At 03:31 PST we have received an alert from our monitoring system about slow responsiveness of node v7228
At 03:37 PST initial investigation showed kernel error messages on the console; the console was unresponsive
At 03:41 PST we managed to login via console and will be gracefully shutting down all VMs in order to perform emergency node reboot
At 03:58 PST all VMs have been gracefully shut down and we will be rebooting the node shortly.
At 04:08 PST Node successfully rebooted without showing any errors. All VMs will start automatically in the next 15 minutes.
We will be closing this incident for now.