Logs and Monitoring

This topic describes how to access logs collected by Tanzu Kubernetes Grid.

Accessing Tanzu Kubernetes Grid Logs

Tanzu Kubernetes Grid collects logs for standalone management cluster and workload cluster deployment and operation. To retrieve audit logs at the Kubernetes API or Linux system level, see Audit Logging.

Local Cluster Logs

Local cluster logs capture the standalone management and workload cluster creation, upgrade, and deletion activity in the ~/.config/tanzu/tkg/logs/CLUSTER-NAME file. These logs can be used to troubleshoot the cluster creation activity and other failures. Successfully deleting a cluster automatically deletes its logs file, but if the delete action fails, the logs file remains and can be reviewed by the user or support team.

Cluster API Logs

If local cluster logs do not provide sufficient information to monitor deployments or troubleshoot component failure, you can retrieve logs using kubectl as follows.

Access Standalone Management Cluster Deployment Logs

To monitor and troubleshoot standalone management cluster deployments, review:

• The log file listed in the terminal output under Logs of the command execution can also be found at…

• The log from your cloud provider module for Cluster API. Retrieve the most recent one as follows:

  1. Search your tanzu management-cluster create output for Bootstrapper created. Kubeconfig: and copy the kubeconfig file path listed. The file is in ~/.kube-tkg/tmp/.
  2. Run the following, based on your cloud provider:
     • vSphere: kubectl logs deployment.apps/capv-controller-manager -n capv-system manager --kubeconfig </path/to/kubeconfig>
     • Amazon Web Services (AWS): kubectl logs deployment.apps/capa-controller-manager -n capa-system manager --kubeconfig </path/to/kubeconfig>
     • Azure: kubectl logs deployment.apps/capz-controller-manager -n capz-system manager --kubeconfig </path/to/kubeconfig>

To troubleshoot Supervisor, see Troubleshooting Supervisor Enablement in the vSphere with Tanzu documentation.

Access Workload Cluster Deployments Logs

After running tanzu cluster create, you can monitor the deployment process in the Cluster API logs on the management cluster.

To access these logs, follow the steps below:

1. Set kubeconfig to your management cluster. For example:

   kubectl config use-context my-management-cluster-admin@my-management-cluster
   

2. Run the following:

   • capi logs:

     kubectl logs deployments/capi-controller-manager -n capi-system manager
     

   • IaaS-specific logs:

     • vSphere: kubectl logs deployments/capv-controller-manager -n capv-system manager
     • AWS: kubectl logs deployments/capa-controller-manager -n capa-system manager
     • Azure: kubectl logs deployments/capz-controller-manager -n capz-system manager

Configuring Logging and Monitoring

You can configure logging and monitoring on workload clusters created by Tanzu Kubernetes Grid:

• To configure log forwarding with Fluent Bit, see Install Fluent Bit for Log Forwarding.
• To configure monitoring with Prometheus and Grafana, see Install Prometheus and Grafana for Monitoring.