We believe that as of a few minutes ago the last customer impact from this issue has been resolved and all queues have cleared. To help you correlate this incident with any issues you may have seen: as far as we can tell the earliest customer impact from this incident started at about 11:00 UTC today.

We believe we have identified and updated the cluster autoscaler. The system is currently working through the backlog of requests.

We are identified an problem with our cluster autoscaler and are continuing to investigate.

We're aware of scheduling delays affecting some models and are investigating the issue.