ocient.lat.paused
(gauge) | Returns 1 if LAT loading is paused |
ocient.lat.bytes_buffered
(gauge) | The number of bytes currently buffered by the loading service |
ocient.lat.complete
(gauge) | Returns 1 if LAT loading is complete |
ocient.lat.workers
(gauge) | The number of workers currently in use by the loading service |
ocient.lat.pipeline.bytes_pushed
(gauge) | The number of bytes pushed by the LAT service to the storage layer |
ocient.lat.pipeline.push_errors
(gauge) | The number of errors from the pipeline |
ocient.lat.pipeline.push_attempts
(gauge) | The number of times the pipeline has attempted to push rows to the storage layer |
ocient.lat.pipeline.rows_pushed
(gauge) | The number of rows pushed to the storage layer since last startup |
ocient.lat.pipeline.record_errors
(gauge) | The number of records which produced an error by the loading service |
ocient.allocation.pending_timeouts
(gauge) | Number of uncompleted storage objects of a given type that will expire and be deallocated in the future |
ocient.blobstore.num_blobs
(gauge) | Total number of tracked blobs within the specified blobstore (per silo)
A blob represents a unit of spill issued by the VM |
ocient.blobstore.pending_dispatches
(gauge) | Number of oustanding callbacks to be processed by the blobstore thread (per silo) |
ocient.blobstore.total_bytes
(gauge) | Tracks accumulative bytes from different operations in the blobstore, along with the number of bytes actively used and bytes available for spill |
ocient.blobstore.verify_hash_operations
(gauge) | Number of verify hash operations {completed, failed} - these operations validate spill requests from the VM as an extra check for correctness |
ocient.blockDeviceContext.bytesRead
(gauge) | Total number of bytes read from the drives |
ocient.cmdcomp.metadata.cache_operations
(gauge) | Number of times table or array column PDFs were found in the cache |
ocient.cmdcomp.metadata.eviction_cycles
(gauge) | Number of times table or array column PDFs were evicted from the cache |
ocient.cmdcomp.metadata.tableToRequestStatsCounter
(gauge) | Number of entries in the structure that maps tables to their statistics |
ocient.cmdcomp.queries
(gauge) | Number of active queries in any state |
ocient.cmdcomp.sql_node_load
(gauge) | The load, as a value between 0.0 and 1.0 for each sql node, used to load balance between sql nodes |
ocient.cmdcomps.server_connections
(gauge) | Number of active connections to the SQL node |
ocient.disk_based_operator_instance.bytes_spilled_compressed
(gauge) | Total bytes written to disk for spill |
ocient.disk_based_operator_instance.bytes_spilled_uncompressed
(gauge) | Total bytes of memory to spill to disk |
ocient.fs.num_file_instances
(gauge) | Number of opened file handles |
ocient.fs.num_loaded_inode_descriptor_blocks
(gauge) | Number of cached inode descriptor blocks - inode descriptor blocks are accessed on files being created and opened |
ocient.fs.num_memory_stalls
(gauge) | Number of requests for IO capable memory that were unable to be fulfilled |
ocient.fs.operations
(gauge) | Tracks metrics on filesystem operations such as total operations of a given type and operations outstanding. |
ocient.fs.volume.fragmentation.avg
(gauge) | Reports the average fragmentation score for the given file size bucket. The fragmentation score is the ratio of the number of block refs for the file against the ideal number of minimum block refs. |
ocient.fs.volume.fragmentation.max
(gauge) | Reports the maximum fragmentation score for the given file size bucket. The fragmentation score is the ratio of the number of block refs for the file against the ideal number of minimum block refs. |
ocient.fs.volume.fragmentation.min
(gauge) | Reports the minimum fragmentation score for the given file size bucket. The fragmentation score is the ratio of the number of block refs for the file against the ideal number of minimum block refs. |
ocient.fs.volume.free_bytes
(gauge) | The number of free unused 4KiB blocks in the drive |
ocient.fs.volume.free_blocks
(gauge) | The number of free bytes in the drive |
ocient.fs.volume.free_inodes
(gauge) | The number of unused inode descriptors in the drive |
ocient.fs.volume.health.status
(gauge) | The result of the startup fsck. If normal, will be 0 |
ocient.fs.volume.inodeOperationsSize
(gauge) | The number of active operations which are manipulating an associated file |
ocient.fs.volume.legacySpaceUsedPct
(gauge) | Total percentage of space used on a drive, which also takes into account any legacy data still present in the local storage service. This is the value that should be used to gauge overall utilization of a drive |
ocient.fs.volume.outstanding_operations
(gauge) | Outstanding metadata mutations from the filesystem |
ocient.fs.volume.space_used_pct
(gauge) | Percentage of space used within the file store volume on a drive. Does not account for legacy data outside of the file store |
ocient.fs.volume.total_blocks
(gauge) | Total number of blocks managed by the file store on a given drive. Equals the number of free blocks plus used blocks |
ocient.fs.volume.total_blocks_used
(gauge) | Total number of 4 KiB blocks in-use within the file store on a given drive |
ocient.fs.volume.total_bytes
(gauge) | Total number of bytes managed by the file store on a given drive. Equals the number of free bytes plus used bytes |
ocient.fs.volume.total_bytes_used
(gauge) | Total number of bytes in-use within the file store on a given drive |
ocient.fs.volume.total_inodes
(gauge) | Total number of inode descriptors that are managed by a file store volume. Equals number of used inode descriptors plus unused inode descriptors |
ocient.fs.volume.total_inodes_used
(gauge) | Total number of inode descriptors that are in-use by a file store volume. |
ocient.fs.volume_health.corruptions
(gauge) | Number of detected corruptions from fsck on startup |
ocient.gsd_buffer.batched_requests.avg
(gauge) | The average number of getSegmentData requests that are sent in a single batch.
Higher values represent more efficient network utilization.
Only applicable when nodes or drives are down and virtual segments are being served |
ocient.healthProtocolInstance.numRunningTasks
(gauge) | The number of distributed tasks that are currently running |
ocient.io.page_scheduler.page_count
(gauge) | Number of pages given out by the page scheduler in a given status |
ocient.jemalloc.stats
(gauge) | Contains stats from jemalloc's internal counters, such as heap bytes allocated |
ocient.loadTocLimit.pendingQueueCount
(gauge) | The number of loadToc actions being rate-limited |
ocient.local_storage_service.available_spare_pct
(gauge) | The percentage of spare capacity remaining on a drive |
ocient.local_storage_service.controller_busy_time
(gauge) | The total time (in minutes) a controller is 'busy' (an operation is outstanding in one of its IO queues) |
ocient.local_storage_service.crc_errors
(gauge) | Number of PCIe interface CRC errors encountered. Preserved across power cycles |
ocient.local_storage_service.data_units_write
(gauge) | The total number of data units written to a drive |
ocient.local_storage_service.device_endurance
(gauge) | Contains a vendor specific estimate of the percentage of life used based on actual usage and the manufacturer's prediction of life. A value of 100 indiciates the estimated endurance as been fully consumed, but the drive may continue to function |
ocient.local_storage_service.device_status
(gauge) | Status of a drive controlled by the local storage service. A status of 10 is active |
ocient.local_storage_service.error_log_entries
(gauge) | Number of entries in the error log over the life of the controller |
ocient.local_storage_service.free_space
(gauge) | The estimated amount of free space on a drive |
ocient.local_storage_service.media_errors
(gauge) | Number of occurances where the controller detected an unrecoverable data integrity error |
ocient.local_storage_service.opal_enabled
(gauge) | OPAL related metrics for a given drive - OPAL allows data encrypted at rest.
A status of 0 indiciates a drive is unlocked and accessible to the database.
For OPAL-compliant drives, enabled and supported should both be 1 |
ocient.local_storage_service.opal_status
(gauge) | The OPAL status of a device |
ocient.local_storage_service.opal_supported
(gauge) | Returns data about whether OPAL is supported on a device |
ocient.local_storage_service.operations
(gauge) | Metrics for NVMe operations for a specific drive. If the number of inflight operations remains elevated and the overflow IO queue is populated, a drive is being saturated |
ocient.local_storage_service.power_cycles
(gauge) | Number of power cycles experienced by the drive |
ocient.local_storage_service.power_on_hours
(gauge) | Number of hours that a controller has been actively powered on |
ocient.local_storage_service.read_commands
(gauge) | Total number of read commands issued |
ocient.local_storage_service.segment_table_entries
(gauge) | Number of tracked partitions in the local storage service - for modern deployments this value should be 1 |
ocient.local_storage_service.space_free_pct
(gauge) | Total free space of a drive, reported as a percentage |
ocient.local_storage_service.temp
(gauge) | The current temperature reading from a device |
ocient.local_storage_service.total_space
(gauge) | Total amount of storage (in bytes) managed by the local storage service |
ocient.local_storage_service.unsafe_shutdowns
(gauge) | Number of reported power losses to the controller |
ocient.local_storage_service.warn_available_spare
(gauge) | Number of reported warnings for available spare falling below the controller's warning threshold |
ocient.local_storage_service.warn_read_only
(gauge) | If 1, the controller has entered read-only mode in response to falling endurance |
ocient.local_storage_service.warn_reliability
(gauge) | If 1, the controller has reported that reliability has fallen to a critical level due to media or internal errors |
ocient.local_storage_service.warn_temp
(gauge) | If 1, the controller's current temperature exceeds a warning threshold |
ocient.local_storage_service.write_block_complete
(gauge) | Total number of logical blocks written |
ocient.local_storage_service.write_block_count
(gauge) | Total number of logical block write requests |
ocient.local_storage_service.write_commands
(gauge) | Total number of write commands issued |
ocient.local_storage_service.write_complete
(gauge) | Total number of write operations completed |
ocient.local_storage_service.write_submit_count
(gauge) | Total number of write opreations started |
ocient.memory.heap
(gauge) | Amount of memory, in bytes currently allocated from the heap |
ocient.memory.huge
(gauge) | Amount of huge page memory currently allocated, in bytes |
ocient.metadata.storage_protocol.task_count
(gauge) | The number of tasks belonging to the Metadata Storage Protocol with the given status or task type |
ocient.metadata.storage_protocol.task_time.avg
(gauge) | The average time taken for a given task in the system, in nanoseconds |
ocient.metadataStorageProtocolInstance.lastDeserializationDuration
(gauge) | Time to deserialize the database metadata, in microseconds |
ocient.metadataStorageProtocolInstance.lastSerializationConfigSize
(gauge) | Size of serialized database metadata, in bytes |
ocient.metadataStorageProtocolInstance.lastSerializationDuration
(gauge) | Time to serialize the database metadata, in microseconds |
ocient.network.tcpChannel.uvBufferPool.bytesInUse
(gauge) | Number of bytes allocated for reading inbound network data |
ocient.db.can_connect
(gauge) | Returns 1 if the database service is reachable and healthy |
ocient.db.gdc_current_count
(gauge) | Number of active entries in the global dictionary compression table |
ocient.db.gdc_max_count
(gauge) | Max allowance entries in the global dictionary compression table |
ocient.db.maxpagetime
(gauge) | The max timestamp of a page |
ocient.db.maxsegtime
(gauge) | The max timestamp of a segment |
ocient.db.minpagetime
(gauge) | The min timestamp of a page |
ocient.db.minsegtime
(gauge) | The min timestamp of a segment |
ocient.db.page_rows
(gauge) | The ratio of current row count to pages |
ocient.db.pagecount
(gauge) | The total number of active pages in the cluster |
ocient.db.query_count
(gauge) | The number of active queries across the entire system |
ocient.db.rows_per_page
(gauge) | The ratio of rows to pages |
ocient.db.rows_per_seg
(gauge) | The ratio of rows to segments |
ocient.db.seg_rows
(gauge) | The number of rows allocated to a segment for a given table |
ocient.db.segcount
(gauge) | The number of segments in use by a database |
ocient.db.segments
(gauge) | Reports the status detail of a segment |
ocient.db.size
(gauge) | The consumed disk space of a database |
ocient.db.tot_rows
(gauge) | The total number of rows in a table |
ocient.operator_summary.num_operators
(gauge) | The number of query operators actively running on a system |
ocient.operator_summary.num_queries
(gauge) | The number queries actively running on a system |
ocient.operator_summary.oldest_age
(gauge) | The largest elapsed time for a query operator actively running on the system |
ocient.partitionProvider.lockSlotsTotal
(gauge) | Total number of cache/lock slots |
ocient.partitionProvider.lockSlotsUsed
(gauge) | Total number of cache/lock slots in use |
ocient.partition_provider.bytes_read
(gauge) | Total number of bytes read from segment partitions |
ocient.partition_provider.cache_bytes
(gauge) | Total number of bytes being allocated to activate cache slots (partition size * number of cache slots in use) |
ocient.partition_provider.cache_operations
(gauge) | Number of partitions requested that were already held in a lock slot |
ocient.partition_provider.cache_raw_bytes
(gauge) | Total number of bytes used by cache slots (less than or equal to cached_bytes as we might not use all the bytes allocated for a slot) |
ocient.partition_provider.cache_size
(gauge) | Total number of bytes for all cache slots active or not |
ocient.protocol.actions
(gauge) | Number of currently active adminstration protocol actions |
ocient.protocol.actions.count
(gauge) | The number of a given action with given status on the given protocol. |
ocient.protocol.actions.time
(gauge) | The amount of time, in milliseconds, taken by a given event on a given protocol. |
ocient.protocol.cached_rows
(gauge) | Number of rows cached by the GlobalDataStorage cache. |
ocient.protocol.cached_tables
(gauge) | Number of tables cached by the GlobalDataStorage cache. |
ocient.protocol.raft.participant_state
(gauge) | Global data Storage raft participant state. SUSPENDED=1, FOLLOWER=2, CANDIDATE=3, LEADER_ESTABLISHING=4, LEADER=5 |
ocient.protocol.raft.snapshot_size
(gauge) | Global Data Storage raft serialized snapshot size |
ocient.protocol.time
(gauge) | The amount of time, in milliseconds, taken by a given event on a given protocol. |
ocient.raftEngine.metadataStorageProtocol.nodeCameOnline.time
(gauge) | Time in MS it took to execute the nodeCameOnline leader method for each metadataNode (reported only if time is in excess of 1 second) |
ocient.resource_manager.pending_hp_memory
(gauge) | Number of HugePage Fragments reserved for use by the resource manager |
ocient.result_cache.queries
(gauge) | The number of queries for which the result set is currently cached. |
ocient.rolehostd.initialization_status
(gauge) | Status code of the phases of the ocient application initialization process |
ocient.segment_activation_limit.pending_queue_count
(gauge) | The number of segment activations being rate-limited |
ocient.segment_store.bytes
(gauge) | Number of bytes in-use for storage objects of a given type |
ocient.segment_store.count
(gauge) | Number of tracked storage objects of a given type |
ocient.segment_store.operations
(gauge) | Number of operations started of a kind for a given storage type |
ocient.segment_transfer.operations.count
(gauge) | Number of requests of a given operation responded to with a given status (success or failure) |
ocient.segment_transfer.operations.duration
(gauge) | Sum of response time minus request time for all requests of a given operation |
ocient.storage_cluster.activating_segments
(gauge) | The number of segments that are currently activating on the node (either all segments or a certain type) |
ocient.storage_cluster.data_stats.avg_row_count
(gauge) | Average number of rows in tracked storage objects of a given type |
ocient.storage_cluster.data_stats.avg_size
(gauge) | Average size (in bytes) of tracked storage objects of a given type |
ocient.storage_cluster.data_stats.num_objects
(gauge) | Total number of tracked storage objects of a given type |
ocient.storage_cluster.data_stats.total_row_count
(gauge) | Total row count of tracked storage objects of a given type |
ocient.storage_cluster.data_stats.total_size
(gauge) | Total size (in bytes) of tracked storage objects of a given type |
ocient.storage_cluster.node_table_segment_size_lookup.size
(gauge) | Total number of bytes tracked in the node table segment size lookup cache |
ocient.storage_cluster.on_put_segment_data.throughput.avg
(gauge) | Average throughput of actions for loading data onto a foundation node in MiB/s |
ocient.storage_cluster.osn_reaps.batched
(gauge) | Number of reap requests that have been given to the OSN reap batcher (a mechanism to reduce the number of RPCs performed) |
ocient.storage_cluster.osn_reaps.duration
(gauge) | Length of time (in nanoseconds) of the last OSN reap leader method that was executed on a given node
Values will seem to freeze is leadership in the cluster is passed to a new node |
ocient.storage_cluster.osns.created
(gauge) | Number of new OSNs created by the storage cluster leader through advancing the legal OSN range |
ocient.storage_cluster.osns.reaped
(gauge) | Number of actual OSN reap leader method requests sent (will be less than number of raw batched requests) |
ocient.storage_cluster.pending_segment_deletions
(gauge) | The number of segments pending deletion, blocked by ongoing, running queries |
ocient.storage_cluster.probes.complete
(gauge) | The number of completed query probes that this node believes has completed |
ocient.storage_cluster.probes.incomplete
(gauge) | The number of incomplete query probes that this node believes are ongoing |
ocient.storage_storage.elapsed_time_for_first_osn_to_activate
(gauge) | The time it takes from when a node starts activating its first OSN to when it completes |
ocient.storage_table_action_cooldown_buffer.queued_actions
(gauge) | The number of storage vtable actions being rate-limited |
ocient.stream_loader.api.push_rows_requests
(gauge) | Total number of pushRows requests received |
ocient.stream_loader.data.page_max_size
(gauge) | Size of largest page from last page set (per-table) |
ocient.stream_loader.data.page_median_count
(gauge) | Size of median page from last page set (per-table) |
ocient.stream_loader.data.page_min_size
(gauge) | Size of smallest page from last page set (per-table) |
ocient.stream_loader.data.tracked_page_bytes
(gauge) | Size of pages awaiting segment generation by this loader (per-bucket) |
ocient.stream_loader.data.tracked_page_count
(gauge) | Count of pages awaiting segment generation by this loader (per-bucket) |
ocient.tkt.pipeline.operations
(gauge) | Number of completed IO requests |
ocient.tkt.segment_service.cache_ops
(gauge) | Number of segment retrieval requests that interact with the segment service cache with a given outcome
(e.g. number of requests that make use of the cache)
The cache is used to speed up the filtering of target segments for query execution |
ocient.tkt.segment_service.cache_size
(gauge) | The number of OSNs whose list of segments are cached in the segment service |
ocient.tkt.segment_service.required_tables
(gauge) | The number of tables tracked within the segment service that have segments which can be served to queries |
ocient.virtual_read_cache.back_writes
(gauge) | Number of blocks evicted from the virtual read heap cache to the disk cache |
ocient.virtual_read_cache.disk.operations
(gauge) | Number of total operations in the disk cache with the given type and outcome |
ocient.virtual_read_cache.heap.operations
(gauge) | Number of total operations in the heap cache with the given type and outcome |
ocient.virtual_read_cache.reads.heap
(gauge) | Number of blocks read from the virtual read cache of the given type (heap or disk) |
ocient.virtual_read_cache.total_blocks.disk
(gauge) | Total number of cache blocks in the virtual segment cache of a given type (heap or disk) |
ocient.vm.active_queries_count
(gauge) | The number of queries currently being processed by this node. |
ocient.vm.datablock_router.network_rate
(gauge) | The current rate of receiving data, in Mbps |
ocient.vm.datablock_router.block_count
(gauge) | The number of non-routed, direct datablocks received from connected peers |
ocient.vm.datablock_router.budget
(gauge) | The number of additional fragments the router is currently allowed to queue |
ocient.vm.datablock_router.byte_count
(gauge) | The number of bytes of the serialized messages received from connected peers |
ocient.vm.fetchingCachedQueriesCount
(gauge) | The number of outstanding requests to the query info cache. |
ocient.vm.huge_block.alloc_count
(gauge) | The number of huge blocks that have been allocated. |
ocient.vm.huge_block.resizes
(gauge) | The number of times a huge block has been shrunk. |
ocient.vm.huge_memory_pool.allocated
(gauge) | The number of bytes of huge memory in silos on this node. |
ocient.vm.long_dispatch_events
(gauge) | The number of times an event has taken more than half a second to process. |
ocient.vm.query_tree_probe.count
(gauge) | The number of times we've run a query tree probe action. |
ocient.vm.queryTreeProbe.timeMs
(gauge) | The amount of time, in milliseconds, we've spent on query tree probes. |
ocient.vm.scheduler.no_work_oom_cycles
(gauge) | The number of cycles (passes through the VM loop) which were used for handling OOM, but did not actually do any work. |
ocient.vm.scheduler.oom_killed_queries
(gauge) | Number of queries which have been killed for OOM since startup. |
ocient.vm.scheduler.opInstRuntimeRatio
(gauge) | Fraction of time being spent in executing operators. |
ocient.vm.stats.pdf_cache_size
(gauge) | Number of bytes of RAM used in the PDF cache |
ocient.vmprotocol.ping.countOver100ms
(gauge) | Number of pings to other nodes which have exceeded 100ms since startup. |
ocient.vmprotocol.ping.current
(gauge) | The elapsed time for the current ping. |
