Hanzo
OpenapiPlatform

Reports whether this control plane can actually deploy anything.

Reports whether this control plane can actually deploy anything. A real probe, not a status page.

GET /v1/platform/health

Addresshttps://api.hanzo.ai/v1/platform/health
MethodGET
Operationget_platform_health
AuthAuthorization: Bearer $HANZO_API_KEY

Reports whether this control plane can actually deploy anything.

A real probe, not a status page. It answers 200 only when the metadata store is open AND the cluster is genuinely reachable — proved by LISTING the operator App CRD, which settles reachability and CRD presence in one bounded call, and which is the exact question every deploy depends on. Anything else is 503 carrying the real reason and whether the CRD was found.

A constructed cluster client proves nothing — it is built from a kubeconfig, not from a reachable apiserver — so this deliberately spends a round trip rather than reporting ok while every deploy fails. Not admin-gated: liveness has to be probe-able without a credential.

Request

GET /v1/platform/health takes no parameters and no body — the credential is the whole request.

Response

StatusBodyMeaning
200readinessok
503readinessservice unavailable

200 body — 5 fields.

FieldInTypeAlwaysDescription
crdbodybooleanCRD is whether the operator App CRD was found, and is absent when no cluster client resolved and the question could not be asked.
errorbodystringError is the real reason the plane is degraded; absent when it is not.
k8sbodybooleanK8s is whether a cluster client resolved at all.
servicebodystringService is always "platform" — which control plane answered.
statusbodystringStatus is "ok" when this plane can deploy, "degraded" when it cannot.

Failure carries the platform error shape — see Errors.

Examples

hanzo platform health

Platform API · All Hanzo APIs · Interactive reference

How is this guide?

On this page