Alert Rules

Alert rules are used to create actionable alerts and send an alert to a correct channel when the health or downtime of a particular JVM or service has been spanning for a certain period. Some examples of such rules are:

  • Send a message to [Hipchat room MyCompany Sysadmins] when [any service] has been unhealthy for more than [1 hour].
  • Send a message to [Hipchat room MyCompany Sysadmins] when [any JVM] has been down for more than [5 minutes].
  • Send [an email to admins@mycompany.com] when the [e-shop@example.com JVM] has been unhealthy for more than [5 minutes].
  • Send [an email to bigboss@mycompany.com] when the [e-shop@example.com JVM] has been unhealthy for more than [30 minutes].

Each such alert rule will trigger the creation of an alert. Alerts are sent out to alert channels, which you can configure separately.

To avoid alert fatigue/flapping, Plumbr sends out alerts only once during the 24 hours when the alert is triggered from the same JVM or service.

By default, Plumbr alert rules are set to trigger an alert when:

  • Any JVM has been unhealthy for more than 10 minutes.
  • Any service has been unhealthy for more than 15 minutes.
  • Any JVM has been down for more than 10 minutes.

You can either modify existing rules or add new rules based on your specific needs.