Skip to content

metrics

The Balerter expose prometheus metrics on serviceAddress if it's defined in the configuration. (see configuration)

Under hood, we use prometheus default library for expose metrics. This library expose some default application metrics:

  • go_gc_duration_seconds
  • go_goroutines
  • go_info
  • go_memstats_alloc_bytes
  • go_memstats_frees_total
  • go_memstats_gc_cpu_fraction
  • go_memstats_heap_alloc_bytes
  • go_threads

...and more

The Balerter add 3 custom metrics, which has the prefix balerter_

  • balerter_info_version
  • balerter_scripts_active
  • balerter_alert_status

balerter_info_version

This metric contains the balerter version as metrics label. The metrics value is always equals to 1.

An example:

balerter_info_version{version="v0.10.0"} 1

balerter_scripts_active

This metric contains the script name as metric label name. The metrics value equals to 1 if the script is active, and 0 if the script is inactive.

An example:

balerter_scripts_active{name="folder.dbg.d1"} 1
balerter_scripts_active{name="folder.dbg.d2"} 1
balerter_scripts_active{name="folder.f2.foo"} 1

balerter_alert_status

This metric contains the current alert statuses. An alert name (id) presents as label name. The metric value present the alert level

  • Success - 1
  • Warning - 2
  • Error - 3

An example:

balerter_alert_status{name="foo"} 3
balerter_alert_status{name="bar"} 1
balerter_alert_status{name="baz"} 2