Errors in kubernetes and debugging

Resources:

I recommend that you read the following books and articles to improve your skills in debugging and in debugging kubernetes:

• Application Introspection and Debugging.
• Debugging: The 9 Indispensable Rules for Finding Even the Most Elusive Software and Hardware Problems
• Effective Debugging: 66 Specific Ways to Debug Software and Systems

Is the problem in the pod?

The first thing is to check if we are having problems in the pods. The first way to do it is by looking at their status:

kubectl get pods

Look for the status, this will tell you if there is a problem if it is not Running.

Then, look for more information about the pod:

kubectl describe pod [pod-name]

At the end you can see the log of recent events related to your Pod. Based on the event message look for the problem in the Error Messages section.

Also, look at the pod logs with:

kubectl logs pod-name

It is also helpful to connect to the pod and try to see if you can execute commands on it to test, with the kubectl command kubectl exec -it .

To go deeper in this topic please read these articles:

• Kubernetes Troubleshooting
• Debug Running Pods
• How to View Kubernetes Pod Logs With Kubectl

Error Messages

ImagePullBackOff

• The problem coud be that you are not using the correct name of the image:

You can test if the image name is correct with docker in you local machine:

docker pull [repository]/[image]:[tag]

0/1 nodes are available 1 node(s) were unschedulable

This is a problem with the nodes not with the pod. We need to debug the nodes to be available to gather information and understand the problem.

First, let’s the pods status:

kubectl get nodes

Second, let’s check the pods that are not READY:

kubectl describe node [node-name]

Now try to match the output to one of these possibilities:

Possibility 1:

Taint: node.kubernetes.io/unschedulable:NoSchedule

Unschedulable:true

Please refer to this post.

Pending status pod

This example is going to show you how to debug a pending pod