Supabase had a significant outage starting June 30, 2026 at 22:23 UTC. It lasted about five days, recovering on July 6 at 16:37 UTC. Supabase has marked it resolved. The status note gives no further detail on what went wrong.
Who this kind of outage hits, and how they usually find out
If you build on Supabase, your app or your automations talk to it constantly. A database that goes quiet for five days is not a blip. It is a week of broken signups, failed reads, stalled workflows, and confused customers. Most operators find out the wrong way: a user emails to say something is broken, or a client asks why their report is empty, or a payment flow silently fails and nobody notices until the money does not arrive. The tool itself rarely sends you a plain message saying “I am having trouble.” You are expected to be watching.
Why this is especially rough if you are not an engineer
There is no error on your screen. Your n8n workflow does not send you a message saying it stopped. Your app just quietly does less than it should. The first signal is often a frustrated customer, and by the time you get that message the problem has been running for hours. You then have to figure out whether the issue is your code, your workflow logic, your hosting, or the tool underneath all of it. Without knowing Supabase was already having trouble, you can spend a long time looking in the wrong place.
Timeline
- June 30, 2026 at 22:23 UTC - Supabase’s own status page showed a bigger problem beginning.
- About five days of disruption - the problem persisted through the first week of July.
- July 6, 2026 at 16:37 UTC - Supabase marked the outage resolved.
How a watcher catches this before your users do
NoCrash reads Supabase’s public status page once every minute. The moment that page flips from working to having trouble, NoCrash sends you a plain-language message, in one place, alongside everything else you build on. You do not have to be watching the status page yourself. You do not have to wait for a customer to tell you.
It also watches the things you ship. If you have n8n workflows, NoCrash watches them through your API token. If you have an app, you can give it a URL or drop in a small JS snippet, and it watches that too. So if Supabase recovers but your own workflow is still stalled, that surfaces as well.
To be clear about what this is: NoCrash reads the tool’s own public report. It does not find the outage before Supabase reports it. What it does is make sure you hear about it within a minute of Supabase’s own report, in plain English, without you having to check anything yourself.
For the authoritative account of this outage, see Supabase’s own status page: https://stspg.io/v2kqtvng9qnt