On June 4th at 8:15pm, we noticed an unusually burst of traffic to our servers causing our database CPU usage to peak to 100%. At 8:23pm this resulted in the CMS going down. At 8:37pm our database scaling kicked in and services started recovering.

On May 31st at 1:45 PM, we noticed an unusually burst of traffic to our servers causing their CPU usage to peak. At 2:45 PM this resulted in the CMS going down. We have blocked all traffic from external sources to reduce the strain on our servers and reduce CPU. At approximately 5:15 PM all servers had recovered and the CMS was back available.

We will continue to block traffic from external sources until the infrastructure team has pinpointed the root cause of the issue.