Both disks have come back online and do not appear to be unhealthy at first glance. We've applied a kernel and various package updates and rebooted the system once more, we will continue to monitor the system for the next several hours. The RAID array is currently re-synchronizing and may affect disk I/O for up to the next 30 minutes.
Posted Jul 19, 2020 - 21:17 UTC
Our monitoring systems have alerted us of a potential disk failure of one of the NVMe SSDs in Los Angeles node 2. In order to ensure the redundancy of the RAID array and stability of the system, we'll be taking the system offline briefly for a reboot to confirm failure. We'll update this posting as soon as more details are available.
Posted Jul 19, 2020 - 20:59 UTC
This incident affected: Prisma - Game Hosting Platform (Los Angeles, CA, USA).