reptyr VS criu

Maybe https://github.com/nelhage/reptyr

https://github.com/nelhage/reptyr I guess this is what your looking for (it reparents a process, while retaining stdin/out using the ptrace syscall and some linux hacks iiuc)

Start CPython and import the libraries. Then keep the process running as a fork server. Whenever a new instance it needed, we make a fork (os.fork), and apply a similar logic as reptyr. Some technical details are here.

Very much reminds me of reptyr

- reptyr: https://github.com/nelhage/reptyr

using mosh means i don't get SIGHUP, and if i use my local display, i can open another mosh, reptr my old session (it's still there!) and keep on hacking.

I know there is reptyr to reattaching running process to another tty. You can probably do the same from within the program. There is an abandoned rust rewrite you can probably ignore.

Alternatively, there's also reptyr, which allows you to reattach to a detached process - thereby giving you access to the standard input and standard output of a running Minecraft server instance.

Checkpoint/Restore I feel is a bigger concept than just saving state. At the zeroth level it's a system that can correctly stop and serialize a running process (as criu https://github.com/checkpoint-restore/criu has shown is a huge pain in the ass to still not be perfect) in a way that can initiated from within the process itself.

The 1st level more-work-but-easier way to do this is to build or use a heavily constrained VM/language you run from within your main application that doesn't allow for most of the hard problems to even exist.

I can't find any ready-made tools to do this that I wouldn't consider an endeavor.

That being said the Checkpoint Restore In Userspace project has been around for a number of years and is the closest thing to what you are talking about: taking a linux process on one machine and moving it to another. It is messy but can be done in some cases. There are folks looking at how to integrate CRIU with k8s but it’s all research at this point.

I would love to make more use of CRIU. E.g. I considered to use CRIU for my Python preloaded logic (https://github.com/albertz/python-preloaded). Unfortunately, at that point in time, CRIU must be used with root access, which was not an option. However, I see that the PR was merged now, so maybe it works now? (https://github.com/checkpoint-restore/criu/pull/1930)

There is also DMTCP (https://github.com/dmtcp/dmtcp/) but this might have other problems for my use case.

My solution was to use a fork server instead, which works almost equally well. There are not really much downsides with this approach. And this is actually quite simple, and also quite cross-platform (except Windows).

CRIU currently needs root access for dump/restore. However, there is ongoing work to support a non-root option in https://github.com/checkpoint-restore/criu/pull/1930.

There have been multiple checkpointing attempts over the years. Criu is the only one I know of that's still kicking. That's probably your best and only bet.