Re: [gentoo-user] separate / and /usr to require initramfs 2013-11-01

[Alan Mackenzie]

'evening, Neil.

On Sat, Sep 28, 2013 at 09:17:02PM +0100, Neil Bothwick wrote:
> On Sat, 28 Sep 2013 19:04:41 +0000, Alan Mackenzie wrote:

> > > I suppose that what I am about to say isn't really relevant, but it is
> > > unfortunate over the past year that people blamed udev specifically
> > > for this. It is true that it does things that don't work if /usr isn't
> > > mounted, but eudev does as well, since it is basically the same code.

> > Who else is there to blame?  We are continually being told that a
> > separate /usr is "broken", as though this were some unfortunate act of
> > <insert your deity here>, much like an earthquake.  This gets
> > patronising really quickly.  (Please note, I'm NOT blaming you here.  I
> > appreciate that you're as much victim as Dale or me or anyone else
> > round here.)

> It's evolution. Linux has for years been moving in this direction, now it
> has reached the point where the Gentoo devs can no longer devote the
> increasing time needed to support what has now become an edge case.

That's precisely the sort of patronising comment I was complaining of in
my previous paragraph.  It isn't "evolution".  It has been a decision of
somebody to move it.  Who?

> > No, this breaking of separate /usr was done by some specific project,
> > some specific person, even, in a supreme display of incompetence,
> > malice, or arrogance.  How come this project and this person have
> > managed to maintain such a low profile?  There seems to have been some
> > sort of conspiracy to do this breakage in secret, each member of the
> > coven pushing the plot until the damage was irrevocable.  Who was it?

> So which was it, one specific person or a coven of conspirators? This is
> open source, secret conspiracies don't really work well. If this really
> was such a bad move, do you really think the likes of Greg K-H would not
> have stepped in? Or is he a conspirator too?

I know not how many people were involved.  Don't you think it noteworthy
that we on this group first learnt of the change when it had already
happened?  I have no idea whether people like GK-H would have been aware
of it either.

> > > If you read flameeyes' blog post, you will get a better idea of what
> > > the issue involves. It is the entire boot process and how to deal
> > > with which software is considered critical for booting.

> > > There is no reason to rebuild your server; we aren't telling you you
> > > have to merge /usr into /. The only thing we are saying is that you
> > > will need to use an initramfs if you are going to keep them separate.

> > "Only"?  ONLY???  You say that as though creating an initramfs were a
> > trifle, trivial, and of no moment.

> For an Ubuntu user, maybe that's true? For someone that feels comfortable
> compiling their own kernel and configuring the entire system by hand,
> running dracut or genkernel should not be too demanding. Even creating
> your own initramfs is hardly rocket science.

It may or may not be demanding for any particular administrator.  It is
undoubtedly tedious and time consuming.  Installing RAID and LVM2 were
(for me) also time consuming, but at least I got something worthwhile out
of them in the end.  Creating an initramfs is a lot of work just to end
up in the same place.

> > Ah yes, the deficiencies of the kernel.  It can only mount one file
> > system when it starts.  It's incapable of mounting LVM2 systems (even
> > though it contains LVM2 code).  It's incapable of mounting encrypted
> > partitions (even though it contains encryption code), ......  So because
> > of these holes, a system must either be constrained in it's makeup (as
> > mine is) or use an ugly hack.  It can (still) mount RAID partitions, I
> > suppose.

> That's plain wrong. The kernel doesn't include LVM code, only the
> device mapper functions that LVM uses, It does include RAID code.

> > > I have a pretty simple setup, but I have been using an initramfs
> > > which I built some time ago with genkernel and I barely know it is
> > > there.

> > Until, after some update, it reminds you of its presence by not booting
> > your machine.  That's the sort of excitement I can do without.

> Do you have any examples of this actually happening? Not "I heard a bloke
> down the pub talking about a mate of a mate who broke his system with an
> initramfs" but actual documented examples of how this can occur in normal
> use.

There have been several times in the past few years when precisely that
could have happened in Gentoo - the updating of the Baselayout in 2011,
the various shenanigans with udev, for example.  Dale's former system
broke because of an initrd.  I get nervous every time something like lvm
get updated.

> --
> Neil Bothwick

-- 
Alan Mackenzie (Nuremberg, Germany).