kubectl-wait

Wait for a deployment to become available

$ kubectl wait --for condition=available deployment/[deployment_name]

$ kubectl wait --for condition=ready pod [[-l|--selector]] [label_key]=[label_value]

$ kubectl wait --for delete pod [pod_name]

$ kubectl wait --for condition=complete job/[job_name] --timeout 120s

kubectl wait [TYPE[/NAME | NAME_PREFIX | /NAME | -l label_selector]] [--for=duration|time|delete|condition=string] [--timeout=duration]

--for=
    Condition to wait for: duration (e.g., 5s, 1m), datetime (e.g., 2023-01-01T00:00:00Z), delete, or condition=string (e.g., Ready, Available). Default: 0

--timeout=
    Max wait time before exit (e.g., 60s, 5m). Default: 30s

TYPE[/NAME | NAME_PREFIX | /NAME | -l label_selector]
    Resource specifier: kind/name, prefix, exact name, labels, file (-f), or stdin

--field-selector=
    Filter by field query (e.g., status.phase=Running). Supports '=', '==', '!='

--grace-period=
    Wait time for deletion finalization before return. Default: -1 (no wait)

kubectl wait is a Kubernetes CLI subcommand that blocks until specified resources satisfy defined conditions, ideal for scripting, automation, and CI/CD pipelines. It polls resources at intervals, checking statuses like pod readiness, deployment availability, or custom conditions.

Target resources by name (e.g., pod/mypod), type with label selectors (e.g., deployments -l app=myapp), filenames, or stdin. Common uses include waiting for pods to be Ready, jobs to complete, or resources to be deleted.

Specify conditions with --for: durations (e.g., 5m), datetimes, delete, or condition=string like condition=Available. Timeouts prevent hangs, defaulting to 30s. Exit codes: 0 on success, 1 on timeout/failure.

This command simplifies orchestration by ensuring prerequisites before proceeding, reducing race conditions in deployments.

Polls resources periodically (not real-time); may miss short-lived states. Timeouts are strict. Does not support all resource conditions—check docs for compatibility. High usage in loops can overload API server.

kubectl wait --for=condition=Ready pod/mypod
kubectl wait --for=condition=Available deployment/myapp --timeout=300s
kubectl wait --for=delete pod/mypod
kubectl wait --for=condition=complete job/myjob -l app=batch

0: Condition met
1: Timeout, error, or condition unmet

Introduced in Kubernetes v1.12 (2018); enhanced in v1.18+ with broader condition support and --for=delete. Evolved alongside kubectl for better automation.

kubectl-get(1), kubectl-describe(1), kubectl-rollout(1), watch(1)