Monitor Kubernetes Events

There are two event types in Kubernetes:

  • Normal events
  • Warning events

See Application Introspection and Debugging for more information on Kubernetes events. Splunk AppDynamics reports a subset of these as Cluster Events. They are comprised of Kubernetes warning events and important informational notices concerning state changes in the cluster. An example of a state change is a pod transitioning from a pending to a running state.

Note: This document contains links to Kubernetes documentation. Splunk AppDynamics makes no representation as to the accuracy of Kubernetes documentation because Kubernetes controls its own documentation.

A Cluster Event of type warning displays a yellow warning icon. When a Pod, Replicaset, or Deployment gets added, updated, or deleted, otherwise known as a state change, the Cluster Agent reports it as an Info type event and displays a blue info icon.

Configure Policy for Kubernetes Events

When you configure the policy for Kubernetes events, the Events page will display the details with the event type as Custom.

You can configure alert policies with the custom Kubernetes events. For more information about configuring policies, see Configure Policies.

You can use Triggers > Custom Events panel in the lower right corner to configure Kubernetes-specific custom events to trigger the policy. See, Configure Policy Triggers.

InAdd Custom Event Filter, you can include the following:

  • Custom Event Type: Specify any of the supported event types.
  • Properties: Filter the events by using a combination of the following properties:
    • clusterName: the name of the cluster

    • namespace: the namespace of the cluster

    • kind: the kind used in the required cluster environment

    • reason: the event reason

    • name: the name of the event

Supported Event Types

  • Kubernetes New Warning
  • Kubernetes Updated Warning
  • Deployment Added
  • Deployment Deleted
  • Replicaset Added
  • Replicaset Deleted
  • Pod Added
  • Pod Deleted
  • Pod Phase Changed

Monitor Cluster Event data

  1. Click the Servers tab.
  2. On the navigation bar, select Events. A panel displays a list of events that have been reported from both servers and clusters. To check for specific events, you can use the Filters option.
  3. Click Add to add a filter and select the desired filters from the available list.
    Note: The filters work only on the Kubernetes events. You can filter the Kubernetes events labeled as Custom events. For supported Event Types, see

    Supported Event Types

    .
  4. Click Apply. The list of events is filtered to show only the selected events.
  5. TheDouble-click the individual Cluster Event. A dialog displays showing a Cluster Event summary. Logs tab displays the log details. See Enable Log Collection for Failing Pods.
  6. Click X to close the dialog.