auth.provider API
    Preparing search index...

    Interface ReadinessRouterOptions

    interface ReadinessRouterOptions {
        includeErrorDetail?: boolean;
        logger?: EventLogger;
        path?: string;
        probes: readonly ReadinessProbe[];
        timeoutMs: number;
    }
    Index
    includeErrorDetail?: boolean

    Include each failing probe's error message in the response body. Defaults to false.

    This endpoint is unauthenticated by design — an orchestrator has no credentials — and probe failures carry text straight from the driver: connect ECONNREFUSED 10.0.3.14:6379 names an internal host and port. On an identity provider that is a free map of the backend network for anyone who can reach the pod. The failing dependency's name is what a probe consumer needs; the message goes to the log, where the operator reading it has already authenticated. Turn this on only when the endpoint is reachable solely from inside the deployment.

    logger?: EventLogger

    When wired, an unready result is logged at warn with the failing checks — including each failure's message, which the response body omits.

    Typed as EventLogger rather than Logger: readiness is what an operator reaches for while things are already broken, and it should not be the endpoint that is hardest to wire.

    path?: string

    Route path. Defaults to /readyz.

    probes: readonly ReadinessProbe[]

    The probes to run, in registration order.

    timeoutMs: number

    Per-probe deadline in milliseconds. Required rather than defaulted: the right value depends on the orchestrator's own probe timeout, which only the composition root knows, and a default here would be a second place for that number to live.