Resolved
Issue has been resolved. And the platform now is stable.
Monitoring
Configuration changes have been deployed, we are seeing autoscaling return to normal. We continue to monitor the situation.
Identified
We have identified the root cause, and currently in the process of deploying configuration changes to restore automatic scaling.
Investigating
We have identified an issue which is preventing workers from scaling. This impacts apps that scale to zero, preventing them from starting as well as bursting new workers from the configured minimum workers. We are investigating the underlying causes.