Google Kubernetes Engine

Google Kubernetes Engine (GKE) provides a managed environment for deploying, managing, and scaling your containerized applications using Google infrastructure.

Observe helps you monitor the health and activity of your GKE Clusters with these features:

• A dataset with the details of individual GKE clusters that allows you to investigate the status and attributes over time.

• A dashboard with at-a-glance insights, such as Node CPU utilization and network throughput.

• Integration with GCP compute and disk resource datasets for monitoring cluster node resources.

• Integration with the Observe Kubernetes app that gathers all events, logs, and metrics within a cluster using open source collectors.

Observe provides these tools as part of the Observe GCP Integration. To use them in your workspace, install the GCP app.

The GCP GKE app installs the following Resource dataset:

• GKE Cluster

Note

Observe does not support GKE Autopilot clusters for ingesting data.

The GCP GKE app also installs the following Event Stream datasets:

• GKE Cluster Events

• GKE Logs

Viewing GKE Activity in Observe

The GCP GKE Monitoring Dashboard

To see an overview of your GKE clusters, go to the GCP/GKE Cluster dataset in Observe. The GCP/GKE Monitoring dashboard shows an overview of the activity and health of your GKE clusters.

Figure 1 - The Monitoring dashboard for GKE

Related Resources using GraphLink

Use GraphLink to explore resources related to your GKE clusters. Filter to one or more clusters and GraphLink to these datasets for additional details:

Compute resources associated with GKE nodes include the following:

• Compute Instance Group - Groups of compute instances that make up a GKE node.

• Compute Instance - Individual instances within a node.

• Compute Disk - Disk performance metrics for Kubernetes volumes.

If you also use the Observe Kubernetes app, access Kubernetes details associated with GKE clusters by using GraphLink to these Kubernetes datasets:

• Resource

  • Cluster - Associated Kubernetes clusters.

  • Node Condition - Condition information about Kubernetes nodes.

  • Physical Container - The container is currently running for a particular instance.

  • Secret - Details about secrets, such as passwords and tokens.

  • Pod - Deployable unit of computing consisting of one or more containers.

• Event Stream

  • Event - Kubernetes state changes, errors, and other system messages.

  • Pod Update - Changes to pods, for example, a deployment or an annotation.

  • Container Logs - Logs from containerized applications.

  • Container Spec - Configuration details of deployed containers.

  • API Update - Change notifications for Kubernetes objects.

Setup

The Observe GCP app includes an overview of GKE cluster information. The Observe Kubernetes app provides detailed monitoring information and metrics for GKE clusters.