We noticed that multiple RENDER instances are currently unreachable.
Some may still appear as booted and operating system is functional, but cannot be reached/pinged.
It also impacts newly created/started RENDER instances, most of the time the boot fails and instances switch back to OFF state.
Issue is currently under investigation, we will update this status as soon as possible.
15/10/20 1530Z (1730LT)
Following a DHCP configuration issue, the GPU instances met some troubles with their public/private IP, and could have been stuck during the boot process (and/or then automatically archived). Running VMs should have recovered their network now.
15/10/20 1100Z (1300LT)
Product team is currently investigating the issue, awaiting further details.