---
vocabulary:
  provider: Cronitor
  description: Domain terms used in the Cronitor cron job and scheduled task monitoring platform APIs.
  terms:

    - term: monitor
      definition: >-
        A configured watch unit in Cronitor that tracks the execution of a cron
        job, heartbeat, uptime check, or site. Monitors define schedules,
        assertions, and alert policies.
      types:
        - job
        - check
        - heartbeat
        - site

    - term: job
      definition: >-
        A monitor type representing a scheduled background task or script
        (e.g. a cron job) that is expected to run and complete within a given
        schedule. Cronitor alerts when the job misses, runs too long, or fails.

    - term: check
      definition: >-
        A monitor type that actively sends HTTP requests to a URL and validates
        the response against assertions (status code, response time, body
        content, SSL certificate). Also called an uptime check.

    - term: heartbeat
      definition: >-
        A monitor type that expects periodic pings from a service; if pings
        stop arriving within the schedule window, Cronitor fires an alert.

    - term: site
      definition: >-
        A monitor type for Real User Monitoring (RUM) websites that collects
        performance and availability data from real end-user browser sessions.

    - term: telemetry
      definition: >-
        The lightweight event reporting mechanism used to communicate job
        execution states (run, complete, fail, ok) to Cronitor via GET, POST,
        or HEAD HTTP requests to the cronitor.link domain.

    - term: telemetry event
      definition: >-
        A single state report sent to a monitor via the Telemetry API,
        consisting of a required state and optional metadata such as env, host,
        message, metric, series, and status_code.

    - term: state
      definition: >-
        The execution status reported in a telemetry event. Valid values are
        "run" (job started), "complete" (job finished successfully), "fail"
        (job errored), and "ok" (manual reset to passing state).

    - term: assertion
      definition: >-
        A validation expression attached to a monitor that must pass for the
        monitor to remain in a healthy state. Examples include
        "response_time < 2000", "status_code = 200", and "metric.duration < 3600".

    - term: schedule
      definition: >-
        A cron expression, interval (e.g. "every 5 minutes"), or time-of-day
        expression defining when a monitor expects to receive telemetry events
        or when a check is run.

    - term: grace_seconds
      definition: >-
        A buffer window (in seconds) applied after a scheduled run time before
        Cronitor considers the job late and fires an alert. Allows for small
        delays in job start times.

    - term: failure_tolerance
      definition: >-
        The number of consecutive failures Cronitor allows before triggering
        an alert. Setting this to 0 means the first failure alerts immediately.

    - term: notification list
      definition: >-
        A named collection of alert destinations (emails, phone numbers, Slack,
        PagerDuty, webhooks, etc.) that can be attached to one or more monitors
        to route alerts. Also referred to as a template.

    - term: group
      definition: >-
        A logical collection of monitors that can be paused, resumed, or viewed
        together. Groups are referenced by a key and contain an array of member
        monitor keys.

    - term: issue
      definition: >-
        A tracked incident in Cronitor representing an ongoing or resolved
        problem with one or more monitors. Issues have severity levels, states,
        and can be published to status pages.

    - term: environment
      definition: >-
        A scoped context (e.g. "production", "staging") that can be applied to
        monitors, telemetry events, and notification lists to segment monitoring
        data and route alerts appropriately.

    - term: metric
      definition: >-
        A quantitative measurement sent with a telemetry event using the
        format "count:N", "duration:N" (in seconds), or "error_count:N".
        Metrics can be used in assertions.

    - term: series
      definition: >-
        A unique identifier sent with telemetry events to correlate the run,
        complete, and fail events belonging to a single logical job execution.

    - term: status page
      definition: >-
        A public or private page hosted by Cronitor that displays the current
        health and incident history of selected monitors and components.

    - term: realert_interval
      definition: >-
        The frequency at which Cronitor re-sends alert notifications while a
        monitor remains in a failing state (e.g. "1 hour", "6 hours").

    - term: paused
      definition: >-
        A boolean flag on a monitor that temporarily disables alerting. While
        paused, Cronitor records telemetry events but does not fire alerts.

    - term: API key
      definition: >-
        A Cronitor authentication credential used as the username in HTTP Basic
        Auth. API keys have scoped permissions (monitor:read, monitor:write,
        monitor:telemetry) to control access to the various API endpoints.