Supabase's project lifecycle operations failed in all regions for 1h 16m on 21 October 2026. Project creation, pausing and restoration jobs broke, and the fault sat on Supabase's side, not yours.
Started
15:36 UTC
Duration
Lasted 1h 16m
Source
IsDown
Next time
Supabase is degraded. Is your product working?
An incident on their side does not always mean an outage on yours, and the only way to know is to already be watching your own endpoints. Set that up in under a minute and you will be notified the next time your own site is affected. And there will be a next time.
50 monitors for your own websites. No card.
What happened?
Supabase logged one incident on this day. It started at 15:36 UTC, when Supabase's status page reported an issue affecting project lifecycle operations such as project creation, pausing and restoration, in all regions. No components list was published alongside this incident, so there is no breakdown of which services sat behind the failing jobs. At 16:12 UTC, Supabase posted that a fix had been applied and that they had observed a reduction in project lifecycle job failures, while continuing to monitor recovery. The incident was marked resolved at 16:51 UTC, with Supabase stating that operations had returned to normal. That gives a total duration of 1h 16m, shorter than the 90 day median of 2 hours 5 minutes for this provider. Supabase's status page is the only record of this incident. No user reports were filed during the window. Supabase's own write-up does not explain what caused the lifecycle jobs to fail, only that a fix reduced the failure rate and that recovery was then confirmed. Over the past 90 days, Supabase has logged 54 incidents, and over all time the rate works out to 7.5 per month. This incident affected project lifecycle operations specifically, not the data plane serving existing projects, so anything already running was not necessarily touched.
Learning
This incident sat in the control plane, the layer that handles creating, pausing and restoring projects, rather than in the layer serving requests from projects already running. If your own uptime checks only ping a live project's API or database endpoint, they would have stayed green through this entire window, because that endpoint never stopped working. Meanwhile anyone trying to spin up a new project, pause one, or restore one during those 76 minutes would have hit failures with no visible cause on their end. This is the general shape of control plane failures at any provider: the thing you built stays up while the thing you need to manage it goes down. Over the last 30 days UptimeRobot recorded 19 incidents for Supabase with a median duration of 1h 44m, all resolved, which gives a sense of how often this layer slips even when the data plane holds steady. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.