Monitoring Linux Workers and Workloads

This topic describes the components and integrations you can use to capture Kubernetes workload and worker node logs and metrics on VMware Tanzu Kubernetes Grid Integrated Edition (TKGI) Linux clusters.

For information about logging and monitoring Kubernetes clusters, workers and workloads on Windows, see Monitoring Windows Worker Clusters and Workers or Logging Windows Worker Clusters and Workers.

Overview

You can use monitoring components to capture Kubernetes workload and worker node logs and metrics on TKGI-provisioned Linux Kubernetes clusters.

For information about monitoring TKGI and TKGI-provisioned cluster VMs on Linux, see Monitoring TKGI and TKGI-Provisioned Clusters.

To capture Kubernetes workload and worker node logs and metrics:

• Configure Monitoring Components and Integrations

Note: When running on worker nodes, the monitoring components and integrations are visible to both TKGI admins and cluster users, such as developers.

Configure Monitoring Components and Integrations

To monitor Kubernetes worker nodes and workloads in your TKGI deployment:

1. Enable one or more supported components and integrations in the Tanzu Kubernetes Grid Integrated Edition tile > In-Cluster Monitoring. For more information about supported components and integrations, see Monitoring Components and Integrations below.

2. To activate TKGI sink resources, Wavefront, or vRealize Operations integration, follow the instructions in In-Cluster Monitoring for your IaaS:

   • Installing Tanzu Kubernetes Grid Integrated Edition on vSphere
   • Installing Tanzu Kubernetes Grid Integrated Edition on vSphere with NSX-T
   • Installing Tanzu Kubernetes Grid Integrated Edition on GCP
   • Installing Tanzu Kubernetes Grid Integrated Edition on AWS
   • Installing Tanzu Kubernetes Grid Integrated Edition on Azure

3. To configure a monitoring integration:

   • Sink Resources: Sink Resources
   • Wavefront:
     • VMware Tanzu Kubernetes Grid Integrated Edition Integration
     • Creating and Managing Sink Resources
   • VMware vRealize Operations Management Pack for Container Monitoring:
     • cAdvisor
     • VMware vRealize Operations Management Pack for Container Monitoring

Monitoring Components and Integrations

The following components and integrations can be used to monitor Kubernetes worker nodes and workloads in your TKGI deployment:

Name	Type	Link
Sink resources	TKGI component	Sink Resources, below.
Wavefront	External integration	VMware Tanzu Kubernetes Grid Integrated Edition Integration in the Wavefront documentation. Or send metrics to Wavefront with a metric sink.
vRealize Operations	External integration, with cAdvisor	on GitHub

Sink Resources

In TKGI, you can deploy log sinks and metric sinks to monitor your Kubernetes worker nodes and workloads that are running on them.

To deploy a log or a metric sink:

1. Enable sink resources in the Tanzu Kubernetes Grid Integrated Edition tile > In-Cluster Monitoring. You can enable both log and metric sink resources or only one of them.
2. (Optional) Enable Node Exporter on worker nodes by selecting the Enable node exporter on workers check box.
3. Create sink resources. For instructions, see Creating and Managing Sink Resources.

For more information about sink resources, see:

• Conceptual information: Sink Architecture in Tanzu Kubernetes Grid Integrated Edition
• Sink resource types, outputs, and identifying strings: Monitoring Clusters with Log Sinks