On February 23rd, we received an unexpectedly large volume of ongoing monitoring updates, triggering a significant number of rechecks. To manage the task backlog, we scaled up our infrastructure. The backlog was cleared within an hour. We are continually working to improve automation performance to prevent such interruptions in the future.