Docs

Volcengine Kubernetes Engine

Event monitoring

Configuring and viewing persistent events

Integrated with TLS, VKE provides the event persistence feature, which can be used to collect cluster events through the event-collector add-on and send event alert notifications. This topic describes how to configure and view cluster events.

Prerequisites

A log project and a log topic have been created in TLS. For details, see Log projects and Log topics.

The event-collector add-on has been installed, and the log project and log topic have been configured. For details, see Installing add-ons.

Limit

When collecting events from a VKE cluster, the event-collector add-on does not collect events generated in the following scenarios:

Historical events before add-on startup: Cluster events generated before the event-collector add-on starts will not be collected retroactively.

Events generated during add-on configuration changes: When the following configuration changes are made to the event-collector add-on, events generated during the change process will not be collected:

Modifying the log project or log topic associated with the add-on.

Enabling or disabling pod scheduling based on taints, or the label selection feature, which triggers add-on restart or rescheduling.

Events generated during add-on listening interruptions: If cluster event listening by the event-collector add-on is interrupted, the listening needs to be retriggered. Cluster events generated during the interruption period cannot be collected.

Event query

In the left navigation bar, click Cluster.

In the cluster list, click the name of the target cluster for which you want to view events.

In the left navigation bar of the cluster management page, click Event management.

Click the Event query tab. Enter the search keywords, select the search time range, and then click Search to view cluster events.

tip

For more information about the index content, see Configuring indexes.

Pod events

Configuring index fields

To view pod events, make sure that the following index fields are configured in the log topic. For configuration methods, see Configuring indexes.

Index field	Type	Field description
involvedObject_kind	String	The Kubernetes resource type of the event, such as Pod, Deployment, Job, and ConfigMap.
type	String	The event level, such as Normal and Warning.
source_component	String	The event source, such as kubelet.
reason	String	The event type, such as Started, Pulling, and Created.
source_host	String	The event source host, such as 192.168.1.100.
message	String	The details of the event.
metadata_name	String	The pod name, such as nginx-deployment-5dc***.
metadata_namespace	String	The namespace where the pod is located, such as default and kube-system.

Viewing pod events

Click the Pod event tab. The system automatically generates charts to record pod events based on log indexes. You can use the filter criteria in the charts to view events for the specified pod.

The following table describes the parameters and filter criteria in the chart.

No.	Parameters	Description
1	PodNamspace	Select the namespace where the target pod is located, such as default and kube-system.
		PodName	Select the target pod.
		Log level	Select the event log level, such as Normal and Warning.
2	Query time	Specify the time period for event queries. You can specify absolute time or relative time, and view historical records.
2		Refresh method	Configure the chart refresh method. The chart is manually refreshed by default. You can enable auto refresh and specify the auto refresh interval.

Last updated: 2026.07.24 14:51:36