Subject: Re: Philosophy of PAM and rc.d
To: Dustin Sallings <dustin@spy.net>
From: None <seebs@plethora.net>
List: current-users
Date: 03/18/1999 11:51:07
In message <Pine.SGI.3.95.990318094446.17190C-100000@bleu.west.spy.net>, Dustin
Sallings writes:
> How many applications support BSD auth out of the box, vs. support
>PAM?
I don't know. On the other hand, it's a very easy thing to add, and it's
not like we're targeting a lot of things we don't have source for.
>For PAM, many of the apps I use (Cyrus, xscreensaver, ssh, etc...)
>can be rebuilt with a different commandline option (maybe a patch) and
>magically work. Not so easy with BSD auth. The main reason I'm
>interested in PAM is so that I only have to solve any given problem once
>and have the most widespread benefit possible.
That's not very NetBSD. Traditionally, the NetBSD answer has been to evaluate
the proposed solutions on technical merit, and choose the best. I think BSD
Auth is a better design, but I haven't got enough experience with PAM to make
a very good judgement. The login class support is part of the BSD Auth
design; I'm not sure whether that's a good thing for everyone, but I sure like
it.
Essentially, if we want that feature, we're stuck altering a lot of software
anyway. :(
-s