaneous
• Description: Quantiles of how long connections waited in the FE Thrift server pending queue before a worker thread picked them up. Sampled for every dequeued connection, including the ones then dropped as expired, so the distribution is not truncated at thrift_server_queue_timeout_ms. This is the leading indicator of Thrift saturation: it climbs while the worker pool is still keeping up, well before thrift_server_rejected_connections_total starts to move.

thrift_server_rejected_connections_total

• Unit: Count
• Type: Cumulative
• Description: Total number of connections the FE Thrift server closed immediately because its worker pool was saturated. Every rejection is counted, including those whose log warning was suppressed by rate limiting. A rising value means clients are being turned away; read it with the thread_pool metrics for thrift-server-pool, where the rate of increase indicates how far worker capacity is short.

thrift_used_clients

• Unit: Count
• Description: Number of thrift clients in use currently.

total_column_pool_bytes (Deprecated)

transaction_streaming_load_bytes

• Unit: Bytes
• Description: Total loading bytes of transaction load.

transaction_streaming_load_current_processing

• Unit: Count
• Description: Number of currently running transactional Stream Load tasks.

transaction_streaming_load_duration_ms

• Unit: ms
• Description: Total time spent on Stream Load transaction Interface.

transaction_streaming_load_requests_total

• Unit: Count
• Description: Total number of transaction load requests.

txn_request

• Unit: -
• Description: Transaction requests of BEGIN, COMMIT, ROLLBACK, and EXEC.

uint8_column_pool_bytes

• Unit: Bytes
• Description: Bytes used by the UINT8 column pool.

unused_rowsets_count

• Unit: Count
• Description: Total number of unused rowsets. Please note that these rowsets will be reclaimed later.

update_apply_queue_count

• Unit: Count
• Description: Queued task count in the Primary Key table transaction APPLY thread pool.

update_compaction_duration_us

• Unit: us
• Description: Total time spent on Primary Key table compactions.

update_compaction_outputs_bytes_total

• Unit: Bytes
• Description: Total bytes written by Primary Key table compactions.

update_compaction_outputs_total

• Unit: Count
• Description: Total number of Primary Key table compactions.

update_compaction_task_byte_per_second

• Unit: Bytes/s
• Description: Estimated rate of Primary Key table compactions.

update_compaction_task_cost_time_ns

• Unit: ns
• Description: Total time spent on the Primary Key table compactions.

update_del_vector_bytes_total

• Unit: Bytes
• Description: Total memory used for caching DELETE vectors in Primary Key tables.

update_del_vector_deletes_new

• Unit: Count
• Description: Total number of newly generated DELETE vectors used in Primary Key tables.

update_del_vector_deletes_total (Deprecated)

update_del_vector_dels_num (Deprecated)

update_del_vector_num

• Unit: Count
• Description: Number of the DELETE vector cache items in Primary Key tables.

update_mem_bytes

• Unit: Bytes
• Description: Memory used by Primary Key table APPLY tasks and Primary Key index.

update_primary_index_bytes_total

• Unit: Bytes
• Description: Total memory cost of the Primary Key index.

update_primary_index_num

• Unit: Count
• Description: Number of Primary Key indexes cached in memory.

update_rowset_commit_apply_duration_us

• Unit: us
• Description: Total time spent on Primary Key table APPLY tasks.

update_rowset_commit_apply_total

• Unit: Count
• Description: Total number of COMMIT and APPLY for Primary Key tables.

update_rowset_commit_request_failed

• Unit: Count
• Description: Total number of failed rowset COMMIT requests in Primary Key tables.

update_rowset_commit_request_total

• Unit: Count
• Description: Total number of rowset COMMIT requests in Primary Key tables.

vacuum_failed

• Unit: Count
• Type: Counter
• Description: Shared-data only. Number of incremental (auto) vacuum rounds on the leader FE in which a Vacuum request for a partition could not be sent or returned an error.

vacuum_success

• Unit: Count
• Type: Counter
• Description: Shared-data only. Number of incremental (auto) vacuum rounds on the leader FE in which every Vacuum request sent to the CNs for a partition returned success. Rounds that send no request are not counted.

vector_index_cache_async_load_failure

• Type: Counter
• Unit: Count
• Description: Cumulative number of background vector index cache-load tasks that started but failed during loading or cache publication. Tasks canceled before execution are not included.

vector_index_cache_async_load_inflight

• Type: Gauge
• Unit: Count
• Description: Current number of vector index cache-load tasks running in background workers.

vector_index_cache_async_load_ns

• Type: Counter
• Unit: Nanoseconds
• Description: Cumulative execution time of background vector index cache-load tasks that started, including successful and failed tasks. Queue wait time and rejected tasks are not included.

vector_index_cache_async_load_queued

• Type: Gauge
• Unit: Count
• Description: Current number of vector index cache-load tasks accepted by the background pool but not yet running.

vector_index_cache_async_load_rejected

• Type: Counter
• Unit: Count
• Description: Cumulative number of background vector index cache-load requests rejected before execution, for example because the cache has zero capacity, the pool is stopped, or its queue cannot accept the task.

vector_index_cache_async_load_success

• Type: Counter
• Unit: Count
• Description: Cumulative number of background vector index cache-load tasks that successfully loaded and published an index. Capacity eviction can remove a successfully published entry immediately when the cache cannot retain it.

vector_index_cache_loading_wait_timeout

• Type: Counter
• Unit: Count
• Description: Cumulative number of synchronous cache callers whose wait for an in-progress vector index load reached vector_index_cache_loading_wait_timeout_ms. This metric counts callers rather than unique indexes; the existing loader continues after a timeout.

wait_base_compaction_task_num

• Unit: Count
• Description: Number of base compaction tasks waiting for execution.

wait_cumulative_compaction_task_num

• Unit: Count
• Description: Number of cumulative compaction tasks waiting for execution.

writable_blocks_total (Deprecated)