splitting courier-authlib into master+slave ports
ports at dino.sk
Wed Apr 20 21:24:58 PDT 2005
On Thursday 21 April 2005 01:48, Yarema wrote:
> --On Wednesday, April 20, 2005 21:44:11 +0200 Jose M Rodriguez
> >> One difference between the courier-authlib-20050408.00.tgz version
> >> and courier-authlib-20050420.00.tgz is that I make --with-authpam
> >> part of the base port's CONFIGURE_ARGS. This prevents libauthpwd.so.0
> >> from being built and instead builds
> >> lib/courier-authlib/libauthpam.so.0. authpwd is discouraged as per
> >> <http://www.courier-mta.org/authlib/README_authlib.html>:
> > We have a FreeBSD supported version without a pam library? I think no.
> Yes, we do have "a FreeBSD supported version without a pam library"
> installed if only the base port is installed. I made this happen to for
> the sake of completness and now I'm presenting arguments that it is a bad
> idea. Thing is that the courier-authlib port, as it is committed NOW, will
> install the no PAM version "libauthpwd.so.0" if NONE of the OPTIONS are
> selected. Yet the PLIST in the current version does not include
> See for yourself. Make sure that you have no WITH_ tunables in
> /etc/make.conf and unselect all the options in 'make config' then 'make
> install' and look in /usr/local/lib/courier-authlib/ -- you'll see that
> there's a "libauthpwd.so.0" in there. Then 'pkg_delete
> courier-authlib-0.55_1' and you'll get:
> pkg_delete: unable to completely remove directory
> pkg_delete: couldn't entirely delete package (perhaps the packing list is
> incorrectly specified?)
> This is how things are NOW. I noticed this when I did my initial rewrite
> and modified the base port PLIST to account for "libauthpwd.so.0". Now I'm
> making a case to do away with it all together by making --with-authpam
> nonoptional. I'm simply arguing to have something removed that noone
> except me noticed existed.
> It's either we have "libauthpwd.so.0" or the more modern "libauthpam.so.0"
> installed in the base port. Both essentially do the same thing by default.
> All the documentation I cited points to PAM being the better choice.
Actually I filled PR ports/79846 for this issue. In light shed by this
discussion simple solution offered there is not the best. It was my first
attempt to solve exactly the issue mentioned above.
Anyway, if possible, I would like the opportunity to get 'clean' base port,
with no auth method included if we are to go with slave ports carrying 'real'
auth methods. It is up to the installer/administrator to select
authentication method best suiting his/her needs in particular installation.
This way we can prevent some unexpected lateral effects, maybe.
More information about the freebsd-ports