Subject: Re: [vserver] Moving away from VServer
From: Herbert Poetzl <herbert@13thfloor.at>
Date: Sun, 13 Jun 2021 14:03:22 +0200

On Fri, Jun 11, 2021 at 06:58:26PM +0200, Romain Rivière wrote:
> Hi there,

Hello Romain,

> Until now, I had managed to maintain a couple of ebuilds in
> order to keep my linux-vserver hosts running with Gentoo as 
> the base distro.

Sounds great!

> However, it seems like this is coming to an end too. 

> Building dietlibc with gcc 10 results in a segmentation fault. 

During build time or when you use the dietlibc with some test
code or in the linux-vserver tools? Can you elaborate on the
segfault and maybe paste some debug info here?

> While it still builds nicely with gcc 9, I feel like this is 
> the curtain call for me. 

> I am probably going to move to Proxmox VE and LXC/KVM QEMU in
> the near future. 

LXC/KVM has gotten pretty stable and also gained a bunch of
features over the years which make them very suitable for easy
isolation as well as full system containerization.

So as long as your setup doesn't depend on Linux-VServer 
commands or isolation features, you should be pretty fine there.

> I really wish I could keep on using LVS, but the hassle of
> maintaining an older kernel has somewhat lost of its appeal :)

Sadly the interest in Linux-VServer has declined over the years
and it seems, that everybody would like to see a patch against
the most recent kernel but nobody wants to spend the time or
money on it to make it happen.

> I would like to thank the whole team for their effort through
> all these years. 

I think I can speak for the team here and say: we are happy to 
hear that and we certainly enjoyed the journey with you over all 
the years.

> In addition to providing us with a great tool and feature set,
> I have learned much by using Linux VServer. 

> I'll keep an eye on things in case something happens though ;)

\o/

All the best,
Herbert

> Cheers!
> -- 
> Romain