Scenario

A memory-intensive application named oom-demo is repeatedly crashing. The pod status shows CrashLoopBackOff, but you need to confirm the underlying cause is an "Out Of Memory" (OOM) error and fix it by increasing the memory limit.

Task

Inspect the existing pod oom-demo in the apps namespace. Confirm the termination reason is OOMKilled. Update the pod definition to increase the memory limit from 20Mi to 100Mi. Verify the pod stabilizes and enters the Running state.

Example

Initial State (Failing):

$ kubectl get pod oom-demo -n apps
NAME       READY   STATUS             RESTARTS
oom-demo   0/1     CrashLoopBackOff   4

Success State (Fixed):

$ kubectl get pod oom-demo -n apps
NAME       READY   STATUS    RESTARTS
oom-demo   1/1     Running   4

Step 1: Diagnose the Crash

Check the previous container instance to confirm it was killed by memory pressure.

kubectl get pod oom-demo -n apps -o jsonpath='{.status.containerStatuses[0].lastState.terminated.reason}'
# Expected Output: OOMKilled

Alternatively, kubectl describe pod oom-demo -n apps will show this in the Events/Status section.

Step 2: Apply the Fix

Since you cannot change the resources of a running Pod directly, you must recreate it. Export the configuration, edit it, and replace the pod.

# 1. Export the current config
kubectl get pod oom-demo -n apps -o yaml > oom-fix.yaml

# 2. Edit the file
nano oom-fix.yaml

Find the resources section. Change the limit from 20Mi to 100Mi.

    resources:
      limits:
        memory: "100Mi"  # <--- Update this value
      requests:
        memory: "10Mi"

Save and exit (Ctrl+O, Enter, Ctrl+X).

# 3. Replace the pod (force deletion to be quick)
kubectl replace -f oom-fix.yaml --force
# Expected Output: 
# pod "oom-demo" deleted
# pod "oom-demo" replaced

Step 3: Verify Stability

Wait a few seconds and check that the pod is now stable.

kubectl get pod oom-demo -n apps
# Expected Output: STATUS: Running