Alerts & notifications

Hear about real problems, not blips.

Send alerts to email, Telegram, Slack or your own system. CuliPulse waits for a problem to repeat, and can ask several locations to agree, before it tells anyone.

Alerts to email, Telegram, Slack or your own system, only after a problem repeats or several locations agree.

Included on every plan

CuliPulse APP
#ops-alerts · 03:12
Checkout API — DOWN
Website · api.example.com/health
Connection timed out after 10s
Locations failing
2 of 2
Down from
Singapore, Frankfurt
Down since Sep 11, 03:12 · Website
View incidentOpen monitor
Example Slack alert · illustration (real capture pending)Example Slack alert · illustration

Where alerts go

Four places. Email is on from the start.

Every channel has a Send test button, so you see what an alert looks like before anything breaks. A test is always marked, so nobody mistakes it for an outage.

Email
your sign-in address

Alerts go to the email address you signed in with. On from the moment you sign up, and you can turn it off anytime.

Already on — nothing to set up
Telegram
you or a group

Tap a link, press Start in Telegram, and you are done. Works for a private chat or a whole team group.

Link opens Telegram → Start
Slack
any channel

Click Add to Slack and pick a channel. Alerts arrive with the details and buttons to the incident and monitor.

Add to Slack → pick a channel
Webhook
your own system

Your endpoint gets a signed JSON message for every alert. We send a test delivery before saving it.

Paste a URL → test → save
Send each monitor to the right place

Telegram, Slack and webhook channels can get every monitor, including ones you add later, or only the ones you pick.

#ops-alertsSlack→Every monitor
Payments teamTelegram→Checkout API, Payments webhook
Incident botWebhook→Every monitor
Example routing · illustration

Route Telegram, Slack and webhook channels to every monitor, or only the ones you pick.

Only when it's real

Two simple settings keep the noise out.

Both live on each monitor. The defaults work for most people.

01
Wait for it to happen again

A single failed check is often a hiccup. Choose how many failed checks in a row it takes before we alert you.

Alert after this many failed checks in a row.

Alert after this many failed checks
12345
2 is the default. We mark it back up after one good check, or more if you prefer.
02
Ask more than one location

A problem on one network path isn't an outage. Check from several places and only alert when more than one sees it.

Only alert if more than one location fails
Singaporefailing
Frankfurtfailing
Office agentok
2 of 3 agree, so the alert goes out.
Check from 1 checkpoint on Free, up to 3 on Solo and 4 on Pro.
2 of 3 agree, so the alert goes out. Up to 3 checkpoints on Solo, 4 on Pro.

What you'll be told

One message when it breaks. One when it's fixed.

No repeats every minute while something stays down. Each message says what happened in plain words.

DownConnection timed out after 10s
Back upBack up after 13 minutes
DegradedGitHub reports "Actions" degraded
Job missedNightly backup did not check in
Expiring soonCertificate expires in 5 days
Agent offlineOffice agent stopped reporting

Send yourself a test alert.

Connect Telegram or Slack and see your first message in under a minute.

See your first message in under a minute.