Skip to content
SilenceWatchSilenceWatchSilenceWatch

Getting started: monitor a job in 5 minutes

This page takes you from nothing to a first monitored job with a working alert. Allow five minutes.

  1. Create an account at app.silencewatch.com. A first project is created with the account. Checks, alert channels and history belong to a project: you can create more (one per customer or per team, say) in Settings › Projects.

  2. Create a check. In Checks, click New check and fill in:

    • the name (“Nightly backup”);
    • the schedule, either “every N seconds” or a cron expression with its time zone;
    • the grace period: how late a run may be before you are alerted;
    • an optional environment (production, staging) so jobs do not get mixed up.

  3. Copy the ping URL. It is shown at the top of the check’s page, with a ready-to-paste curl command. It is the only credential your job needs: keep it secret.

  4. Have your job call that URL at the end of a successful run:

    Terminal window
    /usr/local/bin/backup.sh && curl -fsS -m 10 --retry 3 https://app.silencewatch.com/p/<ping-key>

    For your environment (systemd, Docker, Kubernetes, GitHub Actions…), see the examples.

  5. Add an alert channel. In Alerting, add an email address, a webhook, Slack, Microsoft Teams or Discord, then click Send test. A channel nobody has ever tried is a channel that fails during the incident it was set up for.

  6. Check it works. The check is waiting until a first ping arrives, then goes UP on the first call. If a ping is missing it goes late, then down once the grace period has passed, and every enabled channel is alerted.

Next

Section titled “Next”