[Debichem-devel] Help compiling ACESIII v. 3.0.7

[Veals, Jeffrey]

Hi Michael,

Your suggestion about setting the ACES_EXE_PATH variable appears to have
done the trick.  I have done a few calculations for comparison with
Gaussian, and the individual energies of my test jobs agree very well with
the energies from gaussian ( at least for rhf and mp2 single point
energies).  So I think it is compiled correctly.  I need to try and get
the make file for the test suite up and going in order to make make sure
all other aspects of the program are working.

I did want to ask if you had to recompile the entire program when checking
the test suite,  or were you able to simply run make from the directory
with the test suite Makefile is located?

Jeffrey

On 11/1/12 12:18 PM, "Michael Banck" <mbanck at debian.org> wrote:

>Hi Jeffrey,
>
>On Thu, Nov 01, 2012 at 01:13:01AM +0000, Veals, Jeffrey wrote:
>> I recently downloaded ACES III 3.0.7 from the QTP web page.  I tried
>> some the suggestions from the dialogue between  Michael Banck and
>> Ajith Perera.  From here:
>> 
>> http://web.archiveorange.com/archive/v/304OrIsCcGNywDSkNHbr
> 
>> In particular I tried to model my GNUMakefile.chssi file to resemble
>> as closely as possible to the way Michael describes in his comment on
>> August 2. I have attached a portion of the make file as well as a
>> portion of the log file produced.
>> 
>> The problem appears to occur just as the test suite is about to begin.
>> I don't receive an error from make, but the test suite does not run.
>> When I try and use the executable that is produced(in the main/
>> directory) for the ZMAT input that is found in zmat_1.1.1.1, I receive
>> the following error in my job.out file:
>> 
>> ----------------------------------------------- Cartesian coordinates
>> corresponding to internal coordinate input (Angstroms)
>> ----------------------------------------------------------------
>> Z-matrix Atomic C o o r d i n a t e s Symbol Number X Y Z
>> ---------------------------------------------------------------- AR 18
>> -1.05835450 -1.05835450 0.00000000 AR 18 -1.05835450 1.05835450
>> 0.00000000 AR 18 1.05835450 -1.05835450 0.00000000 AR 18 1.05835450
>> 1.05835450 0.00000000
>> ----------------------------------------------------------------
>> Interatomic distance matrix (Angstroms) AR AR AR AR [ 1] [ 2] [ 3] [
>> 4] AR [ 1] 0.00000 AR [ 2] 2.11671 0.00000 AR [ 3] 2.11671 2.99348
>> 0.00000 AR [ 4] 2.99348 2.11671 2.11671 0.00000 @GEOPT-W, Archive file
>> not created for single-point calculation. Initial xjoda is complete...
>> Using Default jobflow = SCF_RHF_ENERGY Error: Cannot find jobflow
>> SCF_RHF_ENERGY in default_jobflows file.
> 
>I guess the problem is that the driver executable is not finding the
>SIAL files.  Try setting the ACES_EXE_PATH environment variable to
>where sial_config is, usually ./bin/sio from your top source directory.
>
>Also note that I had originally problems with the testsuite files, they
>were missing a trailing blank line, but maybe that is not an issue
>anymore.
>
>We patch the testsuite Makefile a bit to make it run properly, I am
>attaching the patch for your information.  Let me know if you have
>further trouble.
>
>
>Best regards,
>
>Michael