Cron job monitoring tools with a free tier: 7 options compared

Free tiers compared

Here is how the seven tools compare on their free limits, checked on each vendor's pricing page on October 30, 2026:

Runpeek watches each run from the inside; the other six are heartbeat monitors that alert you when a scheduled ping is late. Healthchecks.io has the largest hosted free tier among them.

How cron job monitoring works

You can watch a cron job in two ways: wrap the command, or have it send a heartbeat. A wrapper such as Runpeek's tqdml run starts the job itself, so it sees every line of output and the exit code.

A heartbeat, also called a dead man's switch, is an HTTP request your job makes to a unique URL when it finishes. The service expects that ping every period plus a grace period and alerts you when one is missing.

If you chain the ping with &&, the request only fires when the script exits with 0, so a failed run turns into a missing ping once the grace period runs out. Healthchecks.io and Cronitor also take a start signal to measure duration and catch hung jobs, and a fail signal so you hear about a failure right away instead of waiting for the timer.

0 3 * * * /usr/local/bin/backup.sh && curl -fsS -m 10 --retry 5 -o /dev/null https://hc-ping.com/your-uuid
0 4 * * * curl -fsS -m 10 --retry 5 -o /dev/null https://hc-ping.com/your-uuid/start; /usr/local/bin/report.sh; curl -fsS -m 10 --retry 5 -o /dev/null https://hc-ping.com/your-uuid/$?

The first line only pings if the backup succeeds. The second sends /start before the job and then /$?, which reports the exit code (0 is success, anything else is a failure) the moment the script finishes.

The 7 options

Runpeek

With tqdml run in the crontab line, each run streams its output and tqdm bars to a live link that keeps the last 500 lines (10,000 on Pro) and ends as Finished on exit code 0 or Failed on anything else. A free account keeps runs for 7 days and adds 3 alert rules.

Healthchecks.io

The $0 Hobbyist plan monitors 20 jobs and keeps 100 log entries per job. Jobs ping https://hc-ping.com/<uuid>, and the /start, /fail and /<exit-status> suffixes record run times and failures. SMS and WhatsApp credits start on the $20 Business plan with 100 jobs, and the server is open source (BSD 3-clause) if you'd rather host it yourself.

Better Stack

Better Stack is free for personal projects: 10 monitors and heartbeats, 1 status page, and Slack and email alerts. A heartbeat waits for a request to its secret URL at the frequency you set plus a grace period, and adding /fail to the URL reports a failure right away.

Cronitor

Cronitor's free plan covers 5 monitors with email and Slack alerts but no SMS. Jobs report with curl https://cronitor.link/p/<API_KEY>/<job-name>?state=complete, and state=run and state=fail mark starts and failures. Past 5 monitors you pay $2 per monitor and $5 per user each month.

Dead Man's Snitch

Dead Man's Snitch gives you 1 snitch for free, and the $5 plan has 3. You append && curl https://nosnch.in/<token> to the job, so the check-in only happens when the script exits cleanly. One snitch covers a single backup; a second job means paying.

Sentry Crons

Sentry includes 1 cron monitor on every plan, including the free Developer plan, and charges $0.78 for each extra one on paid plans. Instead of curl, you export your project's SENTRY_DSN and wrap the job with sentry-cli monitors run, which reports missed runs and runs that exceed their maximum runtime.

export SENTRY_DSN="your-project-dsn"
sentry-cli monitors run --schedule "0 3 * * *" nightly-backup -- /usr/local/bin/backup.sh

Uptime Kuma

Uptime Kuma 2.5.8 (October 26, 2026) is open source and self-hosted, so there are no plan limits. Its Push monitor waits for your job to call a URL, next to HTTP, TCP, ping and DNS checks, and it alerts through Telegram, Discord, Slack, email and 90+ other services.

docker run -d --restart=always -p 3001:3001 -v uptime-kuma:/app/data --name uptime-kuma louislam/uptime-kuma:2

Keep it on a different machine than your cron jobs: if both share a box that dies, nothing sends the alert.

Cron mistakes that look like monitoring failures

Many silent failures happen before your script or its curl ever runs. Cron runs commands with /bin/sh, sets LOGNAME and HOME, and uses a short PATH (just /usr/bin:/bin on Debian 13), so a command that works in your terminal can fail with "command not found".

• Cron turns an unescaped % into a newline and sends the rest to the command's stdin, so write date +\%F.
• Redirect output with >> /path/job.log 2>&1, or set MAILTO to mail it to you (MAILTO="" turns mail off).
• Wrap the command in flock -n /tmp/job.lock so a slow run makes the next one exit instead of stacking up.
• On Debian and Ubuntu, journalctl -u cron shows whether cron started the job at all.

MAILTO=you@example.com
PATH=/usr/local/bin:/usr/bin:/bin

*/5 * * * * flock -n /tmp/sync.lock /opt/jobs/sync.sh >> /var/log/sync-$(date +\%F).log 2>&1

How to choose

• Each run's live output, exit code and an alert when it fails or stalls: Runpeek, with tqdml run in the crontab line.
• A hosted heartbeat for a handful of jobs: Healthchecks.io for 20 jobs, or Better Stack for 10 monitors and a status page on a personal project.
• A team already on Sentry: Sentry Crons, at $0.78 per monitor past the first on a paid plan.
• No third-party service: Uptime Kuma's Push monitor on a separate machine.