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

Re: libfoo package depend on libfoog package?

On Fri, 12 Sep 1997, Enrique Zanardi wrote:

> I've been thinking about that, and have some worries about our
> proposed upgrading path. Imagine the following scenario:
> - a user has a 'bo' system installed
> - he decides to upgrade one of his libs, say libelf0, to the latest
> version, found in hamm (it may happen next year, after 'hamm' has been
> released as 'stable', this is not a "hamm is unstable, what do you 
> expected?" case)
> - the new libelf0 package places libelf.so.0.6.4 under /usr/lib/libc5-compat,
> but the old ld.so in 'bo' doesn't know about that path, does it?
> Bang! The user will get errors when trying to run his old bo applications, 
> linked with libelf0, until he upgrades ldso or includes /usr/lib/libc5-compat
> in /etc/ld.so.conf . Am I missing something?
> So, perhaps packages that place libs under /usr/lib/libc5-compat should 
> depend on the new ld.so

Perhaps.  However, we have provided sufficient documentation on the
process of installing new stuff from hamm, if you haven't read it already,
check out <http://www.gate.net/~storm/FAQ/libc5-libc6-Mini-HOWTO.html>,
which is a description of the minimum packages to install to get stuff
from hamm to work.  This includes the new ld.so.  I post a text version of
it here about once a week, and it is in our archive as well.  It seems
like a lot of work to add dependancys for everything that can be fixed
by RTFMing.

                       |       The bond that links your true family is not
    Scott K. Ellis     |      one of blood, but of respect and joy in each
    storm@gate.net     |         other's life. Rarely do members of one
                       |           family grow up under the same roof.
                       |                      -- Illusions

TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-devel-request@lists.debian.org . 
Trouble?  e-mail to templin@bucknell.edu .

Reply to: