Skip to content

Kubernetes Level 02 Day 10: Troubleshoot Deployment Issues

This document outlines the troubleshooting and resolution process for Kubernetes Level 02 Day 10. An existing Redis deployment was broken after manual changes. By analyzing describe logs, two primary configuration errors were identified and corrected.

Table of Contents

Task Overview

Objective: Resolve the failure of the redis-deployment to ensure the Redis app returns to a Running state.

  • Deployment Name: redis-deployment
  • Current State: 0/1 Available, Pod stuck in ContainerCreating.
  • Root Causes Found:
    1. Incorrect ConfigMap name reference (redis-cofig vs redis-config).
    2. Incorrect container image tag (redis:alpin vs redis:alpine).

Troubleshooting Process

1. Identify the Resource Issues

We start by checking the status of the deployment. 

kubectl get deployments
* Observation: AVAILABLE is 0.

Next, we describe the deployment to see the container specifications. 

kubectl describe deployment redis-deployment
* Findings: * The Image is listed as redis:alpin. * The Volume named config is looking for a ConfigMap named redis-cofig.

2. Analyze Pod Failure

We look at the pod events to see why it won't start. 

kubectl describe pod <redis-pod-name>
* Critical Error Message: Warning FailedMount ... MountVolume.SetUp failed for volume "config" : configmap "redis-cofig" not found

Step-by-Step Solution

To resolve these issues, we need to edit the deployment configuration.

Command: 

kubectl edit deployment redis-deployment

1. Fix ConfigMap Typo

In the editor, scroll down to the volumes: section at the bottom.

Change: 

      volumes:
      - name: config
        configMap:
          name: redis-cofig  # Change this
To: 
      volumes:
      - name: config
        configMap:
          name: redis-config # Fixed typo

2. Fix Image Tag Typo

In the same editor, locate the containers: section.

Change: 

    spec:
      containers:
      - name: redis-container
        image: redis:alpin  # Change this
To: 
    spec:
      containers:
      - name: redis-container
        image: redis:alpine # Fixed typo

Save and exit the editor (:wq).

3. Verification

Verify that the deployment triggers a new rollout and the pod starts successfully. 

kubectl rollout status deployment redis-deployment
kubectl get pods
* Expected Output: Pod status should be Running and Ready 1/1.

Deep Dive: Common Deployment Errors

Missing ConfigMaps/Secrets

If a Pod references a ConfigMap or Secret that does not exist, it will stay in the ContainerCreating or Pending state. Kubernetes will keep trying to find the resource, as seen in the "FailedMount" events.

Image Tag Errors

Docker/Container images are case-sensitive and spelling-sensitive. * redis:alpin (Incorrect) * redis:alpine (Correct) If an image tag is wrong, the pod will usually enter ImagePullBackOff. However, because the ConfigMap was also missing in this task, the pod was stuck at the "Mounting Volumes" stage before it even attempted to pull the image.

Idempotency and kubectl edit

kubectl edit allows for live debugging. When you save your changes, Kubernetes notices the difference between the "desired state" (your new YAML) and the "current state". It then terminates the old, broken pod and starts a new one with your fixes.