ringsq / alertdog Goto Github PK

# A list of alertmanager endpoints (used when pushing alerts to alertmanager)
alertmanager_endpoints:
  - http://alertmanager-0:9093
  - http://alertmanager-1:9093


# How often Alertdog checks if a Watchdog has been recieved from an expected
# prometheus (optional) (defaults to 2m)
check_interval: 2m

# How long Alertdog will wait before raising a PagerDuty alert if no webhook
# requests were recieved from alertmanger (optional) (defaults to 5m)
expiry: 5m

# The port that the webhook endpoint is exposed on (optional) (defaults to 9767)
port: 9767

# A PagerDuty EventsV2 API routing key
pager_duty_key: PAGER_DUTY_KEY

# A url for a runbook, to be included in PagerDuty alerts (optional)
pagerduty_runbook_url: https://example.org/alertmanager_down_runbook

# A list of prometheus clusters that we expect to recieve Watchdog alerts from
expected:
  -
    # The labels that match this prometheus cluster, should be set to the
    # Alertname you use for watchdog alerts, and any external labels set on
    # this cluster. Note you need to make sure that your cluster have unique labels.
    match_labels:
      alertname: Watchdog
      owner: team-a

    # How long to wait after reciving a watchdog before raising an alert (optional) (defaults to 4m)
    # Note this value should be longer than `check_interval`.
    # Make sure that alertmanager repeats the alert at least this often,
    # by setting `repeat_interval` & `group_interval` to appropriate values in
    # the alertmanger route configuration see example/alertmanger.yml for an example of this.
    expiry: 4m

    # The configuration of the alert that will be raised in alertmanager if the
    # Watchdog isn't recieved within the configured expiry time
    alert:
      name: PrometheusAlertFailure
      labels:
        severity: critical
        owner: team-a
        service: monitoring
        component: prometheus
      annotations:
        runbook: https://example.org/prometheus-alert-failure
        description: |
          This alert fires when the "Watchdog" alert fails to be received
          from a prometheus instance.
          It could indicate that the prometheus instance is not running, or
          that there is a configuration issue with prometheus or alertmanager.