Top 19 Kubernetes Service Issues and How to Fix Them

Common Kubernetes Issues and Resolutions

Problem 1: Service access failure due to unrecognizable certificate

Cause: Custom or expired certificate.

Solution: Update the certificate.

Problem 2: Service access failure caused by port mapping error

curl: (7) Failed connect to 10.103.22.158:3000; Connection refused

Cause: Service is running but port mapping is incorrect.

Solution: Delete the service and remap the port.

kubectl delete svc nginx-deployment

Problem 3: Service exposure failure (AlreadyExists)

Error from server (AlreadyExists): services "nginx-deployment" already exists

Cause: The service has already been created.

Solution: Delete the existing service and recreate it.

kubectl delete svc nginx-deployment

Problem 4: External access to ClusterIP service fails

Cause: Service type is ClusterIP, not exposed externally.

Solution: Change the service type to NodePort.

kubectl edit svc nginx-deployment

Problem 5: Pod status ErrImagePull

readiness-httpget-pod   0/1   ErrImagePull   0   10s

Cause: Image cannot be pulled.

Solution: Replace the image with a valid one.

Problem 6: Init container stuck in Init:0/2

NAME   READY   STATUS   RESTARTS   AGE
myapp-pod   0/1   Init:0/2   0   20s

Cause: Init container did not finish.

Error from server (BadRequest): container "myapp-container" in pod "myapp-pod" is waiting to start: PodInitializing

Solution: Ensure the init container completes successfully.

Problem 7: CrashLoopBackOff due to image issue

Cause: Faulty image causing container restart failures.

Solution: Replace the image with a working version.

Problem 8: Pod creation failure (Pending/ContainerCreating)

readiness-httpget-pod   0/1   Pending   0   0s
readiness-httpget-pod   0/1   ContainerCreating   0   0s
readiness-httpget-pod   0/1   CrashLoopBackOff   1   4s

Cause: Image problems prevent container start.

Solution: Use a correct image.

Problem 9: Pod ready state never reached

readiness-httpget-pod   0/1   Running   0   116s

Cause: Execution command in the pod fails, preventing resource acquisition.

Solution: Enter the container and create the required resources defined in the YAML.

Problem 10: Pod creation fails due to malformed YAML (Chinese characters)

Cause: Invalid characters in the YAML file.

Solution: Correct the myregistrykey entry.

Problem 11: kube-flannel pod stuck at Init:0/1

Cause: Image pull failure on the slave node.

Solution: Restart Docker on the slave node, manually pull the image, and reinstall the plugin on the master.

kubectl create -f kube-flannel.yml; kubectl get nodes

Problem 12: Service status ErrImagePull

kubectl describe pod test-nginx

Cause: Incorrect image name.

Solution: Delete the faulty pod and redeploy with the correct image.

kubectl delete pod test-nginx; kubectl run test-nginx --image=10.0.0.81:5000/nginx:alpine

Problem 13: Unable to exec into a specific container

Cause: Duplicate containers field in the YAML, causing the target container to be missing.

Solution: Remove the redundant containers entry and recreate the pod.

Problem 14: PersistentVolume creation failure

Cause: Duplicate PV name.

Solution: Change the PV name to a unique value.

Problem 15: PVC mount failure

Cause: Access mode mismatch between PVC and available PVs (requires >1Gi and RWO).

Solution: Adjust accessModes in the YAML or the PV definition.

Problem 16: Pod cannot access data on mounted PV

Cause: NFS volume is empty or permissions are incorrect.

Solution: Create files in the NFS volume and set appropriate permissions.

Problem 17: Node status check fails (missing heapster)

Error from server (NotFound): the server could not find the requested resource (get services http:heapster:)

Cause: Heapster service is not installed.

Solution: Install the Prometheus monitoring component.

Problem 18: Pods remain in Pending state

Cause: All nodes are occupied by pods using the same image, leaving no schedulable nodes.

Solution: Delete existing pods before deploying new ones.

Problem 19: Helm install fails due to missing Chart.yaml

# helm install
Error: This command needs 1 argument: chart name
# helm install ./
Error: no Chart.yaml exists in directory "/root/hello-world"

Cause: Incorrect chart file naming.

Solution: Rename chart.yaml to Chart.yaml.