Previous incidents

A long-running database migration casing a timeout. Kubernetes was retrying the to run the migration, which (unexpectedly) made the Dashboard offline.