Frequently Asked Questions

The Grace Period is a "silent buffer" time you can add to your schedule to account for normal variance in job execution time. During this period, the monitor stays in the **UP** (green) state, even if the scheduled time has passed.

• Formula: LATE Threshold = Last Ping Time + Schedule + Grace Period
• Example: If your job runs every hour (3600s) and takes up to 5 minutes to finish, you might set a grace period of 10 minutes. Your dashboard will stay green for 1 hour and 10 minutes. If no ping arrives by then, it transitions to **LATE**.

A monitor enters the **LATE** status immediately after the Grace Period expires.

• Pre-Alert Warning: If you have an **Alert Delay** configured, the monitor will stay in the **LATE** state (yellow) as a visual warning before an alert is actually sent.
• Formula: DOWN Alert = LATE Threshold + Alert Delay
• Default Behavior: If your Alert Delay is set to 0 (the default), the monitor will transition from **LATE** to **DOWN** almost instantly, and an alert will be sent immediately.

No. We have built-in noise reduction:

• State-Based Alerting: We only alert when the status changes to "DOWN". We won't keep emailing you while it stays down.
• Deduplication: Our dispatcher has a strict lock mechanism that prevents duplicate alerts for the same event from being sent within a short window.

• Email
• Slack
• Discord
• Pushover
• Generic Webhooks (connect to PagerDuty, OpsGenie, or internal tools)

Add a simple HTTP request at the end of your script. For example:

You can also use our /start and /fail endpoints. We support GET parameters or attaching a POST body for logging metadata.

CronRabbit automatically selects distinctive icons and colors based on your environment's name. This helps you visually group your monitors at a glance.

For example, names containing "prod" get an Emerald (green) Globe, "staging" gets an Amber (orange) Layers icon, and "test" gets a Violet (purple) Flask icon. You can see the full mapping in our Visualization Documentation.

Threshold Alerting allows you to reduce notification noise by only triggering an alert if failures persist. This is perfect for jobs that might intermittently fail due to transient network issues but don't require immediate intervention.

• Sequential: Alert only after X failures in a row. A single success resets the count.
• Frequency: Alert if X failures occur within a rolling time window (e.g., more than 3 fails in 1 hour).
• Immediate: The default behavior. Alert as soon as the first failure is detected.

Traffic Monitors allow you to monitor high-frequency jobs or queue workers by tracking the *rate* of pings rather than specific arrival times. It's ideal for tasks that run hundreds or thousands of times an hour where individual failures are less important than overall throughput.

• Expected Pings: The minimum number of pings you expect to receive.
• Period: The rolling time window (e.g., 1 hour, 1 day) used for the count.
• Alerting: Traffic Monitors use **Immediate Alerting**. If the throughput drops below your threshold, you are notified as soon as the discrepancy is detected.

Timezones are critical for calculating when a cron job is "late." CronRabbit uses a hierarchical approach:

• Environment Level: Set a default timezone for your entire environment (e.g., Production). All monitors will inherit this by default.
• Monitor Level: Override the environment timezone for specific jobs that run in different regions.
• Dashboard: All times displayed in your dashboard automatically adapt to the configured timezone, making it easy to see your schedule in the local context.