# Test-fire the VOD lifecycle webhook

Delivers a single synthetic status_changed event — the shape every job transition
posts, with test: true beside it and no results — so an operator can confirm the
lifecycle consumer is reachable and authenticated without submitting a job. Its
job_id names no job, and its state is running, the state a job's first
event carries.

Without a url it goes where the VOD settings send every job's notifications: their
lifecycle_webhook, with any ${ENV} placeholder resolved, under their
lifecycle_webhook_secret and lifecycle_webhook_header. With a url it goes there
instead, under the credential_secret and credential_header beside it in the
request — what a form shows before it is saved.

One attempt, no retries. Unreachable is a result, not an error — the response is
200 with delivered: false — so 400 means there is nothing to send to (no url
and no configured lifecycle_webhook), a destination that is not an absolute http(s)
URL with a host, or a credential_header that is not a legal header name.

Endpoint: POST /probes/lifecycle-webhook
Version: 2.0.0
Security: engineBasic

## Request fields (application/json):

  - `url` (string,null)
    An absolute http(s) URL with a host, tested instead of the settings' lifecycle_webhook, under the request's own credential_secret. A ${ENV} placeholder is not resolved here; leave url out to test the configured one.

  - `credential_secret` (string,null)
    With url: the name of the secret the test event authenticates with, resolved and sent under credential_header as a notification would. Ignored without url.

  - `credential_header` (string,null)
    With url: the header the secret's value goes out under; unset means Authorization. Ignored without url.

## Response 200 fields (application/json):

  - `delivered` (boolean, required)
    Whether the destination answered with a 2xx status.

  - `status` (integer,null)
    The HTTP status the destination answered, when it answered.

  - `error` (string,null)
    Why delivery fell short, when it did; null when delivered.

  - `credential` (string,null)
    What the event travelled under: sent for the resolved credential, none when the listener, or the request beside its url, names no secret, unresolved when the named secret is not configured — the event went without one, so a 401 here is the missing secret, not the consumer's token.
    Enum: "sent", "none", "unresolved"

## Response 400 fields (application/problem+json):

  - `type` (string)
    URI identifying the problem type.

  - `title` (string, required)
    Short, human-readable summary of the problem type.

  - `status` (integer, required)
    The HTTP status code, repeated from the response line.

  - `detail` (string)
    Human-readable explanation specific to this occurrence.

  - `instance` (string)
    URI identifying this specific occurrence of the problem.

  - `cause` (string)
    Why a VOD resume was refused, on a resume's 409 only: credentials_required — resume again with the detector and its credentials in the body; config_mismatch — the body asks for a configuration the job cannot take, so correct the request; config_drift — the source file or the config the job ran with has changed since, so the job cannot be resumed and must be submitted afresh.
    Enum: "credentials_required", "config_mismatch", "config_drift"

## Response default fields (application/problem+json):

  - `type` (string)
    URI identifying the problem type.

  - `title` (string, required)
    Short, human-readable summary of the problem type.

  - `status` (integer, required)
    The HTTP status code, repeated from the response line.

  - `detail` (string)
    Human-readable explanation specific to this occurrence.

  - `instance` (string)
    URI identifying this specific occurrence of the problem.

  - `cause` (string)
    Why a VOD resume was refused, on a resume's 409 only: credentials_required — resume again with the detector and its credentials in the body; config_mismatch — the body asks for a configuration the job cannot take, so correct the request; config_drift — the source file or the config the job ran with has changed since, so the job cannot be resumed and must be submitted afresh.
    Enum: "credentials_required", "config_mismatch", "config_drift"