Know a product is down before a customer tells you.
LopeBase checks each product's address every 15 minutes, records the status and response time, and flags any site that stops answering. Point it at a health address and it watches the database, queue, and other services behind the site too.
No account or key needed. Just the addresses to check.
What each check records
Uptime is the one connector that reads no other tool. LopeBase itself sends a plain GET request to the address you give it, from its own servers, and writes down what came back.
- Whether it answeredAny status under 400 counts as an answer. A 4xx or 5xx, a timeout, or a connection error counts as down.
- HTTP statusThe exact code the site returned, so an outage reads as "HTTP 502" rather than just "down".
- Response timeHow long the answer took, in milliseconds. Slower than 3 seconds is recorded as slow, which still counts as up.
- The error it sawA timeout, a refused connection, or the name of a failing service, kept with the check so the outage list can say what happened.
- One row per serviceWhen the address is a LopeBase health address, each service it reports gets its own row with its own state and timing.
How to set it up
There is nothing to create in another service. You only need the address.
- 01
Open Uptime. A property with no check yet is listed at the bottom of the Overview tab with an Add check button. Press it.
- 02
Paste the address. A health page like /health is best, since it answers fast and says more than a home page. The home page works too.
- 03
The first result lands on the next 15-minute pass. Live status is right from that first check; the longer windows fill in as history builds.
Already running the LopeBase Adapter? Connecting it adds its health address as the property's uptime check for you, unless the property already has one. Full steps are in Uptime checks and incidents.
Watching the services behind a site
A home page can load from a cache while the database behind it is down. A health address closes that gap.
- Declare your servicesWith the LopeBase Adapter, list the services your app depends on, like the database, a queue, a payment provider, or a second API. Its health address is /api/lopebase/health by default.
- Real checks, not status pagesEach service is tested with your app's own credentials and a cheap call, like select 1 or a Stripe balance lookup. That proves the integration works, not just that the vendor's servers are up.
- Critical or notA critical service failing marks the site down. One you mark non-critical marks it degraded instead, shown amber and labeled apart from a real outage.
- Up to 50 servicesLopeBase reads up to 50 services from one health address, and keeps up to 20 per check for status pages.
What shows in the console
Everything lands on the Uptime page, with anything down pinned at the top and a card for every site, problems first.
- Uptime by window24 hours and 7 days from single checks, 30 days on each card, and 90 days and 12 months from daily totals. A window with no checks says "Collecting", never a made-up 100%.
- A 30-day stripOne block per day: green for clean, amber for degraded, red for any failed check. Hover a day for its tally.
- OutagesEvery outage in the last 30 days, with when it started, how long it lasted, and what the first failed check saw.
- History keptSingle checks for 35 days, and a daily total for each site for over a year.
- A public status pageFree on every plan, with your notes, an RSS feed, and a live badge. See Publish a status page.
- AlertsA site going down reaches you by email or browser push, once when it starts and once when it is fixed. See Alerts.
Questions it answers
Uptime is most useful next to the other connections on the same property, so a dip has a cause beside it.
- Is everything up right now?One line at the top of Uptime says so, across every property you run.
- Which site has been flaky this month?Each site's card leads with its 30-day uptime and day strip, so the one with red days stands out.
- What broke, and what do I fix first?The incident review reads 30 days of outages with the top errors from Sentry, Bugsnag, Rollbar, Honeybadger, or PostHog, and drafts a status note after real downtime. It uses AI credits.
- Is it the database or the app?A health address names the failing service. Pair it with Supabase for project health, or Cloudflare for Worker error rates.
- Can my agent check?Over MCP, get_health lists every failing check, uptime included, and connect_source can add a check. See the MCP server.
What it does not do
- No private addressesOnly public addresses are checked. A server only reachable inside your network is out of reach, except through your own health address.
- No redirects followedA redirect counts as an answer, but the check does not follow it to the next address.
- No pagingIt alerts by email and browser push. It does not phone you or run an on-call rota.
- One intervalEvery check runs on the same 15-minute schedule. There is no faster setting for one site.
For a site's speed and crawl health, try the free SEO audit. Weighing tools? See LopeBase vs UptimeRobot, Better Stack, and Instatus.
Uptime checks questions
Something else? Contact us.
How often does it check?
Every 15 minutes, with a year of uptime history. It is not a paging service, so keep one if you need a phone call at 3am.
Can it check a private address?
No. LopeBase only checks public addresses. To watch something private, have your app test it and report it through its LopeBase health address.
How long does a check wait?
Up to 10 seconds. A site that has not answered by then counts as down.
A site loads for me but LopeBase says it is down. Why?
Check the address you gave it. A page behind a login, a firewall that blocks LopeBase's requests, or an error code the browser hides all read as down. The outage list shows the status or error LopeBase got.
Does degraded count against my uptime?
No. Degraded still counts as up in the percentages, but it shows amber and labeled apart from a real outage on the day strip, the status page, and the alert.
Does it ever ping a vendor's status page?
No. It requests only the addresses you give it. Services behind a health address are tested with your app's own credentials.
Put uptime next to everything else.
Free for one property. No card.