Database failover in US East

Updates

1. Resolved

   Aug 30, 04:13 UTC

   Everything is back to normal. Customers in US East saw errors for 22 minutes. Other regions weren't affected.

2. Monitoring

   Aug 30, 04:04 UTC

   Failover is complete and traffic is served by the new primary. Some requests may still be slow while caches warm up.

3. Identified

   Aug 30, 03:55 UTC

   The primary database in US East stopped responding. Automatic failover to the replica has started.

4. Investigating

   Aug 30, 03:51 UTC

   API requests and the dashboard are returning errors for customers in US East. We're investigating a database issue.