Uptime monitoring you can script
Watch your APIs, cron jobs, ports and certificates, create monitors and mute alerts from your pipeline with a REST API, and get a signed webhook when something breaks. Plain HTTP, no agent to install.
- Checks every 60 seconds
- REST API and signed webhooks
- Sign in with GitHub
curl -X POST https://api.alertsleep.com/public/v1/monitors \
-H "AuthToken: $ALERTSLEEP_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"friendly_name": "Orders API",
"url": "https://api.example.com/health",
"type": "http",
"interval_seconds": 60
}'
What you can watch
Six monitor types for the parts of a stack that fail at night.
Wire it into your workflow
Everything below is a plain HTTP request. Use curl, your CI or any HTTP client.
Mute alerts during a deploy
Open a maintenance window from the pipeline and end it when the deploy is done. A window has a maximum length, so a job that dies halfway cannot mute your monitors for good.
# before the deploy: mute alerts for at most 15 minutes
START=$(date -u +%FT%TZ); END=$(date -u -d "+15 min" +%FT%TZ)
WINDOW_ID=$(curl -fsS -X POST https://api.alertsleep.com/public/v1/maintenance-windows \
-H "AuthToken: $ALERTSLEEP_TOKEN" -H "Content-Type: application/json" \
-d "{\"title\":\"Deploy $CI_COMMIT_SHORT_SHA\",\"starts_at\":\"$START\",
\"ends_at\":\"$END\",\"monitor_ids\":[\"$MONITOR_ID\"]}" \
| jq -r .maintenance_window.id)
./deploy.sh
# after it: end the window early
curl -fsS -X POST https://api.alertsleep.com/public/v1/maintenance-windows/$WINDOW_ID/end \
-H "AuthToken: $ALERTSLEEP_TOKEN"Heartbeats from cron
Ping /start when the job begins, so a run that hangs is caught too, and the plain URL when it succeeds. On failure, ping /fail with the exit code and the e-mail goes out at once instead of after the grace period.
# crontab: alert me if the nightly backup fails or never runs
# (one line in crontab)
0 3 * * * curl -fsS https://api.alertsleep.com/v1/ping/YOUR-TOKEN/start; /usr/local/bin/backup.sh && curl -fsS https://api.alertsleep.com/v1/ping/YOUR-TOKEN || curl -fsS https://api.alertsleep.com/v1/ping/YOUR-TOKEN/fail -d exit_code=$?React to an outage in your code
The webhook integration posts JSON on monitor.down, monitor.up and certificate events. With a secret set, every request carries an HMAC-SHA256 signature over the timestamp and the body, so you can reject forged and replayed calls.
// Node: verify an AlertSleep webhook
const ts = req.headers['x-alertsleep-timestamp'];
const expected = 'sha256=' + crypto
.createHmac('sha256', process.env.ALERTSLEEP_SECRET)
.update(ts + '.' + rawBody).digest('hex');
const ok = crypto.timingSafeEqual(
Buffer.from(expected),
Buffer.from(req.headers['x-alertsleep-signature']));
// body: {"version":1,"event":"monitor.down",
// "sent_at":"…","resource":{…},
// "data":{"status":"down","reason":"HTTP 503 …"}}A status badge in the README
Every public status page has an SVG badge with its current state or its 30-day uptime, linked back to the page.
<!-- README.md -->
[](https://status.alertsleep.com/your-page)
<!-- 30-day uptime instead of the current state -->
Public API at /public/v1 with an API token in the AuthToken header, up to 500 requests a minute per token.
Alerts where you already are
Pick the channels per monitor. A failed check is repeated from a second location first, so a blip on our side of the network wakes nobody.
- The alert says what failed: the status code, the timeout, the DNS or TLS error, or the missing keyword.
- A recovery notice follows on the first good check.
- SMS and phone calls are included in the plan, from $59 a year.
- Alerts can be written in English, German, French, Spanish, Polish or Czech.
Is it them or is it you?
When a deploy fails or an API times out, check the services your stack runs on. Each page shows our own checks and the provider's official incidents.
An account you can trust with tokens
Monitors hold credentials to your systems, so the account is protected like one.
Sign in with GitHub
Two-factor authentication
Encrypted credentials
Revocable API tokens
What AlertSleep is not
So you can decide before you sign up.
- There is no CLI, SDK or Terraform provider. The API is plain JSON over HTTPS, so curl and any HTTP library work.
- It is not an APM or a log tool: no traces, no error tracking, no real-user metrics.
- One monitor sends one request. Multi-step flows and JSON schema assertions belong in a health endpoint of your own.
- Checks run from one location and every failure is confirmed from a second one; there is no network of checks from many countries.
Free tools for debugging
No account needed.
Frequently Asked Questions
Does AlertSleep have an API?
Is there a CLI or an SDK?
How do I stop alerts during a deploy?
How do I monitor a cron job or a queue worker?
How do I verify the webhook signature?
Can I monitor APIs behind authentication?
How often are checks run?
Can I try it for free?
Add your first endpoint in a few minutes
Create a monitor in the dashboard or with one API call. Try it free for 14 days, no card required.