labs

ENTR
This lab has been modernised but not yet verified end to end. Tell us if something breaks.

Pods

45 dkkubernetesView on GitHub
LabHintsSolution

Lab Solution

Your Pod spec will look like this - you can use the sample in solution/lab.yaml:

apiVersion: v1
kind: Pod
metadata:
  name: sleep-lab
spec:
  containers:
    - name: app
      image: courselabs/bad-sleep

Deploy it in the usual way with Kubectl:

kubectl apply -f labs/pods/solution/lab.yaml

Now watch the Pod status:

kubectl get pod sleep-lab --watch

After around 30 seconds, the application in the container ends, so the container exits - then Kubernetes restarts the Pod. You'll see a new line in the watch output, with the restart count increased to 1:

NAME        READY   STATUS    RESTARTS   AGE
sleep-lab   1/1     Running   0          3s
sleep-lab   0/1     Completed   0          33s
sleep-lab   1/1     Running     1          35s

Pods restart by deleting the old conainer and creating a replacement not by restarting the existing container

The new container runs until the app exits again after 30 seconds. Kubernetes restarts the Pod - but if the Pod containers keep exiting, Kubernetes adds an increasing delay before restarting.

The status changes to Completed then Running again, but eventually the Pod enters CrashLoopBackOff status:

NAME        READY   STATUS    RESTARTS   AGE
sleep-lab   1/1     Running   0          3s
sleep-lab   0/1     Completed   0          33s
sleep-lab   1/1     Running     1          35s
sleep-lab   0/1     Completed   1          64s
sleep-lab   0/1     CrashLoopBackOff   1          79s
sleep-lab   1/1     Running            2          80s
sleep-lab   0/1     Completed          2          110s
sleep-lab   0/1     CrashLoopBackOff   2          2m4s
sleep-lab   1/1     Running            3          2m17s

You can delete the Pod by name:

kubectl delete pod sleep-lab

Or by using the delete command with your YAML file:

kubectl delete -f labs/pods/solution/lab.yaml

Back to the exercises.