[TBD]Kubernetes events for troubleshooting

Introduction

Laymen explanation

If you have worked on Kubernetes, you might have seen that a POD might not be in ready state for long time. It could be due to delay in fetching image. You might need to know about this and then you might came to know about 'kubectl describe pod' which provides this info. This info is mentioned in the section called event. There are other events which can also help an SRE to troubleshoot the system

Technical explanation

Kubernetes events are a resource type in Kubernetes that are automatically created when other resources have state changes, errors, or other messages that should be broadcast to the system.

Useful points

• Events are specific to object and so, you should look at events in right object. For example, image pull issue will be present in POD related event.

• All events can be viewed at a single place using kubectl get events. This could be the single place if you are not aware of right object to look for.

• Events do not propagate in a way that you might expect. For instance, if your container is unable to start because the docker image cannot be pulled (as seen in the above example) the events are only visible on the pod, not on any ReplicaSet, DaemonSet, StatefulSet, Job, Deployment, or any other resource that created the pod.

• As expected, filtering of events is possible (Ref: below screenshot)

Reference

https://www.bluematador.com/blog/kubernetes-events-explained