Please add kexec-reboot functionality

[verulian]

I'm running Proxmox on a Dell PowerEdge server that takes several minutes to reboot. It's horrible, but I've found that using https://github.com/error10/kexec-reboot works AMAZINGLY well!!!

I don't know the "proper" way to install it, but apt installed the kexec-tools and Ruby from Debian and then git cloned the repo above into ~/src and then symlinked that script to ~/bin and also symlinked reboot to it for my user. Now I rebooting takes just 30-45 seconds versus several minutes.

This feature would be invaluable for everyone since Proxmox so frequently runs on server hardware.

 

[LnxBil]

Ehh ... doesn't this already work in Debian for many years .... and yes it did. I just tried it and it does not work anymore. Fortunately, everything is already in place (as you wrote), but it is a problem with the kexec-tools. I found this forum entry describing the exact problem: http://forums.debian.net/viewtopic.php?f=10&t=137346

If you just load the kernel manually, it works like it should be:

root@pve6 ~ > kexec --load /boot/vmlinuz-5.0.18-1-pve --initrd /boot/initrd.img-5.0.18-1-pve --reuse-cmdline

Next would be to test if Debian Buster is also not working and if so, we should get kexec-tools fixed upstream so that it just works.