Pod Restarted During Node Maintenance
Why a pod can restart during a planned node upgrade without any change on your side, and how to keep a workload available through it.
If a pod restarted without any change on your side, a planned node upgrade is the likely cause. This article explains why and how to avoid it.
When CTO2B performs node upgrades, nodes are drained gracefully — pods are
evicted and rescheduled onto other nodes. A Component running a single replica
will see a brief restart while that happens.
CTO2B notifies you at least 48 hours before a planned maintenance window. If a
workload cannot tolerate a restart, run at least two replicas and raise a
request about configuring a disruption budget for it.
Still having issues?
Raise a request with the namespace, the Component name, the pod status, and what
you have already tried.
Related articles
- Workload Resources and Health Checks — resource requests, limits and health checks
- Deploy Without Downtime or Killing Running Work — replicas, disruption budgets and graceful shutdown
- Pod Keeps Restarting (CrashLoopBackOff) — the pod starts but then fails
- Pod Killed with OOMKilled — the pod exceeded its memory limit
- Pod Stuck in Pending (FailedScheduling) — the pod never starts at all
- Troubleshooting Common Issues — find your symptom
Updated about 1 month ago