The issue affecting flows on the AWS EU (eu-central-1) stack has been resolved. An internal component responsible for advancing flows between phases was overloaded, which caused flows to remain in the processing state after their first phase. We have applied a fix and processed the backlog, and flows are running normally again.
Flows that were delayed have continued automatically and do not need to be re-run. If you manually terminated a flow during the incident, please start it again.
We are sorry for the inconvenience and are working on a permanent fix to prevent this from happening again.
