a misconfiguration job caused a memory limit issue for one of the services, resulting in instability and impacting autoscaling. The issue was resolved by 12:22 UTC after we reverted the job and corrected the memory configuration. The clusters are now back online and connected.