What Is Uptime Monitoring?

What Is Uptime Monitoring?

Uptime monitoring is the practice of automatically checking whether a website, API, or server is reachable and responding correctly - at regular intervals, from external locations, around the clock.

When a check fails, an alert goes to the on-call engineer. When it passes, nothing happens. The goal is simple: know about a problem before your users do.

How it works

A monitoring tool sends a request to your endpoint on a schedule - every 30 seconds, every minute, every five minutes, depending on your configuration. If the response comes back as expected, the check passes. If the response is slow, returns an error code, or does not come back at all, the check fails.

Most tools wait for a second failure before alerting you, to avoid false positives from transient network noise. Once two or three consecutive checks fail, an alert fires.

That alert goes wherever you configure it: email, Slack, PagerDuty, SMS, Discord. Your team knows within seconds, not hours.

The four main types of checks

HTTP/HTTPS monitoring

The most common type. The monitor sends an HTTP request to a URL and checks that the response meets your criteria: a 200 status code, a specific string present in the response body, a response time under a given threshold.

HTTP monitoring catches the most common failure modes: a server that is down, a deployment that broke a route, a CDN that is returning cached errors, a response that is technically a 200 but the page body says "Error connecting to database."

You configure the expected status code and, optionally, a keyword that must appear in the response. If your homepage returns 200 but the word "dashboard" is missing, the check fails.

SSL certificate monitoring

Your SSL certificate has an expiration date. When it expires, every browser shows your users a security warning and blocks them from accessing your site. Search engines de-index you. Enterprise customers trigger incident tickets on their end.

SSL monitoring tracks the expiration date of your certificate and alerts you with enough lead time to renew - typically 30 days, 14 days, and 7 days before expiry. It also alerts you if the certificate becomes invalid for any other reason: domain mismatch, revocation, or misconfiguration after a deployment.

This check runs independently of your HTTP checks. A site can be fully reachable and still have a certificate that expires in four days.

Port monitoring

Port monitoring checks whether a specific TCP port on your server is open and accepting connections. This is used for services that do not speak HTTP: database servers, mail servers, FTP, SSH, or any custom service running on a non-standard port.

If your PostgreSQL instance stops accepting connections on port 5432, or your Redis stops responding on 6379, port monitoring catches it.

Cron job monitoring

Cron job monitoring (also called heartbeat monitoring) works the other way around: instead of your monitor pinging your service, your service pings your monitor on a schedule. If your monitor does not receive a ping within the expected window, it alerts you.

This is for scheduled tasks - nightly database backups, daily email digests, weekly cleanup jobs. These jobs have no URL to ping. They run silently and fail silently. Heartbeat monitoring gives you visibility into whether they actually ran.

Why 99.9% uptime is not the same as 99.99%

Uptime is expressed as a percentage of time a service was available over a given period. The difference between common uptime tiers looks small on paper. In practice, it is not.

The gap between 99.9% and 99.99% is 7 hours and 54 minutes of downtime per year. For a payment processor, a scheduling API, or any SaaS where users depend on your service to do their job, that gap is the difference between a minor inconvenience and a breach of your SLA.

If you have an SLA with enterprise customers, the number on the contract is only as meaningful as your ability to measure it. Uptime monitoring gives you the data to verify your own reliability, report on it honestly, and detect trends before they become SLA violations.

The check interval matters

A monitor that checks every five minutes can only tell you something was down during a five-minute window. If a server crashes and recovers in two minutes, a five-minute interval monitor never fires. You never know it happened.

A 30-second interval catches that same failure. Over a week of production traffic, the difference in detected incidents is significant.

Check interval is also how you calculate the downtime figures you report to customers. A five-minute interval means your smallest measurable downtime unit is five minutes. If you are reporting 99.99% SLA, your check interval needs to match that precision.

What uptime monitoring does not cover

Uptime monitoring tells you whether your service is reachable. It does not tell you whether it is working correctly for every user, every flow, or every region.

A 200 response from your API does not mean a new user can sign up, complete a payment, or export a report. For that, you need synthetic monitoring - scripted browser tests that simulate real user actions.

Uptime monitoring also does not replace error tracking, logging, or performance monitoring. These tools answer different questions. Uptime monitoring answers the most urgent one first: is the service up?

What to monitor

Start with the endpoints your users interact with directly. For most SaaS products, that means:

• The main app URL or dashboard
• Your primary API endpoint
• Your authentication flow
• Any webhooks or integrations that your customers depend on
• Your SSL certificate

Then add infrastructure-level checks: the database port, any background job queues, your email delivery endpoint. Build outward from what breaks your product for users, and work back toward what breaks it silently.

Setting up uptime monitoring with Nodown

Nodown runs checks from 14 regions in parallel. An alert only fires when at least three regions confirm the failure, which eliminates false positives from single-region network noise.

From the dashboard, you can set up HTTP, SSL, port, and heartbeat monitors in under a minute each. Alert channels - Slack, Discord, Telegram, email, SMS, webhooks, PagerDuty - are available on every plan.

Each monitor feeds into a real-time status page you can share with your users. The free plan includes 10 monitors with 1-minute check intervals. Pro ($24/month) adds 100 monitors, on-call scheduling, and a branded status page with a custom domain.

Start monitoring free, no credit card required

Last updated: May 2026.