Deploy Cron Sentinel: Healthchecks [Oct '26]

Deploy and Host Cron Sentinel on Railway

Cron Sentinel is a self-hosted monitor for cron jobs and backups. A job pings a URL when it finishes. If the ping is late or fails, you get an alert.

It runs the Healthchecks image, pinned to v4.4, with PostgreSQL. This template is not affiliated with the Healthchecks project.

About Hosting Cron Sentinel

• Healthchecks listens on the public Railway domain. The database has no public port.
• PostgreSQL 18 keeps check history on a volume. PGDATA is a subdirectory of that volume, so the first boot can initialize the database.
• Signup is closed. The first boot creates one admin. The email is admin@cron-sentinel.local unless you change ADMIN_EMAIL. The password is the generated ADMIN_PASSWORD on the Healthchecks service.
• SECRET_KEY and ADMIN_PASSWORD are generated for each new deploy.
• Apprise is on, so Telegram, Discord, Slack, ntfy, and webhooks work without SMTP.

Why Deploy Cron Sentinel

Use this when you want cron monitoring without filling in required fields, without a public database, and without open registration. Pinning v4.4 keeps the start command and health check on a known image.

Common Use Cases

• Nightly backups, billing jobs, and other scheduled tasks
• A ping URL at the end of a script, plus /fail when the script errors
• Alerts to Telegram, Discord, Slack, ntfy, or a webhook

Do not use it as a website uptime monitor. It watches schedules, not homepages.

Dependencies for Cron Sentinel

No required variables. Optional pieces:

• SMTP (EMAIL_HOST, EMAIL_HOST_USER, EMAIL_HOST_PASSWORD, EMAIL_PORT) if you want email alerts. Many hosts block port 25. Port 587 is the usual choice.
• A Telegram bot token (TELEGRAM_TOKEN and TELEGRAM_BOT_NAME) if you want the built-in Telegram integration. Other chat apps can use an Apprise URL from the dashboard instead.

Deployment Dependencies

After deploy:

1. Open the Healthchecks URL.
2. Sign in with ADMIN_EMAIL and ADMIN_PASSWORD from the Healthchecks service variables.
3. Open My First Check and copy its ping URL.
4. Call that URL at the end of the job:

curl -fsS -m 10 --retry 5 -o /dev/null https://YOUR-DOMAIN/ping/YOUR-CHECK-UUID

Use /start before the job and /fail on the error path when you also want runtime and failure pings.

Changing SECRET_KEY later signs everyone out. The database stays private on purpose.

Docs for checks and integrations: https://healthchecks.io/docs/