Overview

Metrics are statistical information exposed by Hadoop daemons, used for monitoring, performance tuning and debug. There are many metrics available by default and they are very useful for troubleshooting. This page shows the details of the available metrics.

Each section describes each context into which metrics are grouped.

The documentation of Metrics 2.0 framework is here.

jvm context

JvmMetrics

Each metrics record contains tags such as ProcessName, SessionID and Hostname as additional information along with metrics.

Name Description
MemNonHeapUsedM Current non-heap memory used in MB
MemNonHeapCommittedM Current non-heap memory committed in MB
MemNonHeapMaxM Max non-heap memory size in MB
MemHeapUsedM Current heap memory used in MB
MemHeapCommittedM Current heap memory committed in MB
MemHeapMaxM Max heap memory size in MB
MemMaxM Max memory size in MB
ThreadsNew Current number of NEW threads
ThreadsRunnable Current number of RUNNABLE threads
ThreadsBlocked Current number of BLOCKED threads
ThreadsWaiting Current number of WAITING threads
ThreadsTimedWaiting Current number of TIMED_WAITING threads
ThreadsTerminated Current number of TERMINATED threads
GcInfo Total GC count and GC time in msec, grouped by the kind of GC.  ex.) GcCountPS Scavenge=6, GCTimeMillisPS Scavenge=40, GCCountPS MarkSweep=0, GCTimeMillisPS MarkSweep=0
GcCount Total GC count
GcTimeMillis Total GC time in msec
LogFatal Total number of FATAL logs
LogError Total number of ERROR logs
LogWarn Total number of WARN logs
LogInfo Total number of INFO logs
GcNumWarnThresholdExceeded Number of times that the GC warn threshold is exceeded
GcNumInfoThresholdExceeded Number of times that the GC info threshold is exceeded
GcTotalExtraSleepTime Total GC extra sleep time in msec

rpc context

rpc

Each metrics record contains tags such as Hostname and port (number to which server is bound) as additional information along with metrics.

Name Description
ReceivedBytes Total number of received bytes
SentBytes Total number of sent bytes
RpcQueueTimeNumOps Total number of RPC calls
RpcQueueTimeAvgTime Average queue time in milliseconds
RpcProcessingTimeNumOps Total number of RPC calls (same to RpcQueueTimeNumOps)
RpcProcessingAvgTime Average Processing time in milliseconds
RpcAuthenticationFailures Total number of authentication failures
RpcAuthenticationSuccesses Total number of authentication successes
RpcAuthorizationFailures Total number of authorization failures
RpcAuthorizationSuccesses Total number of authorization successes
NumOpenConnections Current number of open connections
CallQueueLength Current length of the call queue
numDroppedConnections Total number of dropped connections
rpcQueueTimenumsNumOps Shows total number of RPC calls (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcQueueTimenums50thPercentileLatency Shows the 50th percentile of RPC queue time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcQueueTimenums75thPercentileLatency Shows the 75th percentile of RPC queue time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcQueueTimenums90thPercentileLatency Shows the 90th percentile of RPC queue time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcQueueTimenums95thPercentileLatency Shows the 95th percentile of RPC queue time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcQueueTimenums99thPercentileLatency Shows the 99th percentile of RPC queue time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcProcessingTimenumsNumOps Shows total number of RPC calls (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcProcessingTimenums50thPercentileLatency Shows the 50th percentile of RPC processing time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcProcessingTimenums75thPercentileLatency Shows the 75th percentile of RPC processing time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcProcessingTimenums90thPercentileLatency Shows the 90th percentile of RPC processing time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcProcessingTimenums95thPercentileLatency Shows the 95th percentile of RPC processing time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.
rpcProcessingTimenums99thPercentileLatency Shows the 99th percentile of RPC processing time in milliseconds (num seconds granularity) if rpc.metrics.quantile.enable is set to true. num is specified by rpc.metrics.percentiles.intervals.

RetryCache/NameNodeRetryCache

RetryCache metrics is useful to monitor NameNode fail-over. Each metrics record contains Hostname tag.

Name Description
CacheHit Total number of RetryCache hit
CacheCleared Total number of RetryCache cleared
CacheUpdated Total number of RetryCache updated

rpcdetailed context

Metrics of rpcdetailed context are exposed in unified manner by RPC layer. Two metrics are exposed for each RPC based on its name. Metrics named “(RPC method name)NumOps” indicates total number of method calls, and metrics named “(RPC method name)AvgTime” shows average turn around time for method calls in milliseconds.

rpcdetailed

Each metrics record contains tags such as Hostname and port (number to which server is bound) as additional information along with metrics.

The Metrics about RPCs which is not called are not included in metrics record.

Name Description
methodnameNumOps Total number of the times the method is called
methodnameAvgTime Average turn around time of the method in milliseconds

dfs context

namenode

Each metrics record contains tags such as ProcessName, SessionId, and Hostname as additional information along with metrics.

Name Description
CreateFileOps Total number of files created
FilesCreated Total number of files and directories created by create or mkdir operations
FilesAppended Total number of files appended
GetBlockLocations Total number of getBlockLocations operations
FilesRenamed Total number of rename operations (NOT number of files/dirs renamed)
GetListingOps Total number of directory listing operations
DeleteFileOps Total number of delete operations
FilesDeleted Total number of files and directories deleted by delete or rename operations
FileInfoOps Total number of getFileInfo and getLinkFileInfo operations
AddBlockOps Total number of addBlock operations succeeded
GetAdditionalDatanodeOps Total number of getAdditionalDatanode operations
CreateSymlinkOps Total number of createSymlink operations
GetLinkTargetOps Total number of getLinkTarget operations
FilesInGetListingOps Total number of files and directories listed by directory listing operations
AllowSnapshotOps Total number of allowSnapshot operations
DisallowSnapshotOps Total number of disallowSnapshot operations
CreateSnapshotOps Total number of createSnapshot operations
DeleteSnapshotOps Total number of deleteSnapshot operations
RenameSnapshotOps Total number of renameSnapshot operations
ListSnapshottableDirOps Total number of snapshottableDirectoryStatus operations
SnapshotDiffReportOps Total number of getSnapshotDiffReport operations
TransactionsNumOps Total number of Journal transactions
TransactionsAvgTime Average time of Journal transactions in milliseconds
SyncsNumOps Total number of Journal syncs
SyncsAvgTime Average time of Journal syncs in milliseconds
TransactionsBatchedInSync Total number of Journal transactions batched in sync
BlockReportNumOps Total number of processing block reports from DataNode
BlockReportAvgTime Average time of processing block reports in milliseconds
CacheReportNumOps Total number of processing cache reports from DataNode
CacheReportAvgTime Average time of processing cache reports in milliseconds
SafeModeTime The interval between FSNameSystem starts and the last time safemode leaves in milliseconds.  (sometimes not equal to the time in SafeMode, see HDFS-5156)
FsImageLoadTime Time loading FS Image at startup in milliseconds
FsImageLoadTime Time loading FS Image at startup in milliseconds
GetEditNumOps Total number of edits downloads from SecondaryNameNode
GetEditAvgTime Average edits download time in milliseconds
GetImageNumOps Total number of fsimage downloads from SecondaryNameNode
GetImageAvgTime Average fsimage download time in milliseconds
PutImageNumOps Total number of fsimage uploads to SecondaryNameNode
PutImageAvgTime Average fsimage upload time in milliseconds
NNStarted Deprecated: Use NNStartedTimeInMillis instead
NNStartedTimeInMillis NameNode start time in milliseconds
GenerateEDEKTimeNumOps Total number of generating EDEK
GenerateEDEKTimeAvgTime Average time of generating EDEK in milliseconds
WarmUpEDEKTimeNumOps Total number of warming up EDEK
WarmUpEDEKTimeAvgTime Average time of warming up EDEK in milliseconds

FSNamesystem

Each metrics record contains tags such as HAState and Hostname as additional information along with metrics.

Name Description
MissingBlocks Current number of missing blocks
ExpiredHeartbeats Total number of expired heartbeats
TransactionsSinceLastCheckpoint Total number of transactions since last checkpoint
TransactionsSinceLastLogRoll Total number of transactions since last edit log roll
LastWrittenTransactionId Last transaction ID written to the edit log
LastCheckpointTime Time in milliseconds since epoch of last checkpoint
CapacityTotal Current raw capacity of DataNodes in bytes
CapacityTotalGB Current raw capacity of DataNodes in GB
CapacityUsed Current used capacity across all DataNodes in bytes
CapacityUsedGB Current used capacity across all DataNodes in GB
CapacityRemaining Current remaining capacity in bytes
CapacityRemainingGB Current remaining capacity in GB
CapacityUsedNonDFS Current space used by DataNodes for non DFS purposes in bytes
TotalLoad Current number of connections
SnapshottableDirectories Current number of snapshottable directories
Snapshots Current number of snapshots
NumEncryptionZones Current number of encryption zones
BlocksTotal Current number of allocated blocks in the system
FilesTotal Current number of files and directories
PendingReplicationBlocks Current number of blocks pending to be replicated
UnderReplicatedBlocks Current number of blocks under replicated
CorruptBlocks Current number of blocks with corrupt replicas.
ScheduledReplicationBlocks Current number of blocks scheduled for replications
PendingDeletionBlocks Current number of blocks pending deletion
ExcessBlocks Current number of excess blocks
PostponedMisreplicatedBlocks (HA-only) Current number of blocks postponed to replicate
PendingDataNodeMessageCount (HA-only) Current number of pending block-related messages for later processing in the standby NameNode
MillisSinceLastLoadedEdits (HA-only) Time in milliseconds since the last time standby NameNode load edit log. In active NameNode, set to 0
BlockCapacity Current number of block capacity
StaleDataNodes Current number of DataNodes marked stale due to delayed heartbeat
TotalFiles Deprecated: Use FilesTotal instead
MissingReplOneBlocks Current number of missing blocks with replication factor 1
NumFilesUnderConstruction Current number of files under construction
NumActiveClients Current number of active clients holding lease
HAState (HA-only) Current state of the NameNode: initializing or active or standby or stopping state
FSState Current state of the file system: Safemode or Operational
LockQueueLength Number of threads waiting to acquire FSNameSystem lock
TotalSyncCount Total number of sync operations performed by edit log
TotalSyncTimes Total number of milliseconds spent by various edit logs in sync operation
NameDirSize NameNode name directories size in bytes
NumTimedOutPendingReplications The number of timed out replications. Not the number of unique blocks that timed out. Note: The metric name will be changed to NumTimedOutPendingReconstructions in Hadoop 3 release.
FSN(Read|Write)LockOperationNameNumOps Total number of acquiring lock by operations
FSN(Read|Write)LockOperationNameAvgTime Average time of holding the lock by operations in milliseconds

JournalNode

The server-side metrics for a journal from the JournalNode’s perspective. Each metrics record contains Hostname tag as additional information along with metrics.

Name Description
Syncs60sNumOps Number of sync operations (1 minute granularity)
Syncs60s50thPercentileLatencyMicros The 50th percentile of sync latency in microseconds (1 minute granularity)
Syncs60s75thPercentileLatencyMicros The 75th percentile of sync latency in microseconds (1 minute granularity)
Syncs60s90thPercentileLatencyMicros The 90th percentile of sync latency in microseconds (1 minute granularity)
Syncs60s95thPercentileLatencyMicros The 95th percentile of sync latency in microseconds (1 minute granularity)
Syncs60s99thPercentileLatencyMicros The 99th percentile of sync latency in microseconds (1 minute granularity)
Syncs300sNumOps Number of sync operations (5 minutes granularity)
Syncs300s50thPercentileLatencyMicros The 50th percentile of sync latency in microseconds (5 minutes granularity)
Syncs300s75thPercentileLatencyMicros The 75th percentile of sync latency in microseconds (5 minutes granularity)
Syncs300s90thPercentileLatencyMicros The 90th percentile of sync latency in microseconds (5 minutes granularity)
Syncs300s95thPercentileLatencyMicros The 95th percentile of sync latency in microseconds (5 minutes granularity)
Syncs300s99thPercentileLatencyMicros The 99th percentile of sync latency in microseconds (5 minutes granularity)
Syncs3600sNumOps Number of sync operations (1 hour granularity)
Syncs3600s50thPercentileLatencyMicros The 50th percentile of sync latency in microseconds (1 hour granularity)
Syncs3600s75thPercentileLatencyMicros The 75th percentile of sync latency in microseconds (1 hour granularity)
Syncs3600s90thPercentileLatencyMicros The 90th percentile of sync latency in microseconds (1 hour granularity)
Syncs3600s95thPercentileLatencyMicros The 95th percentile of sync latency in microseconds (1 hour granularity)
Syncs3600s99thPercentileLatencyMicros The 99th percentile of sync latency in microseconds (1 hour granularity)
BatchesWritten Total number of batches written since startup
TxnsWritten Total number of transactions written since startup
BytesWritten Total number of bytes written since startup
BatchesWrittenWhileLagging Total number of batches written where this node was lagging
LastWriterEpoch Current writer’s epoch number
CurrentLagTxns The number of transactions that this JournalNode is lagging
LastWrittenTxId The highest transaction id stored on this JournalNode
LastPromisedEpoch The last epoch number which this node has promised not to accept any lower epoch, or 0 if no promises have been made
LastJournalTimestamp The timestamp of last successfully written transaction

datanode

Each metrics record contains tags such as SessionId and Hostname as additional information along with metrics.

Name Description
BytesWritten Total number of bytes written to DataNode
BytesRead Total number of bytes read from DataNode
BlocksWritten Total number of blocks written to DataNode
BlocksRead Total number of blocks read from DataNode
BlocksReplicated Total number of blocks replicated
BlocksRemoved Total number of blocks removed
BlocksVerified Total number of blocks verified
BlockVerificationFailures Total number of verifications failures
BlocksCached Total number of blocks cached
BlocksUncached Total number of blocks uncached
ReadsFromLocalClient Total number of read operations from local client
ReadsFromRemoteClient Total number of read operations from remote client
WritesFromLocalClient Total number of write operations from local client
WritesFromRemoteClient Total number of write operations from remote client
BlocksGetLocalPathInfo Total number of operations to get local path names of blocks
RamDiskBlocksWrite Total number of blocks written to memory
RamDiskBlocksWriteFallback Total number of blocks written to memory but not satisfied (failed-over to disk)
RamDiskBytesWrite Total number of bytes written to memory
RamDiskBlocksReadHits Total number of times a block in memory was read
RamDiskBlocksEvicted Total number of blocks evicted in memory
RamDiskBlocksEvictedWithoutRead Total number of blocks evicted in memory without ever being read from memory
RamDiskBlocksEvictionWindowMsNumOps Number of blocks evicted in memory
RamDiskBlocksEvictionWindowMsAvgTime Average time of blocks in memory before being evicted in milliseconds
RamDiskBlocksEvictionWindowsnums(50|75|90|95|99)thPercentileLatency The 50/75/90/95/99th percentile of latency between memory write and eviction in milliseconds. Percentile measurement is off by default, by watching no intervals. The intervals are specified by dfs.metrics.percentiles.intervals.
RamDiskBlocksLazyPersisted Total number of blocks written to disk by lazy writer
RamDiskBlocksDeletedBeforeLazyPersisted Total number of blocks deleted by application before being persisted to disk
RamDiskBytesLazyPersisted Total number of bytes written to disk by lazy writer
RamDiskBlocksLazyPersistWindowMsNumOps Number of blocks written to disk by lazy writer
RamDiskBlocksLazyPersistWindowMsAvgTime Average time of blocks written to disk by lazy writer in milliseconds
RamDiskBlocksLazyPersistWindowsnums(50|75|90|95|99)thPercentileLatency The 50/75/90/95/99th percentile of latency between memory write and disk persist in milliseconds. Percentile measurement is off by default, by watching no intervals. The intervals are specified by dfs.metrics.percentiles.intervals.
FsyncCount Total number of fsync
VolumeFailures Total number of volume failures occurred
ReadBlockOpNumOps Total number of read operations
ReadBlockOpAvgTime Average time of read operations in milliseconds
WriteBlockOpNumOps Total number of write operations
WriteBlockOpAvgTime Average time of write operations in milliseconds
BlockChecksumOpNumOps Total number of blockChecksum operations
BlockChecksumOpAvgTime Average time of blockChecksum operations in milliseconds
CopyBlockOpNumOps Total number of block copy operations
CopyBlockOpAvgTime Average time of block copy operations in milliseconds
ReplaceBlockOpNumOps Total number of block replace operations
ReplaceBlockOpAvgTime Average time of block replace operations in milliseconds
HeartbeatsNumOps Total number of heartbeats
HeartbeatsAvgTime Average heartbeat time in milliseconds
HeartbeatsTotalNumOps Total number of heartbeats which is a duplicate of HeartbeatsNumOps
HeartbeatsTotalAvgTime Average total heartbeat time in milliseconds
LifelinesNumOps Total number of lifeline messages
LifelinesAvgTime Average lifeline message processing time in milliseconds
BlockReportsNumOps Total number of block report operations
BlockReportsAvgTime Average time of block report operations in milliseconds
IncrementalBlockReportsNumOps Total number of incremental block report operations
IncrementalBlockReportsAvgTime Average time of incremental block report operations in milliseconds
CacheReportsNumOps Total number of cache report operations
CacheReportsAvgTime Average time of cache report operations in milliseconds
PacketAckRoundTripTimeNanosNumOps Total number of ack round trip
PacketAckRoundTripTimeNanosAvgTime Average time from ack send to receive minus the downstream ack time in nanoseconds
FlushNanosNumOps Total number of flushes
FlushNanosAvgTime Average flush time in nanoseconds
FsyncNanosNumOps Total number of fsync
FsyncNanosAvgTime Average fsync time in nanoseconds
SendDataPacketBlockedOnNetworkNanosNumOps Total number of sending packets
SendDataPacketBlockedOnNetworkNanosAvgTime Average waiting time of sending packets in nanoseconds
SendDataPacketTransferNanosNumOps Total number of sending packets
SendDataPacketTransferNanosAvgTime Average transfer time of sending packets in nanoseconds

yarn context

ClusterMetrics

ClusterMetrics shows the metrics of the YARN cluster from the ResourceManager’s perspective. Each metrics record contains Hostname tag as additional information along with metrics.

Name Description
NumActiveNMs Current number of active NodeManagers
numDecommissioningNMs Current number of NodeManagers being decommissioned
NumDecommissionedNMs Current number of decommissioned NodeManagers
NumShutdownNMs Current number of NodeManagers shut down gracefully. Note that this does not count NodeManagers that are forcefully killed.
NumLostNMs Current number of lost NodeManagers for not sending heartbeats.
NumUnhealthyNMs Current number of unhealthy NodeManagers
NumRebootedNMs Current number of rebooted NodeManagers
AMLaunchDelayNumOps Total number of AMs launched
AMLaunchDelayAvgTime Average time in milliseconds RM spends to launch AM containers after the AM container is allocated
AMRegisterDelayNumOps Total number of AMs registered
AMRegisterDelayAvgTime Average time in milliseconds AM spends to register with RM after the AM container gets launched

QueueMetrics

QueueMetrics shows an application queue from the ResourceManager’s perspective. Each metrics record shows the statistics of each queue, and contains tags such as queue name and Hostname as additional information along with metrics.

In running_num metrics such as running_0, you can set the property yarn.resourcemanager.metrics.runtime.buckets in yarn-site.xml to change the buckets. The default values is 60,300,1440.

Name Description
running_0 Current number of running applications whose elapsed time are less than 60 minutes
running_60 Current number of running applications whose elapsed time are between 60 and 300 minutes
running_300 Current number of running applications whose elapsed time are between 300 and 1440 minutes
running_1440 Current number of running applications elapsed time are more than 1440 minutes
AppsSubmitted Total number of submitted applications
AppsRunning Current number of running applications
AppsPending Current number of applications that have not yet been assigned by any containers
AppsCompleted Total number of completed applications
AppsKilled Total number of killed applications
AppsFailed Total number of failed applications
AllocatedMB Current allocated memory in MB
AllocatedVCores Current allocated CPU in virtual cores
AllocatedContainers Current number of allocated containers
AggregateContainersAllocated Total number of allocated containers
aggregateNodeLocalContainersAllocated Total number of node local containers allocated
aggregateRackLocalContainersAllocated Total number of rack local containers allocated
aggregateOffSwitchContainersAllocated Total number of off switch containers allocated
AggregateContainersReleased Total number of released containers
AvailableMB Current available memory in MB
AvailableVCores Current available CPU in virtual cores
PendingMB Current memory requests in MB that are pending to be fulfilled by the scheduler
PendingVCores Current CPU requests in virtual cores that are pending to be fulfilled by the scheduler
PendingContainers Current number of containers that are pending to be fulfilled by the scheduler
ReservedMB Current reserved memory in MB
ReservedVCores Current reserved CPU in virtual cores
ReservedContainers Current number of reserved containers
ActiveUsers Current number of active users
ActiveApplications Current number of active applications
AppAttemptFirstContainerAllocationDelayNumOps Total number of first container allocated for all attempts
AppAttemptFirstContainerAllocationDelayAvgTime Average time RM spends to allocate the first container for all attempts. For managed AM, the first container is AM container. So, this indicates the time duration to allocate AM container. For unmanaged AM, this is the time duration to allocate the first container asked by unmanaged AM.
FairShareMB (FairScheduler only) Current fair share of memory in MB
FairShareVCores (FairScheduler only) Current fair share of CPU in virtual cores
MinShareMB (FairScheduler only) Minimum share of memory in MB
MinShareVCores (FairScheduler only) Minimum share of CPU in virtual cores
MaxShareMB (FairScheduler only) Maximum share of memory in MB
MaxShareVCores (FairScheduler only) Maximum share of CPU in virtual cores

NodeManagerMetrics

NodeManagerMetrics shows the statistics of the containers in the node. Each metrics record contains Hostname tag as additional information along with metrics.

Name Description
containersLaunched Total number of launched containers
containersCompleted Total number of successfully completed containers
containersFailed Total number of failed containers
containersKilled Total number of killed containers
containersIniting Current number of initializing containers
containersRunning Current number of running containers
allocatedContainers Current number of allocated containers
allocatedGB Current allocated memory in GB
availableGB Current available memory in GB
allocatedVcores Current used vcores
availableVcores Current available vcores
containerLaunchDuration Average time duration in milliseconds NM takes to launch a container
badLocalDirs Current number of bad local directories. Currently, a disk that cannot be read/written/executed by NM process or A disk being full is considered as bad.
badLogDirs Current number of bad log directories. Currently, a disk that cannot be read/written/executed by NM process or A disk being full is considered as bad.
goodLocalDirsDiskUtilizationPerc Current disk utilization percentage across all good local directories
goodLogDirsDiskUtilizationPerc Current disk utilization percentage across all good log directories

ugi context

UgiMetrics

UgiMetrics is related to user and group information. Each metrics record contains Hostname tag as additional information along with metrics.

Name Description
LoginSuccessNumOps Total number of successful kerberos logins
LoginSuccessAvgTime Average time for successful kerberos logins in milliseconds
LoginFailureNumOps Total number of failed kerberos logins
LoginFailureAvgTime Average time for failed kerberos logins in milliseconds
getGroupsNumOps Total number of group resolutions
getGroupsAvgTime Average time for group resolution in milliseconds
getGroupsnumsNumOps Total number of group resolutions (num seconds granularity). num is specified by hadoop.user.group.metrics.percentiles.intervals.
getGroupsnums50thPercentileLatency Shows the 50th percentile of group resolution time in milliseconds (num seconds granularity). num is specified by hadoop.user.group.metrics.percentiles.intervals.
getGroupsnums75thPercentileLatency Shows the 75th percentile of group resolution time in milliseconds (num seconds granularity). num is specified by hadoop.user.group.metrics.percentiles.intervals.
getGroupsnums90thPercentileLatency Shows the 90th percentile of group resolution time in milliseconds (num seconds granularity). num is specified by hadoop.user.group.metrics.percentiles.intervals.
getGroupsnums95thPercentileLatency Shows the 95th percentile of group resolution time in milliseconds (num seconds granularity). num is specified by hadoop.user.group.metrics.percentiles.intervals.
getGroupsnums99thPercentileLatency Shows the 99th percentile of group resolution time in milliseconds (num seconds granularity). num is specified by hadoop.user.group.metrics.percentiles.intervals.

metricssystem context

MetricsSystem

MetricsSystem shows the statistics for metrics snapshots and publishes. Each metrics record contains Hostname tag as additional information along with metrics.

Name Description
NumActiveSources Current number of active metrics sources
NumAllSources Total number of metrics sources
NumActiveSinks Current number of active sinks
NumAllSinks Total number of sinks  (BUT usually less than NumActiveSinks, see HADOOP-9946)
SnapshotNumOps Total number of operations to snapshot statistics from a metrics source
SnapshotAvgTime Average time in milliseconds to snapshot statistics from a metrics source
PublishNumOps Total number of operations to publish statistics to a sink
PublishAvgTime Average time in milliseconds to publish statistics to a sink
DroppedPubAll Total number of dropped publishes
Sink_instanceNumOps Total number of sink operations for the instance
Sink_instanceAvgTime Average time in milliseconds of sink operations for the instance
Sink_instanceDropped Total number of dropped sink operations for the instance
Sink_instanceQsize Current queue length of sink operations  (BUT always set to 0 because nothing to increment this metrics, see HADOOP-9941)

default context

StartupProgress

StartupProgress metrics shows the statistics of NameNode startup. Four metrics are exposed for each startup phase based on its name. The startup *phase*s are LoadingFsImage, LoadingEdits, SavingCheckpoint, and SafeMode. Each metrics record contains Hostname tag as additional information along with metrics.

Name Description
ElapsedTime Total elapsed time in milliseconds
PercentComplete Current rate completed in NameNode startup progress  (The max value is not 100 but 1.0)
phaseCount Total number of steps completed in the phase
phaseElapsedTime Total elapsed time in the phase in milliseconds
phaseTotal Total number of steps in the phase
phasePercentComplete Current rate completed in the phase  (The max value is not 100 but 1.0)