Your site is up. Your background jobs might be dead.
Stripe syncs, invoice emails, trial-expiration jobs, backups, queue workers — the work that runs your SaaS fails quietly. Your jobs ping Holter when they finish; if a ping goes missing, Holter opens an incident and tells you.
No credit card · free plan · up and watching in minutes.
The problem
Uptime monitoring watches your website. It says nothing about the Stripe sync that stopped last Tuesday, the nightly backup that has failed all week, or the invoice emails that quietly stopped sending. As a founder, you find out when a customer complains — or when the month-end numbers do not add up.
Set up in three steps
Add a heartbeat monitor
Name the job and set how often it should check in — every few minutes for a worker, daily for a backup.
Ping on success
Append a curl to the job so it calls the heartbeat URL when it finishes. Curl the /fail URL to alert the moment a run errors.
Get alerted when it goes quiet
If the expected ping does not arrive, Holter opens an incident and routes it to email, Slack, Discord, or a webhook.
What Holter does
Holter watches the jobs a status check cannot see. Add one curl call at the end of a cron, queue worker, or scheduled task, and Holter expects a ping on your schedule. Miss the window and it opens an incident and alerts your channels — email, Slack, Discord, or a webhook. No agent, no SDK, no access to your servers.
- ✓ One curl at the end of a cron — no agent, no SDK
- ✓ Covers backups, queue workers, scheduled jobs, and pipelines
- ✓ A /fail URL alerts the instant a run errors
- ✓ Free for up to 5 monitors, no credit card
Simple, transparent pricing
Free is enough to prove the heartbeat flow on your most critical job. Pro adds faster checks, more monitors, and Slack/Discord/webhook routing, from $15/mo.
Free plan needs no card. Every paid plan includes a 14-day free trial.
See full pricingQuestions, answered
Anything that runs on a schedule and can make an HTTP request when it finishes: cron jobs, queue workers, nightly backups, ETL pipelines, billing runs, and report generation.
No. You add one curl call (or a single HTTP request) at the end of the job. There is nothing to install and no access to your servers.
As soon as the expected ping is late past the grace you set. You can also curl a /fail URL to alert the instant a run errors, before the window closes.
No. The Free plan covers up to 5 monitors with no card.
Your site is up. Your background jobs might be dead.
No credit card required. Free plan available.
Create your first heartbeat