Kubernetes error explainer
Paste kubectl describe output or a pod log. Known states are matched first, then explained with the likely cause and the commands to run.
Kubernetes error explainer
Output
The result shows up here. Your input is not written to the log.
Errors
- CrashLoopBackOffThe container starts, exits, and Kubernetes waits longer between each restart.
- OOMKilledThe container used more memory than its limit and the kernel killed it.
- ImagePullBackOffKubernetes failed to pull the image and is backing off before trying again.
- ErrImagePullThe kubelet could not pull the container image.
- CreateContainerConfigErrorKubernetes could not build the container config from the pod spec.
- EvictedThe kubelet evicted the pod to reclaim node resources.
- PendingThe pod has not been scheduled onto a node.
- FailedSchedulingThe scheduler tried and rejected every node.
- RunContainerErrorThe container runtime could not start the container.
- ContainerCreatingThe pod is stuck while the runtime sets the container up.
- ErrorThe container exited with a non-zero status.
- NodeNotReadyThe node stopped reporting healthy, so pods on it are disrupted.
Questions
›Which errors have their own page?
CrashLoopBackOff, OOMKilled, ImagePullBackOff, ErrImagePull, CreateContainerConfigError, Evicted, Pending, FailedScheduling, RunContainerError, ContainerCreating, Error, and NodeNotReady.
›Do you run commands against my cluster?
No. Paste the text. Nothing connects to a Kubernetes API.
›What if the text matches more than one state?
Every match is listed. A pod can be Pending because of FailedScheduling, or OOMKilled and then CrashLoopBackOff. The explanation covers the set.
›Is the pasted output stored?
Not unless you are on the paid plan and tick save history. Logs and the database otherwise keep only a hashed usage record.