Rebooting an instance that is locked up

Question

One of my instances is locked up because of a major memory leak. I want to reboot it, but the only way to do so preserving scratch disk data is to SSH into it. When doing so, no commands can be executed because of a lack of memory.

Is the only way of resolving this to delete the instance and provision another? I can't imagine that being the case, really.

I've looked through the documentation, and it states SSH is the only option, so I'm curious to hear if anyone reckons otherwise.

Answer 1

We don't support explicit reboots from the outside/API yet.

This is something that we know we need to support but I don't have a date or schedule for it yet. If you are using persistent disk to store your data (which you probably should for anything important) you can terminate and restart the instance.