On 9/2/07, Randy McMurchy <[EMAIL PROTECTED]> wrote:
> Dan Nicholson wrote:
>
> > Randy, maybe we should make this a little more robust so the .so
> > versions aren't hardcoded (although, they're probably entities in the
> > book source). I've been using a loop with readlink in my script. I
> > think the following would work.
> >
> > mv -v /usr/lib/libpam*.so.0* /lib &&
> > for lib in libpam{,c,misc}.so; do
> > link=`readlink /usr/lib/$lib` &&
> > ln -v -sf ../../lib/$link /usr/lib/$lib
> > done
>
> I don't have access to the book sources right now, but if I
> recall correctly, they are not entities and hard-coded requiring
> us to properly update the ln commands.
>
> However, the book looks correct to me, so no telling how Alan's
> installation got hosed.
The book is correct right now. My guess is that Alan had himself
hardcoded the numbers on a previous build of pam.
> I do like your idea. If you're certain it is reliable, please
> update the book, or make a Trac ticket. This doesn't have
> anything to do with Alan's installation, however.
I'll test it out, but I've been something similar in my scripts (the
difference being that use --libdir=/lib and then move the .so link to
/usr/lib).
--
Dan
P.S. You'll be happy to know that I'm working on the cyrus-sasl update
right now. Only 15 months to take action :)
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
