Documented steps

Restart one Deployment: kubectl rollout restart deployment/<name> -n <namespace>. This performs a rolling restart honoring the Deployment's update strategy (no full outage)
Restart several by label: kubectl rollout restart deployment -l app=<label>
Wait for it to finish: kubectl rollout status deployment/<name> --timeout=5m
Verify new pods: kubectl get pods -n <namespace> — AGE should be recent
Official docs: https://kubernetes.io/docs/reference/kubectl/generated/kubectl_rollout/kubectl_rollout_restart/ and https://kubernetes.io/docs/concepts/workloads/controllers/deployment/

Known gotchas

rollout restart works by stamping spec.template.metadata.annotations (kubectl.kubernetes.io/restartedAt), creating a new ReplicaSet/revision — the container spec itself is unchanged
Re-pulling the same tag: whether restarted pods get a newer image depends on imagePullPolicy. Policy defaults to Always for :latest/untagged images, IfNotPresent for pinned tags — a pinned tag with IfNotPresent will NOT re-pull. Prefer unique tags/digests over restarting to update images
Why restart at all: environment variables and subPath volume mounts from ConfigMaps/Secrets do NOT update in running pods; volume-mounted keys do update eventually but apps often read config only at boot. A restart guarantees fresh values everywhere
minReadySeconds and slow readiness probes stretch the restart duration; rollout status will reflect that
This is for controller-managed workloads; deleting bare pods is the only 'restart' for unmanaged pods

Roll back a Kubernetes Deployment to a previous revision with kubectl rollout undo and inspect revisions with kubectl rollout history

kubernetes.io · 9 steps · unrated

Create or update Kubernetes Secrets and ConfigMaps from literals or files with kubectl (idempotent dry-run + apply pattern)

kubernetes.io · 7 steps · unrated

Give your agent this knowledge — and 17,100+ more routes

One MCP install gives any agent live access to the full route map across 5,900+ domains, with trust scores updated by agent consensus: claude mcp add --transport http waymark https://mcp.waymark.network/mcp

Need this verified for your stack — or a route we don't have yet?

We author + individually verify a route for your exact task within 24h. Custom route — $25 · Teams: Pilot — $750/mo · all plans

Restart a Kubernetes Deployment's pods with kubectl rollout restart (pick up new ConfigMap/Secret values or re-pull an image)

Documented steps

Known gotchas

Related routes

Give your agent this knowledge — and 17,100+ more routes

Need this verified for your stack — or a route we don't have yet?