Alerts Configuration
Setting up email alerts so you're notified within minutes when your app goes down — and when it recovers.
A monitor that checks your app every 5 minutes is useless unless it tells you when something is wrong. Alert contacts are the mechanism that converts a failed check into an email in your inbox.
UptimeRobot separates the configuration into two steps: first you create an alert contact (who to notify), then you attach that contact to your monitors (when to notify them). This design means you can have one contact list and reuse it across all your monitors.
The "Why First" Scenario
You set up UptimeRobot, created two monitors, and they're both showing green. You feel like you're covered.
Three weeks later, your Supabase project exceeds the free tier row limit at 2am. Supabase pauses the project. The monitor detects the failure — but nobody configured an alert contact. UptimeRobot logs the failure internally but sends no email.
You find out Monday morning when a client calls.
The monitor without an alert contact is like a smoke detector with no battery. It detects the fire. It does nothing about it.
Create an Alert Contact
Go to My Settings
Click your profile icon in the top right → "My Settings". Or look for the "Alert Contacts" section in the main navigation.
Click "Alert Contacts" tab
Alert Contacts page. Click "+ Add Alert Contact" to add an email destination. Each contact must verify their email address before alerts will route to it.
You'll see the alert contacts list — currently empty.
Click "+ Add Alert Contact"
A dialog opens.
Set Alert Contact Type to "E-mail"
The free plan supports email alerts. Paid plans also support SMS, voice calls, Slack, Teams, Telegram, PagerDuty, and webhooks. Email is the correct choice for this workshop.
Enter Friendly Name
This is just your label for this contact. Use your name:
For trainees working on EduTrack:
Enter the Email Address
Use the email address where you want to receive alerts. This should be an email you actually check — not one you set up for a project and rarely open.
For the EduTrack project, this is the lead developer's email (or Subhash's email for production incidents).
Click "Save Changes"
UptimeRobot sends a verification email to the address you entered. Open it and click the confirmation link. The contact will show as unverified until you do this — and unverified contacts do not receive alerts.
Check your spam folder for the verification email. Email services sometimes flag UptimeRobot's verification emails as promotional or spam. If you don't see the email within 5 minutes, check spam before assuming it didn't arrive.
Attach the Alert Contact to Your Monitors
Creating the alert contact is not enough. You must also attach it to each monitor you want it to watch. UptimeRobot does not auto-attach new contacts to existing monitors.
Go to My Monitors (the main dashboard)
Click on your first monitor
Click the name of the monitor — "EduTrack — Production". This opens the monitor detail view.
Click "Edit"
You'll see all the settings you configured when creating the monitor.
Scroll down to "Alert Contacts to Notify"
You'll see a list of your alert contacts. Check the box next to the contact you just created.
Set "Alert After X Minutes" to 0 minutes
This setting controls how long UptimeRobot waits before sending the downtime alert after detecting a failure. Setting it to 0 means you're notified on the first confirmed outage, with no additional delay.
How UptimeRobot confirms an outage: It doesn't alert on the very first failed check — it waits for 2 consecutive failures before confirming an outage. This prevents false alarms from momentary network blips. So in practice, the alert arrives after 2 check intervals (10 minutes on the free plan) even with "alert after 0 minutes" set.
Click "Save Changes"
Repeat for your Supabase monitor
Go back to the monitor list. Click "EduTrack — Supabase API" → Edit → add the same alert contact → Save Changes.
What the Alerts Look Like
Downtime alert email:
This tells you: what went down, which URL, why (the HTTP error code), and exactly when.
Recovery alert email:
This tells you the app is back and how long it was down. If you see "14 minutes", that's likely one incident — a bad Vercel build that was quickly rolled back.
Reading the HTTP Error Codes in Alerts
The "Reason" field in the downtime alert tells you what the monitor received. Each code points to a different cause.
| Error code | What happened | Likely cause |
|---|---|---|
| HTTP 500 | Server error | Bad deploy, unhandled code error |
| HTTP 503 | Service unavailable | Vercel scaling issue, server overloaded |
| HTTP 404 | Not found | Wrong URL in the monitor, or page was deleted |
| Timeout | No response received | Server completely unreachable, Vercel outage |
| Connection refused | Server actively refused | Firewall rule, server shutdown |
| DNS resolution failed | URL doesn't resolve | Domain expired, DNS misconfigured |
When you get an alert, the error code is your first diagnostic clue before you even open your browser.
Free Plan Alert Limits
Free plan sends unlimited email alerts. There is no cap on how many downtime/recovery emails UptimeRobot sends. If your app goes up and down 20 times in a day, you get 40 emails. The free plan limits are on the number of monitors (50) and check frequency (5 minutes) — not on alert volume.
When to Add More Alert Contacts
For EduTrack specifically, consider adding:
| Contact | When to add |
|---|---|
| Your own email | Always — you're the primary |
| Subhash's email | When a trainee is the primary developer and Subhash is the backup |
| A team group email | When multiple people need to be aware of outages |
For personal projects and CA firm websites during training: one email contact is sufficient.
Silencing Alerts During Maintenance
When you're doing a planned maintenance — running a migration, deploying a major update — you know the app may be briefly down. You don't want a false alarm during that window.
To pause a monitor temporarily:
- Click the monitor → Pause button
- The monitor stops checking — no alerts fire
- When your maintenance is done → Resume
Always resume the monitor when you're done. A paused monitor provides no protection.
What You Have Now
- An alert contact pointing to your monitoring email
- Both monitors (production app + Supabase API) attached to that contact
- Email alerts will fire within 10 minutes of any downtime starting
- Recovery emails will confirm when the app comes back up
The next step is creating a public status page — so your clients can check app status themselves instead of calling you.