Alerts
The only message that counts is the one you still read
A tool that alerts too much is a tool you mute, and a mute never gets turned back on. PostShip sends nothing while everything is green.
“We ended up creating a dedicated channel so we'd stop seeing them.”
Alert fatigue comes from lost trust more than from volume: as soon as one message in three requires no action, the brain learns to ignore them all — including the right one.
- without duplicates
- 10 min
- message per incident
- 1
- channels to choose from
- 4
- to mute everything
- 1 click
without duplicates
message per incident
channels to choose from
to mute everything
The rules that make the silence
01Nothing when all is well
No daily “all green” report, no check receipt.
02One incident, one message
Five URLs that fall with the same deploy make one alert with five lines, not five alerts.
03Your hours, your time zone
Quiet hours per project, in your time zone and not the server's.
04Where you already are
Email, Discord, Slack, Telegram, or your own signed webhook if you'd rather plug in yours.
Channels
- Discord
- Slack
- Telegram
- Outgoing webhook
What we're often asked
Do I still get the recovery message?
Yes, and it's the only other message: knowing it's back saves you from going to check. It can be turned off independently of the failure alert.
I'm going on holiday, what do I do?
A maintenance window on the project, or “Mute 1 h / 4 h / 24 h” from the Incidents page — also available as a command from the Telegram bot. Checks keep running, only the alerts go quiet.
And if your own server goes down?
It can't send you an alert to tell you — no monitoring tool can. What it does is show publicly the last run of its scheduler, so that silence can be verified instead of merely reassuring.
Paste your site's URL. PostShip checks it in front of you, no account and no card.