[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: native vs emulated configuration



On Sat, 9 Jul 2011 00:12:43 +0200
Hector Oron <hector.oron@gmail.com> wrote:

> 2011/7/8 Lisandro Damián Nicanor Pérez Meyer <perezmeyer@gmail.com>:
> > On Vie 08 Jul 2011 15:27:17 Neil Williams escribió:
> > [snip]
> >> > By going step by step through the things that polystrap does, let me
> >> > list what would have to change in multistrap to make it build foreign
> >>
> >> > rootfs without superuser priviliges:
> >> Multistrap already does all of this, the only thing for which some
> >> people might want emulation is configuration. That's device-specific.
> >> There's no reason to fold that into multistrap.
> >
> > Then why is multistrap installed under /usr/sbin?
> 
> That's good question, but if you do not run multistrap as root, then
> the system is unusable (partially because effective UID does not match
> real UID).

i.e. it is in /usr/sbin because the default is that root access
(usually under sudo) is necessary. Fakeroot exists exactly for this
situation and multistrap can work under fakeroot, where that is
desirable. It's not entirely that simple because fakeroot has
complications about running multiple commands but there are Makefile
snippets out there which will support this.

The point is that there is no need for a separate wrapper just to run
multistrap without root access - that program already exists and it's
called fakeroot. Whichever way you do it, multistrap and
other tools like debootstrap cannot live in /usr/bin.

This "polystrap" wrapper uses fakeroot / fakechroot which do the same
things.

> http://wiki.debian.org/EmDebian/CrossDebootstrap#Generating_cross_images_as_non-root_user
> 
> There are two different ways already, with schroot or with fakechroot

Exactly - but some form of root access, either real or fake, is
necessary for any kind of rootfs processing.

-- 


Neil Williams
=============
http://www.linux.codehelp.co.uk/

Attachment: pgpaztPbL4WwY.pgp
Description: PGP signature


Reply to: