Kubernetes: Pod is Pending forever

Step-by-step Solution

1. 1Describe the pod: `kubectl describe pod <name>` — scroll to the **Events** section at the bottom.
2. 2If the event says `no nodes available that match all of the predicates`, the scheduler can't place the pod. Check resource requests: is your pod asking for more CPU/RAM than any node has free? Reduce `resources.requests` or add nodes.
3. 3If events mention `taints`, your pod lacks the tolerations for the target node. Add `tolerations:` to the pod spec or remove the taint from the node.
4. 4If events mention `PersistentVolumeClaim` not bound, the pod is waiting for a PVC that has no matching PV. Check `kubectl get pvc` and the underlying StorageClass provisioner.
5. 5If nothing helpful in events, check the scheduler itself: `kubectl -n kube-system get pods` — is `kube-scheduler` healthy? On managed clusters (EKS, GKE, AKS) this is rare but possible.
6. 6For quick diagnosis, temporarily reduce the deployment's `resources.requests` to a tiny value and see if the pod schedules — that isolates whether the issue is capacity or something else.