Website uptime monitoring
Website uptime monitoring requests your site on a schedule, every 5 minutes on a free plan, and alerts you the moment it answers with an error or stops answering, so you hear about downtime before a visitor emails you.
Without a monitor, the usual way to learn a site is down is a message from a customer, an hour in. Uptime monitoring replaces that with a request every few minutes from outside your hosting and an alert on the first bad answer. The gap between failure and alert is the check interval: 5 minutes on the Logdash free plan, 1 minute on Builder, 15 seconds on Pro.
What a free 5-minute check is worth
At a 5-minute interval a site gets 288 checks a day, and an outage that starts right after a check goes unnoticed for up to 5 minutes. For a blog or a landing page that is fine. For a checkout it is a support thread, which is the case for Builder at 1 minute or Pro at 15 seconds. Either way the check has to come from outside your hosting. A monitor on the same server as the site goes down with it and tells you nothing.
What website monitoring catches, and what it misses
- Catches: the server answering 5xx, the host refusing connections, a hostname that stops resolving, and no reply within 10 seconds.
- Catches: a redirect loop. The check follows up to 5 redirects, so http to https to www is fine and a sixth redirect is down.
- Misses: a page that returns 200 but renders blank because a script failed to load. The check reads the status line, not the page.
- Misses: wrong text on the page. Logdash has no keyword checks yet.
- Misses: a CDN serving a cached 200 while your origin is down. Monitor a route the cache cannot answer.
The catch-all trap
Single-page apps and many hosting setups answer every path with the same HTML and a 200. Check the home page of such a site and you are checking that the host serves a file, not that your app works. The backend can be down for a day while the monitor shows 100%. Run this to see whether your site does it.
site=https://example.com
home=$(curl -sL --max-time 10 "$site/" | cksum)
fake=$(curl -sL --max-time 10 "$site/no-such-page-$RANDOM" | cksum)
if [ "$home" = "$fake" ]; then
echo "catch-all: monitor a health route instead of /"
else
echo "ok: / is a real page"
fiLogdash runs the same test when you add a URL. It requests your address and a made-up path on the same host, and if the bodies match it warns that the check only proves the host is up. It also tries /health, /api/health and /up, and offers any that answer as a one-click swap.
Website downtime alert in three steps
- Add the site Paste the URL into Logdash. If the catch-all warning appears, take one of the health routes it offers.
- Connect Telegram Link the Logdash bot to your Telegram chat once. Every monitor in the domain can then alert there, and a webhook can run beside it.
- Take the site down Stop the server or point the route at a 500. On the next check a Telegram message arrives with a red dot, the site name, the status code and the first lines of the error, and a second one when it recovers.
Logdash vs Better Stack
| Feature | Logdash | Better Stack |
|---|---|---|
| Free monitors and interval | 5 sites every 5 minutes | 10 monitors every 3 minutes |
| False alarm filter | Alerts on the first failed check | Checks from at least 4 locations, opens an incident when 3 fail |
| Free alert channels | Telegram and webhook | Email and Slack |
| Public status page on the free plan | One, custom domain on Pro | One |
| Source code | AGPL-3.0, public repository | Closed source |
When Better Stack is the better pick
- You run more than five sites and want them all on a free plan with shorter gaps.
- A single failed request from one location should not wake you. Their multi-location confirmation filters that out by default.
- You also need SSL certificate or domain expiry checks, which Logdash does not have.