AMBER Archive (2006)

Subject: Re: AMBER: compile amber8 on IBM-sp4

From: Rachel (comeonsos_at_googlemail.com)
Date: Thu Nov 23 2006 - 07:14:09 CST

Hi, Xuebin,

Thanks a lot for your reply, after I changed "sysdir='Machines/ibm_aix' " to
"sysdir='lib'', the installation processed without any errors, however, when
i started test by using:
 cd ../test
make clean
make

I got the following error messages:

####################################################

diffing prmtop.save with prmtop
possible FAILURE: check prmtop.dif
==============================================================
        cd antechamber/tp; ./Run.tp_mol2
Running: /hpcx/home/e39/e39/uq281/amber8/exe/bondtype -i
ANTECHAMBER_BOND_TYPE.A
C -o ANTECHAMBER_BOND_TYPE.AC -f ac -j full

Running: /hpcx/home/e39/e39/uq281/amber8/exe/atomtype -i ANTECHAMBER_AC.AC0
-o
A
NTECHAMBER_AC.AC -p gaff

Total number of electrons: 58; net charge: 0

Running: /hpcx/home/e39/e39/uq281/amber8/exe/divcon

Running: /hpcx/home/e39/e39/uq281/amber8/exe/am1bcc -i
ANTECHAMBER_AM1BCC_PRE.AC
-o ANTECHAMBER_AM1BCC.AC -f ac -p
/hpcx/home/e39/e39/uq281/amber8/dat/antechamb
er/BCCPARM.DAT -j 4

diffing tp.mol2.save with tp.mol2
PASSED
==============================================================
diffing frcmod.save with frcmod
PASSED
==============================================================
diffing prmtop.mol2.save with prmtop.mol2
possible FAILURE: check prmtop.mol2.dif
==============================================================
        cd antechamber/ash; ./Run.ash
Running: /hpcx/home/e39/e39/uq281/amber8/exe/bondtype -i
ANTECHAMBER_BOND_TYPE.A
C -o ANTECHAMBER_BOND_TYPE.AC -f ac -j full

Running: /hpcx/home/e39/e39/uq281/amber8/exe/atomtype -i ANTECHAMBER_AC.AC0
-o
A
NTECHAMBER_AC.AC -p gaff

Total number of electrons: 100; net charge: 0

Running: /hpcx/home/e39/e39/uq281/amber8/exe/divcon
Unable to find mopac charges in divcon.out
  ./Run.ash: Program error
make: 1254-004 The error code from the last command is 1.

Stop.
########################################################################

There are several failures, do I need to check all of them or it is the last
one which is the mopac charges in divon.out? do you have any ideas how to
get pass these errors?

And after all these tests passed, can I start to compile the parallel parts?
are there any particular points I need to pay attention to?

Thanks very very much.

Best regards,
Rachel

On 11/23/06, Xuebin Qiao <xbqiao_at_gmail.com> wrote:
>
> Hi, Rachel:
>
> Sorry for previous broken mail body. This is the right one.
>
>
> On 11/23/06, Rachel < comeonsos_at_googlemail.com> wrote:
> >
> > Hi, Xuebin,
> >
> > Thanka for your reply. After I removed the -traditional as you
> > suggested, the first mistake is gone, however, about the 2nd one, althought
> > I commented out the "Machines/ibm_aix" out, I still have the following
> > error:
> > #############################################################
> >
> > 1501-510 Compilation successful for file _lapack.f.
> > cd ../ ; make sys.a
> >
> ^^^^^^^^^^^^^^^^^
> This line means that the value of SYSDIR in Makefile is null. You could
> try following steps.
>
> First, I guess you did not comment out *whole line* in configure file
> (maybe simply delete word Machines/ibm_aix), if so, remember to comment out
> whole line, and run configure script.
>
> If above not work, try to change "sysdir='Machines/ibm_aix' " to
> "sysdir='lib' ", run configure again.
>
> Finally, if you do not bother to tweak the configure file. Let's do it
> manually. Open file 'src/Makefile', you'll see following lines:
>
>
> serial: configured_serial
> ...
> cd lib; make install
> cd addles; make install
> cd sander; make serial
> cd ptraj; make install
> ...
>
> This part of Makefile is quit self-explanatory. It means the actual
> compile process is to go to *each* subdirectory under 'src' and run 'make
> install' except that the sander subdirectory should use 'make serial' or
> 'make parallel' instead.
>
> --
> ... there have been two really clean,
> consistent models of programming so far:
> the C model and the Lisp model.
> These two seem points of high ground,
> with swampy lowlands between them.
>
> --Paul Graham
>

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu