platform status · independent triage guide

Is Lovable down — or is it just your app?

The indicator below is pulled live from status.lovable.dev. If Lovable reports an incident, the editor and previews are affected platform-wide. But your published app mostly doesn't run on Lovable's editor — it runs on your hosting and your Supabase backend. A paused Supabase project, a failing edge function, or a rotated API key takes your app down while Lovable stays perfectly green. Check the platform first, then your own stack.

Checking Lovable's official status… live from status.lovable.dev

Nightlamp is not affiliated with Lovable and this is not the official status page. For platform incidents, the source of truth is status.lovable.dev. This page exists for the other half of the question: what to do when the platform is green and your app still isn't working.

what their status page covers

What status.lovable.dev does tell you

  • The Lovable editor and AI generation pipeline
  • Preview environments
  • Lovable-managed publishing/hosting incidents
what it can't tell you

Lovable can be green while your app is down

Your Supabase project paused itself

Free-tier Supabase projects pause after about a week of inactivity. The moment that happens, login, signup, and every database read in your app fails — and neither Lovable's nor your app's hosting reports anything wrong.

An edge function is throwing errors

The AI wrote a Supabase edge function for emails, payments, or webhooks; a dependency update or a malformed payload now makes it 500. The page renders fine, the action silently fails.

An API key was rotated or hit its quota

OpenAI, Stripe, or Resend keys living in your project's secrets expire, get rotated, or run out of quota. The app deploys cleanly and still can't do the one thing users came for.

A schema change broke your RLS policies

After a database migration, row-level-security policies can start rejecting legitimate reads. Users see empty screens or infinite spinners; every status page involved stays green.

five-minute triage

Platform problem or your problem? Run this checklist

  1. Check status.lovable.dev — editor or publishing incidents are platform-wide.
  2. Open your published app in a private window: does it load at all?
  3. Check your Supabase dashboard: is the project active (not paused), and are there errors in the logs?
  4. Test login/signup specifically — auth depends on Supabase and email delivery, the two most fragile links.
  5. Check edge function logs for recent failures.
  6. Verify the API keys your app calls (OpenAI, Stripe, Resend) are valid and within quota.

If sign-in emails are part of the broken flow, our free magic-link tester checks the full email round-trip in one click.

Checklist points at your own app? Our step-by-step guide to fixing a broken Lovable app walks through each failure symptom by symptom.

Stop checking status pages after the fact

Nightlamp watches your Lovable app's actual flows — checkout, signup, login emails, scheduled jobs — and a real engineer sends a plain-English diagnosis when one breaks, whether or not the platform is having an incident.

Start 14-day trial · no card
faq

Frequently asked questions

Lovable's status page is green but my app is down. Why?

Your published app runs on your hosting and your Supabase backend, not on Lovable's editor. Paused Supabase projects, failing edge functions, rotated API keys, and broken RLS policies are the usual culprits — none of them appear on status.lovable.dev.

Why did login stop working on my Lovable app when nothing changed?

The most common cause is a free-tier Supabase project auto-pausing after inactivity, which kills auth and database access at once. The second most common is auth emails (magic links, confirmations) failing to deliver. Both look like “the app is down” to users.

How do I get alerted when my Lovable app breaks but Lovable is up?

Monitor the app's real flows — login, signup emails, the edge functions behind your key features — rather than the platform. Nightlamp probes those continuously and a real engineer tells you in plain English what broke and how to fix it.

Newsletter

One real no-code incident + the fix, monthly

A post-mortem from a real no-code outage — what broke, how it was found, and the fix — once a month. Confirming also gets you the no-code incident runbook pack.

Double opt-in. One-click unsubscribe. No spam, ever.