Email and Telegram notifications

An alert at three in the morning for a disruption that already fixed itself isn't caution, it's noise. And after enough noise, even the real alert, the one that matters, ends up ignored along with the rest. Etizei's alert logic starts from there: a disruption only becomes a real incident, and triggers a notification, after a number of consecutive failed checks that you decide on, not at the first isolated hiccup.

A concrete case

Raise the confirmation threshold to just two or three consecutive failed checks, and a twenty-second timeout that resolved on its own before you'd even opened your eyes stops waking you up in the middle of the night. What's left, almost always, is an alert that genuinely deserves attention.

In a team, the same logic scales up: you can choose whether an alert goes out by email, to a shared Telegram channel, or both, so whoever's on call knows without checking the dashboard out of habit. A shared channel also works as a record: anyone joining later can scroll back and understand what happened.

Self-service or assisted setup

The starting values (two failed checks before an alert, no silenced hours) work well for most sites. If your case is different, say a service with predictable traffic spikes or scheduled maintenance, we can help you set thresholds and quiet hours that fit better, as part of the rest of your monitoring setup.