System Status

Connection Logging feature is working fine for all customers now.

We are actively correcting an issue with the Management API: • Postgres Config changes only save the latest values sent, past changes are not being saved For now please refrain from not making any Postgres Config changes via the [Management API](https://.com/docs/reference/api/introduction) or in the [Database Settings Dashboard](https://.com/dashboard/project/\_/database/settings)

This issue is now resolved.

We are actively investigating some reports of failed PG15 -> PG17 upgrades. We have identified an error in an AMI release from today and are rolling it back to unblock upgrades.

We have deployed the fix for degraded logs, and smoke tests completed successfully at 12:20 UTC. Service is now operating normally. We apologise for the disruption

We are currently investigating degraded Logs functionality in multiple regions. Some users may experience delayed, missing, or unavailable logs in the Dashboard and Logs API. Other services, including your database and APIs, are not affected at this time. Our team is actively investigating. We'll post another update as soon as we have one.

This issue has been resolved and our response times are back to operational.

We have noticed issues with requests coming to our API gateway. We are investigating.

Error rates and latency have remained at normal levels following the mitigation. This incident is now resolved.

We are investigating elevated error rates and increased response times on the Management API across multiple regions. Some operations that depend on it, such as creating new projects, may fail or be slow.

Error rates have remained at normal levels following the mitigation. This incident is now resolved.

We identified increased error rates in project lifecycle operations (create, restore, pauses, etc.) in the EU region since 21 Oct 23:00 UTC. A mitigation has been put in place and error rates have reduced to normal levels. We are monitoring the situation and are working on fixing affected projects.

Between 22:10 and 22:51 UTC on October 9, 2026, about 1% of new database connections made through the Shared Pooler (connection pooler) in us-east-1 failed with "Failed to connect to database" timeout errors. The cause of the failures was an incident at our cloud provider which affected network traffic between availability zones. Project databases were not affected, and connections that had already been open prior to the incident remained open. The issue was resolved at 22:51 UTC. Clients that retried on connection errors would have reconnected successfully.

Capture was restored at 12:51Z on 26 October and has written continuously since; affected prices are current again.

The capture path stopped writing sports prices while its jobs kept reporting success, so the stop went unalerted. Other bookmakers and racing were unaffected.

Sports prices for the affected bookmakers stopped updating; events were serving their last captured price.

Realtime has fully recovered.

We are experiencing increased latency with Realtime and are actively investigating.

Edge function deployments have remained stable. We are closing the incident.

We are experiencing significant degraded performance across edge functions, mainly affecting deploy and retrieving function body. We have identified the root cause and are working on the fix now.

The database is reachable again and the API is serving normally. A read failure now returns 503 with retry_after instead of 401.

The API could not reach its database. Requests are failing with 401 and Service temporarily unavailable.

Between 05:50–06:05 UTC some customer using connection pooler in us-east-1 may have experienced brief dropped connections or connection disruptions. This issue was caused by an unexpected restart of connection pooler instances in the affected cluster. We apologize for any disruption caused.

We observed latency issues over the weekend resulting in 504 errors primarily impacting customers on the Free tier. The incident has been resolved.

This incident has been resolved.

We are investigating reports of increased Edge Function latency and elevated rates of HTTP 503 responses, beginning at approximately 06:00 UTC. Our team is working to identify the cause and will provide further updates as they become available.

This incident has been resolved.

Projects running Postgres version 14 in us-east-1 and ap-northeast-2 may experience query slowness. We've identified the cause and are working on a fix.

This issue has been resolved and our error rates have returned back to normal. During this incident users would have observed unauthorized/5xx failures across multiple operations on the project operations and dashboard functions. We have since rolled out a fix and all functionality is restored.

We are investigating an increase in errors on project update flows including creation, pausing/unpausing and upgrading across multiple regions.

A fix for this issue has been implemented and our data shows previous impacted users have recovered.

We have identified an issue with increased latency for clients in the eastern US during spikes caused by bursty traffic. This is most noticeable during EDT working hours around the :00 and :30 hour marks. We are actively working on a fix. We will provide updates as the fix is implemented.

This issue has been resolved and our error rates have returned back to normal. During this incident users would have observed unauthorized/5xx failures across multiple operations on the project operations and dashboard functions. We have since rolled out a fix and all functionality is restored.

We are investigating an increase in errors on project update flows including creation, pausing/unpausing and upgrading across multiple regions.