[systemd-devel] cryptsetup vs. swapon/fsck (some kind of race condition)

[Lennart Poettering]

On Fri, 16.05.14 03:20, Andy Kittner (andy.kittner at gmail.com) wrote:

> >> | May 15 22:06:47 pinky systemd[1]: Mounted Runtime Directory.
> >> | May 15 22:06:47 pinky systemd[1]: Mounted Lock Directory.
> > 
> > This looks like remains from a *really* old systemd version. THis does not
> > exist anymore. What distribution is this?
>
> Gentoo

Oh, man. Gentoo is really getting on my nerves... 

That completly backwards /usr migration thing they are doing. The
completely misunderstood multib setup of Gentoo, and now these idea of
mknod from userspace on today's kernel.... Jezuz, guys. And Gentoo wants
to be taken seriously... --omg-optifucked...

I am tempted to just redirect all systemd questions coming from Gentoo
users directly back to tgen Gentoo ML. They need to clean up their own
mess first...

> > This really should be turned off in libdm. It's really stupid and
> > broken. We start these things in parallel, they create these races
> > without reason. Since ages we don't support non-devtmpfs kernels
> > anymore, [...]
> And that's fine for systemd, but it doesn't necessarily mean that other 
> projects work under the same assumption. And IMHO especially for 
> critical things like lvm it is a good thing to support minimal 
> configurations.

"Minimal" configurations? Are you kidding?

If you want something "minimal", then just use devtmpfs, and don't
involve userspace at all in creating device nodes. This is not about
being "minimal" or not, this is about actually getting basic things
right...

Anyway, systemd is not compatible with Gentoo's way of doing LVM/DM,
sorry. Please file a bug against Gentoo.

> That sounds like this can only end in a systemd vs. the rest of the 
> world flamewar and I have no intention to get involved in one of
> those,

Dunno, every distro actually appears to get this right these days, but
one...

Lennart

-- 
Lennart Poettering, Red Hat