Supabase favicon

Supabase outage on 2026-10-31

supabase.com
October 2026 31 Minor incident

On 31 October 2026, Supabase's PG15 to PG17 database upgrades started failing for 1h 10m. The fault sat in a bad AMI release Supabase shipped that day, not in anything you did to your own database.

Started

20:06 UTC

Duration

Lasted 1h 10m

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?

The incident started at 20:06 UTC when Supabase reported active investigation into failed PG15 to PG17 upgrades. Within the same update, Supabase said it had identified an error in an AMI release from that day and was rolling it back to unblock upgrades. This was a single incident, not a cluster of separate faults, and it did not affect a listed set of components on Supabase's status page, since none are named in this record. By 20:42 UTC, Supabase reported that upgrades were succeeding again and that error rates had returned to normal levels, while it continued monitoring for stability. The incident was marked resolved at 21:14 UTC, giving a total duration of 1h 10m from first report to close. Supabase published three updates across the window: identified, monitoring, and resolved, and no further detail on the AMI error itself was given beyond naming it as the cause. No user reports were filed during this window, so there is no independent account of impact to weigh against Supabase's own timeline. For anyone running an upgrade from PG15 to PG17 during that hour, the practical effect would have been a failed or stuck upgrade attempt rather than downtime on an already-running database. Over the past 90 days, Supabase has logged 51 incidents with a median duration of 2 hours 5 minutes, which makes this one shorter than typical. Across all time, Supabase has recorded 474 incidents, averaging 7.5 per month.

Learning

This incident lived entirely in Supabase's upgrade pipeline, a backend process triggered when you ask to move a database from PG15 to PG17, and that kind of failure would not show up on an external check of your own running endpoints. If your product was pointed at a database that was mid-upgrade, your own monitoring of your application layer could stay green the whole time, since the request path your users hit was unaffected until the upgrade either completed or stalled. That is the shape of this kind of fault: it blocks a specific action rather than taking down a running service, so uptime checks built around your live endpoints simply have nothing to catch. Over the past 30 days, UptimeRobot's own checks against Supabase recorded 17 incidents, all resolved, with a median duration of 2h 13m, a cadence that lines up with how often Supabase itself reports trouble. Watching Supabase's status page or your own stack in isolation only gives you half the picture of where a fault like this sits. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.

4.7
stars out of 5
284+ reviews on

Start monitoring in 30 seconds.

There's nothing to install. No credit card required. 50 monitors for free.