Fix: Kubernetes Pod Stuck in Pending State

!Symptoms

• Pod stays in Pending state indefinitely
• kubectl describe pod shows 'Unschedulable' or 'Insufficient' events
• No node has enough resources to schedule the pod
• Pod is waiting for a PersistentVolume to be bound

?Root Causes

• Insufficient CPU or memory on available nodes
• Node selector, affinity, or taint/toleration mismatch
• PersistentVolumeClaim cannot be bound (no matching PV or storage class)
• Resource quotas exceeded in the namespace
• Cluster autoscaler not configured or scaling too slowly

#Diagnosis Steps

1. 1Describe the pod: `kubectl describe pod <pod-name>` and check Events
2. 2Check node resources: `kubectl describe nodes | grep -A 5 'Allocated resources'`
3. 3Check PVC status: `kubectl get pvc` — look for Pending PVCs
4. 4Check resource quotas: `kubectl describe resourcequota -n <namespace>`
5. 5Check node taints: `kubectl describe nodes | grep Taints`

>Fix

1. 1Reduce resource requests in the pod spec to fit available nodes
2. 2Add or resize nodes in the cluster to provide more capacity
3. 3Fix node selector/affinity rules to match available nodes
4. 4Add tolerations for node taints that are blocking scheduling
5. 5Create the required PersistentVolume or fix the StorageClass
6. 6Increase namespace resource quotas if they are too restrictive

*Prevention

• Enable Cluster Autoscaler to automatically add nodes when needed
• Set appropriate resource requests (not too high, not too low)
• Use Pod Disruption Budgets to prevent over-packing nodes
• Monitor cluster capacity and set alerts at 80% utilization
• Use LimitRanges to set default resource requests for pods

Related Error Messages