[torqueusers] location of pam module

[Lloyd Brown]

On 12/14/10 7:22 PM, Garrick Staples wrote:
>
> What does configure say for "checking host system type..."?
>  
> Have you generated configure yourself?
>


I'm not quite sure what you mean by whether or not I generated configure
myself.  I did run autoconf in this case, at the instructions of
Adaptive, to increment the version number for the patched version. 
There weren't any special parameters to autoconf.  I just ran it in that
directory.  Admittedly I'm not very familiar with autoconf, so I'm not
entirely sure what happened.  We just changed the "AC_INIT" in
"configure.ac" to have the version number "2.4.7.8901patch", then ran
autoconf.  From there on, it was the same "configure", "make", "make
rpm" process I've always used.

Looking through the "configure.log", I find these lines; Is this the
info you're after?

> ## ----------- ##
> ## Core tests. ##
> ## ----------- ##
>
> configure:1726: checking build system type
> configure:1744: result: x86_64-unknown-linux-gnu
> configure:1752: checking host system type
> configure:1766: result: x86_64-unknown-linux-gnu



>
> So why do you disable rpp?
>

That's an old config parameter that I haven't changed since 2.2.x or
before.  I don't remember exactly why for certain, but it seemed like
there was some concern about the scalability for large clusters.  I
can't find that recommendation in the Torque manual anymore, though, so
I don't have anything to back it up.  It's very possible that this is a
bad idea.  Unless you tell me otherwise, though, I'm going to assume
that this is a I'm-just-curious kind of question, and not really germane
to the question at hand.  I'm willing to take feedback or have a
discussion about this, but unless it's related to the pam module's
location, lets do this in a separate list thread, or off-list if you'd
prefer.



-- 


Lloyd Brown
Systems Administrator
Fulton Supercomputing Lab
Brigham Young University
http://marylou.byu.edu