A monitor is only as useful as the alert it sends. Email works, but email is slow to notice, easy to miss, and often silenced overnight. Most people want the alert where they already are: their phone or their team chat. Here is how to set that up with ServerMonitor.
Before you start
You need a ServerMonitor account and at least one site added. Telegram alerts are available on every plan, including the free one. Slack and webhook alerts are on the paid plans (Pro and Ultimate). See the plans.
Open Settings, then Notifications. This is where every channel is created.
Telegram
Telegram is a good first choice. Alerts arrive on your phone within seconds, with a notification sound.
- In Settings, Notifications, choose Add a channel, pick Telegram, give it a name such as "My phone", and add it.
- ServerMonitor shows a link. Open it in Telegram and press Start in the chat with the bot.
- The page confirms with Connected.
- Press Send test to receive a test message, and Sample alert to see exactly what a real outage and a recovery message look like.
Each person who wants alerts on their own phone connects their own Telegram channel the same way.
Slack
- In Slack, create an incoming webhook for the channel where alerts should appear, using the Slack app settings (search for "Incoming Webhooks"). Slack gives you a URL that starts with
https://hooks.slack.com/services/. - In ServerMonitor, choose Add a channel, pick Slack, give it a name, and paste the URL.
- Press Send test. A message appears in the Slack channel.
Treat the webhook URL as a secret, because anyone who has it can post to that channel.
Webhooks, for your own tools
A webhook sends each alert as an HTTP request to an address you control. It is the way to connect ServerMonitor to anything else: a status board, an on-call tool, an automation platform, or your own code.
- Choose Add a channel, pick Webhook, and enter the address that should receive the alerts. Production webhooks must use
https. - ServerMonitor creates a signing secret and shows it once. Save it.
- Your endpoint can verify each request by recomputing the signature with that secret, so you know the request really came from ServerMonitor and was not tampered with.
- Press Send test to see the payload.
Choose who gets which alerts
If you have several channels, you can decide which one is the default. A common setup is Telegram for yourself, a Slack channel for the team, and email as a backup that is always on.
Quiet hours: sleep without missing real emergencies
Open a channel's Quiet hours to choose a time window, such as 22:00 to 07:00, and a timezone. Alerts that arrive during that time are held and sent as a summary afterwards. You can allow critical problems to get through anyway, so a full outage of your main site can still wake you.
Keep alerts useful
- Confirmed failures only. ServerMonitor re-checks a failure before it alerts you, so a one-second network blip does not wake anyone.
- Throttling. If many events happen in a short time, they are summarised instead of sent one by one.
- Maintenance windows. Schedule planned downtime so incidents are recorded but nobody is paged.
- Test regularly. Send a test message now and then, and especially after you change a Slack channel or a phone.
A good starting setup
| Channel | Used for |
|---|---|
| Telegram | Your own phone, always on |
| Slack | The team, in a channel about operations |
| A backup, and a record |
With that in place, every outage, expiring certificate and security finding reaches you first. For what to monitor in the first place, see what to monitor on your website, and how often.