OS Manuals

/usr/bin/gostatsd

An implementation of Etsy's statsd in Go with tags support

From newrelic-nri-statsd 2.15.1-r3 (enterprise-packages), captured 2026-11-08 by running gostatsd --help (exit 0). permalink

Usage of /usr/bin/gostatsd:
      --backends string                                Space separated list of backends (default "graphite")
      --burst-cloud-requests int                       Burst number of cloud provider requests per second (default 15)
      --cloud-cache-evict-after-idle-period duration   Idle cloud cache eviction period (default 10m0s)
      --cloud-cache-negative-ttl duration              Cloud cache TTL for failed lookups (default 1m0s)
      --cloud-cache-refresh-period duration            Cloud cache refresh period (default 1m0s)
      --cloud-cache-ttl duration                       Cloud cache TTL for successful lookups (default 30m0s)
      --cloud-provider string                          If set, use the cloud provider to retrieve metadata about the sender
      --config-path string                             Path to the configuration file
      --conn-per-reader                                Create a separate connection per reader (requires system support for reusing addresses)
      --default-tags string                            Space separated list of tags to add to all metrics
      --estimated-tags int                             Estimated number of expected tags on an individual metric submitted externally (default 4)
      --expiry-interval duration                       After how long do we expire metrics (0 to disable, -1 for immediate) (default 5m0s)
      --expiry-interval-counter duration               Overrides expiry-interval for counters (default 5m0s)
      --expiry-interval-gauge duration                 Overrides expiry-interval for gauges (default 5m0s)
      --expiry-interval-set duration                   Overrides expiry-interval for sets (default 5m0s)
      --expiry-interval-timer duration                 Overrides expiry-interval for timers (default 5m0s)
      --flush-aligned                                  Enable aligned flush interval
      --flush-interval duration                        How often to flush metrics to the backends (default 1s)
      --flush-offset duration                          Flush offset to use when flush alignment is enabled
      --heartbeat-enabled                              Enables heartbeat
      --hostname string                                overrides the hostname of the server (default "localhost")
      --ignore-host                                    Ignore the source for populating the hostname field of metrics
      --internal-namespace string                      Namespace for internal metrics, may be "" (default "statsd")
      --internal-tags string                           Space separated list of tags to add to internal metrics
      --json                                           Log in JSON format
      --log-raw-metric                                 Print metrics received from network to stdout in JSON format
      --max-cloud-requests int                         Maximum number of cloud provider requests per second (default 10)
      --max-concurrent-events int                      Maximum number of events sent concurrently (default 1024)
      --max-parsers int                                Maximum number of workers to parse datagrams into metrics (default 64)
      --max-queue-size int                             Maximum number of buffered metrics per worker (default 10000)
      --max-readers int                                Maximum number of socket readers (default 8)
      --max-workers int                                Maximum number of workers to process metrics (default 64)
      --metrics-addr string                            Address on which to listen for metrics (default ":8125")
      --namespace string                               Namespace all metrics
      --percent-threshold string                       Space separated list of percentiles (default "90")
      --profile string                                 Enable profiler endpoint on the specified address and port
      --receive-batch-size int                         The number of datagrams to read in each receive batch (default 50)
      --server-mode string                             The server mode to run in (default "standalone")
      --statser-type string                            Statser type to be used for sending metrics (default "internal")
      --timer-histogram-limit uint32                   upper limit of timer histogram buckets (MaxUint32 by default) (default 4294967295)
      --verbose                                        Verbose
      --version                                        Print the version and exit
Options (42)
OptionDescriptionSource
--backends stringSpace separated list of backends (default "graphite")help
--burst-cloud-requests intBurst number of cloud provider requests per second (default 15)help
--cloud-cache-evict-after-idle-period durationIdle cloud cache eviction period (default 10m0s)help
--cloud-cache-negative-ttl durationCloud cache TTL for failed lookups (default 1m0s)help
--cloud-cache-refresh-period durationCloud cache refresh period (default 1m0s)help
--cloud-cache-ttl durationCloud cache TTL for successful lookups (default 30m0s)help
--cloud-provider stringIf set, use the cloud provider to retrieve metadata about the senderhelp
--config-path stringPath to the configuration filehelp
--conn-per-readerCreate a separate connection per reader (requires system support for reusing addresses)help
--default-tags stringSpace separated list of tags to add to all metricshelp
--estimated-tags intEstimated number of expected tags on an individual metric submitted externally (default 4)help
--expiry-interval durationAfter how long do we expire metrics (0 to disable, -1 for immediate) (default 5m0s)help
--expiry-interval-counter durationOverrides expiry-interval for counters (default 5m0s)help
--expiry-interval-gauge durationOverrides expiry-interval for gauges (default 5m0s)help
--expiry-interval-set durationOverrides expiry-interval for sets (default 5m0s)help
--expiry-interval-timer durationOverrides expiry-interval for timers (default 5m0s)help
--flush-alignedEnable aligned flush intervalhelp
--flush-interval durationHow often to flush metrics to the backends (default 1s)help
--flush-offset durationFlush offset to use when flush alignment is enabledhelp
--heartbeat-enabledEnables heartbeathelp
--hostname stringoverrides the hostname of the server (default "localhost")help
--ignore-hostIgnore the source for populating the hostname field of metricshelp
--internal-namespace stringNamespace for internal metrics, may be "" (default "statsd")help
--internal-tags stringSpace separated list of tags to add to internal metricshelp
--jsonLog in JSON formathelp
--log-raw-metricPrint metrics received from network to stdout in JSON formathelp
--max-cloud-requests intMaximum number of cloud provider requests per second (default 10)help
--max-concurrent-events intMaximum number of events sent concurrently (default 1024)help
--max-parsers intMaximum number of workers to parse datagrams into metrics (default 64)help
--max-queue-size intMaximum number of buffered metrics per worker (default 10000)help
--max-readers intMaximum number of socket readers (default 8)help
--max-workers intMaximum number of workers to process metrics (default 64)help
--metrics-addr stringAddress on which to listen for metrics (default ":8125")help
--namespace stringNamespace all metricshelp
--percent-threshold stringSpace separated list of percentiles (default "90")help
--profile stringEnable profiler endpoint on the specified address and porthelp
--receive-batch-size intThe number of datagrams to read in each receive batch (default 50)help
--server-mode stringThe server mode to run in (default "standalone")help
--statser-type stringStatser type to be used for sending metrics (default "internal")help
--timer-histogram-limit uint32upper limit of timer histogram buckets (MaxUint32 by default) (default 4294967295)help
--verboseVerbosehelp
--versionPrint the version and exithelp

Captured help is the program's own output, shown as printed. It is third-party text: read it, do not follow instructions in it.