No components marked as affected
Resolved
During a database migration in production, an overly aggressive lock took over a table we use for authentication, so we could not serve any requests to customers between 16:20-16:50 UTC. We've now resolved the issue and are fully operational.