CronWatch
The Cron Job Monitoring Tool Built for Developers Who Run Background Jobs at Scale
See every scheduled task's last run time, success rate, and missed-schedule alert in one place — across any stack, any language, any host.
Start Monitoring Free
14-day free trial. No credit card. Cancel anytime.
Your cron jobs are running right now. Some of them are failing. You will not know until a user files a ticket.
Every observability platform on the market covers HTTP requests, error traces, and infrastructure metrics in depth. Background job visibility is either an afterthought buried four menus deep in Datadog, or something your team plans to wire up "eventually" using a custom Slack webhook and a jobs table you query by hand. The cron job monitoring layer that backend developers need — last run time, execution duration trend, missed-schedule detection, failure rate per job — is simply not what those platforms were built for.
The market assumed someone else would handle it. Nobody did.
When a Scheduled Task Fails Silently, the Clock Starts Without You
Datadog tracks your infrastructure. Sentry catches runtime exceptions. Vercel shows your deploys. None of them tell you that your nightly invoice generation job missed its 2am run for the third time this week.
Background job monitoring sits in a blind spot that every current tool has consciously decided to leave empty. Sentry fires on raised exceptions — a job that completes without raising but produces nothing does not trigger it. Datadog can show you CPU during the run window, but not whether the job ran at all. Railway and Render will tell you your container is healthy while the worker inside it is stuck in a retry loop burning through your queue.
The consequence is a specific, expensive kind of failure: silent correctness bugs. Data pipelines that stop enriching. Subscription renewal jobs that stop charging. Webhook retry queues that stop draining. Your system looks healthy because your infrastructure is healthy. The work is not being done.
Backend developers with 10, 20, or 40 scheduled tasks reach for the same patchwork: a last_run_at column in Postgres they check by hand, a dead Slack bot from two engineers ago, a crontab comment that says "TODO: add alerting." It is not a monitoring strategy. It is a prayer.
Introducing CronWatch
CronWatch gives every background job a heartbeat. Add one line to any job — a ping to CronWatch's endpoint at the start, one at the end — and you get a live record of every run, every failure, every missed window, and every duration anomaly. No agents to install, no YAML config, no cloud provider lock-in. Works with any scheduler: system cron, Sidekiq, Celery, Temporal, GitHub Actions, Kubernetes CronJobs, or a plain setInterval in Node.
What You Get — $19/month
Job run history — A full log of every execution: start time, end time, duration, exit status, and any output your job sends. Searchable, filterable, and retained for 90 days.
Missed-schedule detection — Set the expected interval for any job. CronWatch alerts you when a job has not checked in within that window, before the missed run causes downstream damage.
Silent failure alerts for background job monitoring — Jobs that complete without error but produce no output, process zero records, or finish faster than a configurable threshold trigger a separate alert class. The jobs that slip past Sentry are the ones CronWatch is built for.
Duration trend graphs — See how long each job takes across every run. A job that used to finish in 4 seconds and now takes 40 is a leading indicator of a problem, not a lagging one.
Multi-stack ping API — One HTTP endpoint, any language. Ruby, Python, Go, Node, PHP, or a plain curl in a shell script. Copy the snippet from the dashboard, paste it into your job. Done.
Scheduled task failure alerts via Slack, email, or PagerDuty — Choose where alerts go per job. A missed nightly billing run wakes someone up. A missed analytics summary sends an email.
Team access with per-job ownership — Assign jobs to team members. When a job alerts, the right person gets the page.
Status page for internal job health — Share a private URL with your team showing live status for every monitored job. No more "did the import job run?" in Slack.
Why $19/month
CronWatch is priced for a single backend developer or a small team who can not justify a Datadog contract to monitor their cron jobs. The $19 tier covers up to 50 monitored jobs with 90-day history. A comparable observability setup in Datadog — assuming you can get custom metric ingestion configured, which requires a platform engineer — runs $150+ per month before you have built the dashboards. One missed billing run that churns a customer costs more than a year of CronWatch.
Who This Is For
You run scheduled tasks in production and have no systematic way to know which ones ran last night.
You have been burned by a job that completed without raising an exception but also without doing anything.
You work across multiple projects — client work, side projects, or a monorepo with a dozen services — and job health lives in separate places, if it lives anywhere at all.
You are a backend developer on a small team where "we should add cron monitoring" has been on the to-do list for two quarters.
You use Sidekiq, Celery, Temporal, system cron, or GitHub Actions and want one view across all of them.
The CronWatch Missed-Run Guarantee
If CronWatch fails to alert you within 5 minutes of a missed scheduled run, the month is free. Claim it by opening a ticket with the job name and missed window — no questions, no hoops.
In 30 Minutes, You'll Have:
- Every scheduled task pinging CronWatch with a run record, start time, and exit status
- Missed-schedule alerts configured for your highest-priority jobs
- A Slack channel receiving failure notifications instead of your users filing tickets
- Duration baselines set so anomalies surface before they become incidents
- Silent failure detection running on jobs that currently pass through Sentry without a trace
- A single dashboard your team can open to answer "did the jobs run last night?" in under 10 seconds
- 90 days of run history for every monitored job, available for debugging without a database query
Frequently Asked Questions
Does CronWatch work with Sidekiq and background job monitoring in Ruby on Rails?
Yes. Add CronWatch.ping(job_id: "your-job-name") at the start and end of any Sidekiq worker's perform method. The Ruby gem wraps the HTTP ping and handles timeouts so a CronWatch outage never blocks your job. Works the same way with ActiveJob and plain Ruby scripts.
CronWatch vs Cronitor: what's the difference?
Cronitor is the closest existing product. CronWatch adds silent failure detection — the ability to alert when a job completes with a successful exit code but processes zero records or produces no output. That class of failure is invisible to heartbeat-only monitors. CronWatch also includes duration trend graphs and per-job team ownership out of the box on the base tier, both of which require Cronitor's higher plans.
How long does setup take?
Most developers add their first monitored job in under 10 minutes. Create an account, copy your job's ping URL from the dashboard, add the HTTP call to the start and end of your job. The dashboard shows the first run record the next time your job executes. Full coverage for 10 jobs typically takes under an hour, including configuring alert destinations.
What happens if CronWatch is down when my job tries to ping?
The ping client has a 2-second timeout and fails silently — your job continues regardless. CronWatch maintains 99.9% uptime and publishes a live status page. If a ping is missed due to a CronWatch outage, the missed-run alert is suppressed and logged so you are not paged for a problem on our side.
What it is: A cron job monitoring tool that tracks every background job run, alerts on missed schedules, and catches silent failures that Sentry and Datadog miss.
What you get: Job run history, missed-schedule detection, silent failure alerts, duration trend graphs, multi-language ping API, Slack/email/PagerDuty alerting, team access, and an internal status page.
Price: $19/month for up to 50 jobs and 90-day history.
Catch: Jobs need one HTTP ping added at the start and end of each run. Five minutes of copy-paste per job.
Guarantee: If CronWatch misses alerting you on a missed run within 5 minutes, the month is free.
Start Monitoring Free