Sheldon Hearn wrote: > On (2003/11/27 17:19), Bill Hacker wrote:
>
>
>>It worked this time. IF it failed last time, it should have aborted the
>>whole process, but apparently did not....
>
>
> I've never seen a port dependency failure ignored by the ports system.
> :-)
Depends on *how* it fails.
Problem is usually not with the retreival or build of the sources, but
rather with cp, mv, rm, mkdir, chown or chmod calls wherin the author
has either not checks for a possible conflict as a precondition or has
not checked errors returned.
>
> I'm still curious as to why you ended up with /etc/rc.d/exim.sh instead
> of /usr/local/etc/rc.d/exim (or even /etc/rc.d/exim).
You may have missed it. It *was* intalled in /usr/local/etc/rc.d. I
moved it to /etc/rc.d.
Historically we used to symlink /usr/local/etc/rc.d to /etc/rc.d
The out-of-the-source 4.9 doesn't have an /etc/rc.d, whereas the 5.1 has
it full of its *own* stuff, so we have quit that...
>
> But I'm starting to get that "local weirdness" feeling. :-)
>
Actually, I am trying to 'unweird' that weirdest of all surviving OS'es.
With all the games *BSD is playing with where they put *their* stuff,
from one release to another, I will probably create a root/mount-point
dir such as "/jfdi" - put its path into /etc/rc.conf, and start keeping
my startup/shutdown scripts in there.
