State: Terminated
Reason: OOMKilled
Exit Code: 137

Your container used more memory than its limit allows. Kubernetes killed it.

Fix 1: Increase Memory Limit

# Check current limits
kubectl describe pod my-pod | grep -A 5 Limits

# Increase in deployment
resources:
  requests:
    memory: "256Mi"
  limits:
    memory: "512Mi"  # Increase this

Fix 2: Find the Memory Leak

# Check actual memory usage
kubectl top pod my-pod

# Check memory over time
kubectl top pod my-pod --containers

# Get detailed metrics
kubectl exec my-pod -- cat /sys/fs/cgroup/memory/memory.usage_in_bytes

Fix 3: JVM Heap Size (Java Apps)

# ❌ JVM uses more memory than container limit
# ✅ Set JVM heap to ~75% of container limit
env:
  - name: JAVA_OPTS
    value: "-Xmx384m -Xms256m"  # For 512Mi container limit

Fix 4: Node.js Memory Limit

env:
  - name: NODE_OPTIONS
    value: "--max-old-space-size=384"  # For 512Mi container limit

Fix 5: Set Requests Equal to Limits

# ✅ Prevents overcommit — pod gets guaranteed memory
resources:
  requests:
    memory: "512Mi"
  limits:
    memory: "512Mi"  # Same as request = Guaranteed QoS

Fix 6: Horizontal Scaling Instead

# Instead of giving one pod more memory, run more pods
kubectl scale deployment my-app --replicas=3

Debugging

# Check why it was killed
kubectl describe pod my-pod | grep -A 10 "Last State"

# Check events
kubectl get events --field-selector involvedObject.name=my-pod

# Check previous container logs
kubectl logs my-pod --previous

Related resources

• kubectl CrashLoopBackOff fix
• Kubernetes kubectl cheat sheet
• Docker ENOMEM fix