dcos/telemetry-net

Name: telemetry-net

Owner: DC/OS

Description: null

Created: 2016-04-15 20:06:43.0

Updated: 2018-04-22 08:28:29.0

Pushed: 2018-03-30 16:23:59.0

Homepage: null

Size: 2687

Language: Erlang

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

Telemetry-net

Telemetry-net is a library for aggregation of metrics across many systems. It supports counters and histograms. You can also register a function to be called periodically to populate a value. You can also specify aggregation tags on the client to specify that certain axes of tags on metrics should be collapsed.

Configuration

elemetry,
[
  % Aggregators retain metrics over time, otherwise all metrics are dropped after trying to forward them.
  {is_aggregator, false},
  % Periodically send metrics to a destination
  {forward_metrics, true},
  % Don't allow receiving metrics
  {receive_metrics, false},
  % Send metrics to all hosts that are in this DNS A record
  {forwarder_destinations, ["master.mesos"]},
  % Send metrics every 60 seconds
  {interval_seconds, 60},
  % Add 20 seconds of jitter to the interval to avoid thundering herd
  {splay_seconds, 20},
]

Usage

 = #{host => "host-324242", destination => "10.1.2.3:5"},
gregate pass-through (include all tags), collapse on hostname to get
obal metrics across all hostnames for various destinations, and also
gregate on both hostname and destination for global metrics across
l hosts and backends.
ags = [[], [hostname], [hostname, destination]],
metry:counter(connect_successes, Tags, AggTags, 1),
metry:histogram(connect_latency, Tags, AggTags, TimeDelta),