StackWitness

Compare

Heartbeat and cron monitoring, plus the outside view

Heartbeat monitoring catches silent job failures: your job pings a URL each time it runs, and if the expected ping does not arrive in time, you get alerted. It is how you learn a nightly job, queue worker, or backup quietly stopped.

StackWitness does heartbeat monitoring and adds the outside-in view: measurement of the dependencies you rely on from multiple regions, plus your own endpoint monitors, all on one honest status page. If all you need is "did my cron run?", a dedicated monitor is simple and enough. If you also care whether your dependencies are up, one tool can cover both.

Two questions about background work

  • Dedicated heartbeat / cron monitors

    You ping a URL each time a scheduled job runs; if the expected ping does not arrive in time, they alert. Purpose-built for silent job failures.

    Suits: Teams whose main need is knowing a cron, queue worker, or backup stopped running.

  • StackWitness heartbeats plus outside-in monitoring

    Set how often a job should check in and a grace window for the same silent-failure alert, alongside measurement of the dependencies you rely on from multiple regions, plus your own endpoint monitors, on one status page.

    Suits: Teams that want job check-ins and dependency truth in one place, from the same measurement.

What each one covers

Dedicated heartbeat monitorStackWitness
Silent job failure (cron stopped)Core purposeSupported: expected check-in period plus a grace window
Outside-in dependency measurementUsually out of scopeCore: multi-region probes of the hosts you depend on
Your own endpoint monitoringOften limited or separateIncluded: HTTP monitors with keyword checks
One honest status pageSometimes a basic status viewCustomer-facing status backed by all of the above

How to choose

  • The only thing you need is "did my nightly job run?" A small dedicated heartbeat monitor is simple and enough.
  • You also want to know whether your dependencies are up. StackWitness folds heartbeats in next to multi-region dependency measurement and your own endpoint checks.
  • You want one honest status page. Job check-ins, dependency truth, and your endpoints all feed the same page, instead of three disconnected tools.

Where StackWitness fits

StackWitness is independent dependency and uptime monitoring with honest status pages. We measure the services you depend on ourselves, from multiple regions, and read the vendor's own status feed alongside, so when something breaks you know within minutes whether it is your provider or you. When our regions disagree, we show it as measured instead of averaging it into a comfortable green.

The paid product gives you your own monitors, clear attribution the moment a third party is the cause, and a customer-facing status page backed by that same measurement. It is the evidence you want in front of customers and your own team at 3am.

Start monitoring See live dependency truth

Public dependency truth pages are free to read. Paid product: your monitors, attribution, and a status page backed by the same measurement discipline. How StackWitness works · Blog.

Questions

How does a StackWitness heartbeat monitor work?

You tell us how often a job should check in and set a grace window. Your job sends a ping each time it runs. If a ping does not arrive within the expected period plus the grace, we alert you, the same idea as a dedicated cron monitor.

Does StackWitness parse cron expressions?

Heartbeats are configured as an expected check-in period plus a grace window rather than a full cron expression. For most nightly jobs, queues, and backups that is all you need to catch a silent failure.

Why put heartbeats next to dependency monitoring?

Because at 3am you want one place that tells you whether your background jobs are running, whether your dependencies are up, and shows customers an honest status, all from the same measurement instead of three disconnected tools.

Start with evidence you can read. Public dependency truth pages use the same measurement discipline as the paid product.

Start monitoring See live dependency truth