Certified Kubernetes Administrator (CKA) Practice Exam: Part 1

Count the Number of Nodes That Are Ready to Run Normal Workloads

Determine how many nodes in the cluster are ready to run normal workloads (i.e., workloads that do not have any special tolerations).

Output this number to the file /k8s/0001/count.txt

Linux

kubectl config use-context acgk8s
kubectl get nodes

I just grepped for taints first to see what was currently set to see what is ready to run normal workloads. Only control was not ready.

Linux

cloud_user@acgk8s-control:~$ kubectl describe node acgk8s-worker1 | grep -i taints
cloud_user@acgk8s-control:~$ kubectl describe node acgk8s-worker2 | grep -i taints
cloud_user@acgk8s-control:~$ kubectl describe node acgk8s-control | grep -i taints

Linux

kubectl get nodes -o='custom-columns=NodeName:.metadata.name,TaintKey:.spec.taints[*].key' | grep "<none>" | wc -l > /k8s/0001/count.txt
cat /k8s/0001/count.txt

Retrieve Error Messages from a Container Log

In the backend namespace, check the log for the proc container in the data-handler Pod.

Save the lines which contain the text ERROR to the file /k8s/0002/errors.txt.

Linux

kubectl config set-context --current --namespace=backend
kubectl get pods

Linux

kubectl logs -n backend data-handler -c proc | grep "ERROR" > /k8s/0002/errors.txt

Find the Pod with a Label of app=auth in the Web Namespace That Is Utilizing the Most CPU

Before doing this step, please wait a minute or two to give our backend script time to generate CPU load. Determine which Pod in the web namespace with the label app=auth is using the most CPU. Save the name of this Pod to the file /k8s/0003/cpu-pod.txt.

Linux

kubectl config set-context --current --namespace=web
kubectl get pods

Linux

kubectl top pods -n web --sort-by cpu --selector=app=auth 

Linux

kubectl top pods -n web --sort-by cpu --selector=app=auth | grep auth-web | awk '{print $1}' > /k8s/0003/cpu-pod.txt