rkallos / peep

Small, simple, unsampled metrics

Peep

Telemetry.Metrics reporter for Prometheus and StatsD (including Datadog).

Peep has some important differences from libraries like TelemetryMetricsPrometheus.Core and TelemetryMetricsStatsd:

• Instead of sampling or on-demand aggregation of samples, Peep estimates distributions using histograms.
• Instead of sending one datagram per telemetry event, Peep's StatsD reporting runs periodically, batching all lines into the smallest number of datagram packets possible while still obeying the configured :mtu setting.

To use it, start a reporter with start_link/1, providing a keyword list of options (see Peep.Options for the schema against which options are validated).

import Telemetry.Metrics

Peep.start_link(
  name: MyPeep,
  metrics: [
    counter("http.request.count"),
    sum("http.request.payload_size"),
    last_value("vm.memory.total")
  ]
)

or put it under a supervisor:

import Telemetry.Metrics

children = [
  {Peep, [
    name: MyPeep,
    metrics: [
      counter("http.request.count"),
      sum("http.request.payload_size"),
      last_value("vm.memory.total")
    ]
  ]}
]

Supervisor.start_link(children, ...)

By default, Peep does not emit StatsD data. It can be enabled by passing in configuration with the statsd keyword. Peep's StatsD reporting supports Unix Domain Sockets.

What's Missing

Currently, there's no implementation of 'summary' metrics. Since histograms are relatively inexpensive in Peep, we suggest you use 'distribution' metrics instead.

Installation

Peep package can be installed by adding peep to your list of dependencies in mix.exs:

def deps do
  [
    {:peep, "~> 2.0"}
  ]
end

Documentation can be generated with ExDoc and published on HexDocs. Once published, the docs can be found at https://hexdocs.pm/peep.