Collect logs

Collect logs to identify root cause of an issue.

For troubleshooting vSphere Cloud Provider issues, you must gather the following log files:

Before you begin, ensure that kubectl CLI is set up. For more information, see Accessing your cluster from the Kubernetes CLI (kubectl).

Controller manager log

1. Get the Kubernetes master pod name.

   kubectl -n kube-system get pod | grep k8s-master
   

   Following is a sample output:

   k8s-master-<nodename>                                        4/4       Running     3          1d
   

2. Get the controller manager log.

   kubectl -n kube-system logs k8s-master-<nodename> -c controller-manager > controller-manager-<nodename>.log
   

API server log

1. Get the Kubernetes master pod name.

   kubectl -n kube-system get pod | grep k8s-master
   

   Following is a sample output:

   k8s-master-<nodename>                                        4/4       Running     3          1d
   

2. Get the API server log.

   kubectl -n kube-system logs k8s-master-<nodename> -c apiserver > apiserver-<nodename>.log
   

Kubelet log

On each node, run the following command to get the kubelet log from that node.

journalctl -u kubelet > kubelet-<nodename>.log