Service recovery is complete, and the cluster has been scaled up. We hopefully don't expect this issue to recur after this.

Apologies again for all the inconvenience!

Additional capacity has been added and we've rebooted the service. Service recovery is underway and is expected to complete momentarily.

We're investigating similar crash as yesterday, and are restarting the service. Additionally, we are adding more capacity to the service, as we think the crash is load-based.