Kubernetes Pod Debugging Cheat Sheet

kubectl get pods -A --field-selector=status.phase=Pending

# Find pods stuck before startup.

kubectl describe pod <pod> -n <namespace>

# Look for FailedScheduling, taints, or PVC issues.

kubectl get pvc -A

# Check if pod scheduling is blocked on volume binding.

kubectl describe pvc <claim> -n <namespace>

# Inspect storage class, events, and binding failures.

kubectl describe node <node> | sed -n '/Taints:/,/Unschedulable:/p'

# Check whether tolerations are missing.

kubectl get pod <pod> -n <namespace> -o jsonpath='{.spec.nodeSelector}{"
"}{.spec.affinity}{"
"}'

# Inspect scheduling constraints from raw YAML.

kubectl get pods -A | grep CrashLoopBackOff

# List pods that are restarting and failing to stay up.

kubectl logs <pod> -n <namespace> -c <container> --previous

# Very useful when a restart wipes current state.

kubectl describe pod <pod> -n <namespace> | sed -n '/Events:/,$p'

# Inspect ErrImagePull and ImagePullBackOff reasons.

kubectl get pod <pod> -n <namespace> -o jsonpath='{.spec.imagePullSecrets[*].name}{"
"}'

# Verify that imagePullSecrets are wired correctly.

kubectl describe secret <secret> -n <namespace>

# Check secret type and metadata for image pulls.

kubectl get pod <pod> -n <namespace> -o jsonpath='{.status.containerStatuses[0].state.waiting.reason}{"
"}'

# Compact way to inspect wait state reasons.

kubectl get pod <pod> -n <namespace> -o jsonpath='{.spec.containers[*].readinessProbe}{"
"}{.spec.containers[*].livenessProbe}{"
"}{.spec.containers[*].startupProbe}{"
"}'

# View readiness, liveness, and startup probe definitions.

kubectl describe pod <pod> -n <namespace> | grep -A5 -i probe

# Describe a pod and inspect event messages for failing probes.

kubectl exec -it <pod> -n <namespace> -c <container> -- <probe-command>

# Validate probe behavior inside the container context.

kubectl port-forward pod/<pod> -n <namespace> 8080:<container-port>

# Verify whether the probe endpoint behaves as expected.

curl -i http://127.0.0.1:8080/healthz

# Use curl locally against the forwarded health endpoint.

kubectl get pod <pod> -n <namespace> -o jsonpath='{.status.initContainerStatuses}{"
"}'

# See waiting, terminated, and exit details for init containers.

kubectl logs <pod> -n <namespace> -c <init-container>

# Inspect the failing setup step before app containers start.

kubectl describe pod <pod> -n <namespace>

# Read events and termination details from the pod description.

kubectl debug <pod> -n <namespace> --copy-to=<pod>-copy --set-image='*=busybox:1.36' -it

# Use kubectl debug to clone a pod with a safer command for inspection.

kubectl top pod <pod> -n <namespace>

# Compare live usage against requests and limits.

kubectl get pod <pod> -n <namespace> -o jsonpath='{.spec.containers[*].resources}{"
"}'

# Inspect CPU and memory resource settings quickly.

kubectl get pod <pod> -n <namespace> -o jsonpath='{.status.containerStatuses[0].lastState.terminated.reason}{"
"}'

# Check whether the last exit reason was OOMKilled.

kubectl get pod <pod> -n <namespace> -o jsonpath='{.status.containerStatuses[0].lastState.terminated.exitCode}{"
"}'

# Inspect the numeric exit code for the container.