We are looking into the issue now.
Performance has been restored. Reviewing the logs, we found a share that surpassed our rate limits causing a cascading CPU issue on the http frontend. We will be tunning those rules to prevent this issue for reoccurring.
We increased resources & restarted the services. System performance is restoring now. We will continue to monitor.
Our http frontend is reporting 100% CPU utilization & autoscaling was unable to keep up. Still looking for the initial cause.
