Skip to main content

Deployment Troubleshooting: When Room Management Has Problems

Deployments fail. Room management has problems. That's reality. But we can fix them. That's troubleshooting.

🎯 The Big Picture​

Think of deployment troubleshooting like fixing hotel room management problems. Rooms not created (pods not created). Update failed (renovation failed). Rooms crashing (pods crashing). That's deployment troubleshooting.

Deployment troubleshooting is systematic. Check deployment. Check ReplicaSet. Check pods. Find the problem. Fix it.

The Hotel Room Management Problem Analogy​

Think of deployment problems like room management problems:

Pods not created: Rooms not created Update stuck: Renovation stuck Pods crashing: Rooms have problems ReplicaSet issues: Floor manager problems Resource issues: Not enough capacity

Once you see it this way, troubleshooting makes perfect sense.

Systematic Troubleshooting Approach​

Step-by-step process:

Step 1: Check Deployment Status

kubectl get deployment hotel-management

Step 2: Describe Deployment

kubectl describe deployment hotel-management

Step 3: Check ReplicaSet

kubectl get replicaset -l app=hotel-app
kubectl describe replicaset <replicaset-name>

Step 4: Check Pods

kubectl get pods -l app=hotel-app
kubectl describe pod <pod-name>

Step 5: Check Logs

kubectl logs <pod-name>
kubectl logs <pod-name> --previous

Step 6: Check Events

kubectl get events --sort-by=.metadata.creationTimestamp

Think of it as: Systematic inspection. Check everything. Find problem.

Common Deployment Problems​

Problem 1: Deployment Not Creating Pods​

Symptoms:

  • Deployment created
  • No pods running
  • Desired > Available

Troubleshooting:

# Check deployment
kubectl describe deployment hotel-management

# Check ReplicaSet
kubectl get replicaset -l app=hotel-app

# Check events
kubectl get events

Common causes:

  • Selector mismatch
  • Resource constraints
  • Node taints
  • Image pull issues

Solution:

  • Fix selector
  • Add resources
  • Remove taints
  • Fix image

Think of it as: Room management created. But no rooms. Check why.

Problem 2: Update Stuck​

Symptoms:

  • Update in progress
  • Not completing
  • Pods not updating

Troubleshooting:

# Check rollout status
kubectl rollout status deployment/hotel-management

# Check deployment
kubectl describe deployment hotel-management

# Check pods
kubectl get pods -l app=hotel-app

Common causes:

  • Health checks failing
  • Resource constraints
  • Image pull issues
  • Application errors

Solution:

  • Fix health checks
  • Add resources
  • Fix image
  • Fix application

Think of it as: Renovation stuck. Not completing. Check why.

Problem 3: Pods Crashing​

Symptoms:

  • Pods created
  • Immediately crash
  • CrashLoopBackOff

Troubleshooting:

# Check pods
kubectl get pods -l app=hotel-app

# Check logs
kubectl logs <pod-name>
kubectl logs <pod-name> --previous

# Describe pod
kubectl describe pod <pod-name>

Common causes:

  • Application errors
  • Configuration errors
  • Missing dependencies
  • Resource limits

Solution:

  • Fix application
  • Fix configuration
  • Add dependencies
  • Adjust limits

Think of it as: Rooms created. But guests can't stay. Check why.

Problem 4: ReplicaSet Issues​

Symptoms:

  • ReplicaSet not creating pods
  • Pods not matching selector
  • ReplicaSet scaling issues

Troubleshooting:

# Check ReplicaSet
kubectl describe replicaset <replicaset-name>

# Check selector
kubectl get replicaset <replicaset-name> -o yaml | grep selector

# Check pods
kubectl get pods --show-labels

Common causes:

  • Selector mismatch
  • Template issues
  • Resource constraints

Solution:

  • Fix selector
  • Fix template
  • Add resources

Think of it as: Floor manager problems. Can't manage rooms. Check why.

Real-World Example: Complete Troubleshooting​

Scenario: Deployment update stuck

Step 1: Check deployment:

kubectl get deployment hotel-management

Output:

NAME              READY   UP-TO-DATE   AVAILABLE   AGE
hotel-management  3/5     2            3           10m

Problem: Update stuck. Only 2 of 5 updated.

Step 2: Check rollout status:

kubectl rollout status deployment/hotel-management

Output:

Waiting for rollout to finish: 2 of 5 new replicas have been updated...

Step 3: Check new pods:

kubectl get pods -l app=hotel-app

Output:

NAME                              READY   STATUS    RESTARTS   AGE
hotel-management-xxxxx (old)      1/1     Running   0          10m
hotel-management-yyyyy (old)      1/1     Running   0          10m
hotel-management-zzzzz (old)      1/1     Running   0          10m
hotel-management-aaaaa (new)      0/1     CrashLoopBackOff  3    2m
hotel-management-bbbbb (new)      0/1     CrashLoopBackOff  3    2m

Problem: New pods crashing.

Step 4: Check logs:

kubectl logs hotel-management-aaaaa

Output:

Error: Cannot connect to database

Problem: Database connection issue.

Step 5: Check configuration:

kubectl describe pod hotel-management-aaaaa | grep -A 10 Env

Found: Missing DATABASE_URL environment variable.

Step 6: Fix:

# Update deployment
env:
- name: DATABASE_URL
  valueFrom:
    secretKeyRef:
      name: db-secret
      key: url

Step 7: Apply fix:

kubectl apply -f deployment.yaml

Step 8: Verify:

kubectl rollout status deployment/hotel-management
kubectl get pods -l app=hotel-app
# All pods running

That's troubleshooting. Systematic. Effective.

Troubleshooting Tools​

kubectl describe deployment​

Best friend for troubleshooting:

kubectl describe deployment hotel-management

Shows:

  • Deployment status
  • ReplicaSet status
  • Pod status
  • Events
  • Conditions

Think of it as: Complete management inspection. Everything visible.

kubectl rollout status​

Monitor rollout:

kubectl rollout status deployment/hotel-management

Shows:

  • Rollout progress
  • Current state
  • Completion status

Think of it as: Renovation progress. See status.

kubectl rollout history​

View history:

kubectl rollout history deployment/hotel-management

Shows:

  • Revision history
  • Change causes
  • Rollback options

Think of it as: Renovation history. Previous designs.

My Take: Troubleshooting Strategy​

Here's what I do:

When deployment fails:

  1. Check deployment status
  2. Describe deployment
  3. Check ReplicaSet
  4. Check pods
  5. Check logs
  6. Check events
  7. Fix and verify

The key: Systematic approach. Check everything. Find root cause. Fix it.

Memory Tip: The Hotel Room Management Problem Analogy​

Deployment troubleshooting = Fixing room management problems

Pods not created: Rooms not created Update stuck: Renovation stuck Pods crashing: Rooms have problems ReplicaSet issues: Floor manager problems

Once you see it this way, troubleshooting makes perfect sense.

Common Mistakes​

  1. Not checking deployment: Missing deployment-level issues
  2. Not checking ReplicaSet: Missing ReplicaSet issues
  3. Not checking pods: Missing pod-level issues
  4. Not checking logs: Missing application errors
  5. Not systematic: Random troubleshooting

Key Takeaways​

  1. Troubleshooting is systematic - Follow a process
  2. Check deployment first - Know the state
  3. Check ReplicaSet - Foundation level
  4. Check pods - Execution level
  5. Check logs - Application level
  6. Most problems are fixable - Don't give up

What's Next?​

Congratulations! You've completed the Deployments & ReplicaSets module. Now let's learn about services. Next: Understanding Services.

Remember: Deployment troubleshooting is like fixing room management problems. Systematic. Check deployment. Check ReplicaSet. Check pods. Find problem. Fix it. Most problems are fixable.