We are experiencing performance degradation and higher than usual error rates on the API and the web interface.
As we are currently making changes to our infrastructure we initially thought it was the cause and rolled it back, but the service is still degraded and it does not seem to be related to our migration.
We are investigating the source further.