Re: [AMBER-Developers] sander.MPI tests randomly stall in Ubuntu 9.10

From: Lachele Foley <lfoley.ccrc.uga.edu>
Date: Wed, 12 May 2010 14:03:40 -0400

Actually... it's the other compiler. But, not for lack of trying.

I totally did not expect this...

Here, the user has gcc aliased at login:

[installer.aarya src]$ which gcc
gcc: aliased to /usr/bin/gcc44
[installer.aarya src]$ gcc --version
gcc44 (GCC) 4.4.0 20090514 (Red Hat 4.4.0-6)

...but the config is finding something different...

========================================================
[installer.aarya src]$ ./configure gnu
Your AMBERHOME environment variable should be set to /usr/local/programs64/amber11
 

Obtaining the gnu suite version:
      gcc -v
The version is 4.1.2
========================================================

GAAAAUUUGHHHHH!!!!
Is it because the script is in sh?


It obviously pays attention to the .cshrc file, because when I made a directory $HOME/bin, set that first in the path, with sym links inside to the various places, it found the right compilers.

sh doesn't recognize the term "alias"...

:-) Lachele 
--
B. Lachele Foley, PhD '92,'02
Assistant Research Scientist
Complex Carbohydrate Research Center, UGA
706-542-0263
lfoley.ccrc.uga.edu
----- Original Message -----
From: Volodymyr Babin
[mailto:vbabin.ncsu.edu]
To: AMBER Developers Mailing List
[mailto:amber-developers.ambermd.org]
Sent: Wed, 12 May 2010 13:15:12
-0400
Subject: Re: [AMBER-Developers] sander.MPI tests randomly stall in
Ubuntu 9.10
> Hi Lachele,
> 
> thank you. It might be indeed something else. If it fails again with
> 4.4.0, please give me the outputs [or .diff(s)] to look at. I am not
> sure to what extent this can be relevant, but 'man alias' says
> 
> ...
> the alias definition shall not affect the parent process of the current
> shell nor any utility environment invoked by the shell; see Shell Exe&#8208;
> cution Environment
> ...
> 
> and it thus may be worthwhile to try making 'gcc->gcc44' symlinks
> somewhere and then add this somewhere to the front of the PATH.
> 
> Thank you & best regards,
> 
> Volodymyr
> 
> On Wed, May 12, 2010 12:03, Lachele Foley wrote:
> > 4.4.0
> >
> > I'll have her re-do from a clean un-tar (even if she did this time), just
> > to be sure.  I checked her paths.  She has gcc aliased to /usr/bin/gcc44
> > and so forth, and gcc --version yields 4.4.0, etc.  The MPI execs all just
> > point to "gcc", so they should go wherever her gcc points.
> >
> > It might not be RHEL.  Could be the machine.  Has an AMD processor... any
> > chance that matters?
> >
> > :-) Lachele
> > --
> > B. Lachele Foley, PhD '92,'02
> > Assistant Research Scientist
> > Complex Carbohydrate Research Center, UGA
> > 706-542-0263
> > lfoley.ccrc.uga.edu
> >
> >
> > ----- Original Message -----
> > From: case
> > [mailto:case.biomaps.rutgers.edu]
> > To: AMBER Developers Mailing List
> > [mailto:amber-developers.ambermd.org]
> > Sent: Wed, 12 May 2010 11:49:33
> > -0400
> > Subject: Re: [AMBER-Developers] sander.MPI tests randomly stall in
> > Ubuntu 9.10
> >
> >
> >> On Wed, May 12, 2010, Lachele Foley wrote:
> >>
> >> >
> >> > The failures are all from the ncsu set:
> >>
> >> Is this with gcc 4.1.2 or gcc 4.4.0?  The ncsu errors with 4.1.2 are
> >> well
> >> known (and will probably never be fixed); if you had errors with 4.4.0,
> >> it would be of interest to know what they are.  I've used gcc44 on
> >> Ubuntu
> >> and on CentOS with no such failures, fwiw.
> >>
> >> ....thx...dave
> >>
> >>
> >> _______________________________________________
> >> AMBER-Developers mailing list
> >> AMBER-Developers.ambermd.org
> >> http://lists.ambermd.org/mailman/listinfo/amber-developers
> >>
> >
> > _______________________________________________
> > AMBER-Developers mailing list
> > AMBER-Developers.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber-developers
> >
> 
> 
> _______________________________________________
> AMBER-Developers mailing list
> AMBER-Developers.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber-developers
> 
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Wed May 12 2010 - 11:30:03 PDT
Custom Search