With the help of AI, I created a persistence patch, build the Debian packages with mise, and new CLI commands for VM restarts, allow-list updates and arbitrary initial binary without going through sh (useful for nixos/nix images without rebuilding the image). https://github.com/leonardopedro/matchlock-patch-debian-restart
It seems there are some tradeoffs (some MBs more RAM needed per VM, for example) , so maybe a small part of it could be added to matchlock as an option at compile time, for example. In any case, choosing to apply or not apply the patch matchlock-patch-debian-restart is an effective option at compile time. Progressively, this could be integrated into matchlock as a compile-time option.
I do not know much about programming for many people, my experience is about programming for specific Physics calculations, so it is very different. matchlock-patch-debian-restart is similar, I did it for me, now I am sharing it as it is.
With the help of AI, I created a persistence patch, build the Debian packages with mise, and new CLI commands for VM restarts, allow-list updates and arbitrary initial binary without going through sh (useful for nixos/nix images without rebuilding the image). https://github.com/leonardopedro/matchlock-patch-debian-restart
It seems there are some tradeoffs (some MBs more RAM needed per VM, for example) , so maybe a small part of it could be added to matchlock as an option at compile time, for example. In any case, choosing to apply or not apply the patch matchlock-patch-debian-restart is an effective option at compile time. Progressively, this could be integrated into matchlock as a compile-time option.
I do not know much about programming for many people, my experience is about programming for specific Physics calculations, so it is very different. matchlock-patch-debian-restart is similar, I did it for me, now I am sharing it as it is.