[AMBER] FW: AMBER Digest, Vol 3348, Issue 1

From: Markus Metz <mmetz.omeros.com> Date: Thu, 29 Apr 2021 19:09:23 +0000 · >I am trying to compile a parallel version of Amber.

--
Dr. Charo I. del Genio
Senior Lecturer in Statistical Physics
Applied Mathematics Research Centre (AMRC)
Design Hub
Coventry University Technology Park
Coventry CV1 5FB
UK
https://charodelgenio.weebly.com<https://charodelgenio.weebly.com>
------------------------------
Message: 8
Date: Thu, 29 Apr 2021 12:43:12 +0200
From: Miguel Garav?s <mgaravis.gmail.com<mailto:mgaravis.gmail.com>>
Subject: Re: [AMBER] Problems to install AmberTools20
To: AMBER Mailing List <amber.ambermd.org<mailto:amber.ambermd.org>>
Message-ID:
<CAJVSO568D8BFDxV0pfLnDy1jJuZ2-wE=V=MyrY8UtOvpV9yTFQ.mail.gmail.com<mailto:CAJVSO568D8BFDxV0pfLnDy1jJuZ2-wE=V=MyrY8UtOvpV9yTFQ.mail.gmail.com>>
Content-Type: text/plain; charset="UTF-8"
Thanks again.
I did a deep cleaning and tried to install it again. The result is still
the same.
In case it helps, I paste here some of the lines concerning existing errors:
-- Failed to find Perl modules: Chemistry::Mol (missing:
PERLMODULES_CHEMISTRY_MOL_MODULE)
-- Could NOT find Boost (missing: Boost_INCLUDE_DIR thread system
program_options iostreams regex timer chrono filesystem graph)
-- Cannot find PLUMED. You will still be able to load it at runtime. If
you want to link it at build time, set PLUMED_ROOT to where you installed
it.
CMake Error at cmake/LibraryUtils.cmake:75 (message):
Could not determine whether
"/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX11.1.sdk/usr/lib/libbz2.tbd"
is a static or shared library, it does not have a known suffix.
The OSX version I have is MacOS Catalina 10.15.7 (19H15)
The version of XCode is: 12.4 (12D4e)
And the clang version is clang-1200.0.32.29
MGC
El mi?, 28 abr 2021 a las 18:41, David A Case (<david.case.rutgers.edu<mailto:david.case.rutgers.edu>>)
escribi?:
> On Wed, Apr 28, 2021, Miguel Garav?s wrote:
> >
> >I can confirm that update.11 is applied. It appears in the list after
> >typing the command you indicated.
>
> OK, we have to search for something else.
>
> First, I think it's worth trying a "clean" build: re-download the
> AmberTools20.tar.bz2 file (unless you still have it on your computer),
> un-tar and try a "run_cmake" job again. It's possible that some cached
> file
> is interfering with what you did before.
>
> If that doesn't help, please let us know what version of OSX you have, and
> which version of Xcode (open /Applications/Xcode.app and look at the "about
> Xcode menu item.) Also, type "clang --version" at a terminal prompt, and
> report the result.
>
> ...thanks...dac
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org<mailto:AMBER.ambermd.org>
> http://lists.ambermd.org/mailman/listinfo/amber<http://lists.ambermd.org/mailman/listinfo/amber>
>
------------------------------
Message: 9
Date: Thu, 29 Apr 2021 08:09:35 -0400
From: David A Case <david.case.rutgers.edu<mailto:david.case.rutgers.edu>>
Subject: Re: [AMBER] ??: [ambermd.org??]Re: ??: [ambermd.org??]Re:
Max relative errors questions in parallel and cuda.serial test
To: AMBER Mailing List <amber.ambermd.org<mailto:amber.ambermd.org>>
Message-ID:
<20210429120935.guzxnqmkym4ff4hn.rulx-presler-l.rad.rutgers.edu<mailto:20210429120935.guzxnqmkym4ff4hn.rulx-presler-l.rad.rutgers.edu>>
Content-Type: text/plain; charset=utf-8; format=flowed
On Thu, Apr 29, 2021, ??? wrote:
>
>Thanks for your follow-up. If the errors are with GB calculation, then
>that's quite serious because the most common post-processing tasks we will
>do is MMPB(GB)SA.
Do you use pmemd.cuda for such tasks? Is it really significantly faster
than pmemd or sander on a CPU? I ask because getting single frame energies
involves a lot of initialization tasks that reduce the speed advantage of
the GB calculation itself.
>
>David, what will be the potential way to solve this DPFP
>errors? Re-compiling with different ways? Install on a different
>workstation/cluster.....?
Both of the above are good ideas. It's certainly possible that you are
seeing a hardware problem, so trying a different GPU might help.
Changing compilers or OS might also help identify the problem. Right now,
we don't have any way to reproduce the problem at our end.
...regards....dac
------------------------------
Message: 10
Date: Thu, 29 Apr 2021 08:21:53 -0400
From: David A Case <david.case.rutgers.edu<mailto:david.case.rutgers.edu>>
Subject: Re: [AMBER] Problems to install AmberTools20
To: AMBER Mailing List <amber.ambermd.org<mailto:amber.ambermd.org>>
Message-ID:
<20210429122153.4nhifvuntn5vozjk.rulx-presler-l.rad.rutgers.edu<mailto:20210429122153.4nhifvuntn5vozjk.rulx-presler-l.rad.rutgers.edu>>
Content-Type: text/plain; charset=iso-8859-1; format=flowed
On Thu, Apr 29, 2021, Miguel Garav?s wrote:
>
>I did a deep cleaning and tried to install it again. The result is still
>the same.
>
>CMake Error at cmake/LibraryUtils.cmake:75 (message):
> Could not determine whether
>"/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX11.1.sdk/usr/lib/libbz2.tbd"
> is a static or shared library, it does not have a known suffix.
>
>The OSX version I have is MacOS Catalina 10.15.7 (19H15)
>The version of XCode is: 12.4 (12D4e)
>And the clang version is clang-1200.0.32.29
I don't have any more suggestions for now. AmberTools21 will be released
today or tomorrow, so you might try that when it is available.
If you have access (via a friend?) to any other OSX machine, you might try
an installation there. If it works, one could try to see what difference
there is between the two computers.
If time is critical, you could try using the configure option in place of
cmake. See Section 2.5 of the Amber 2020 Reference Manual. Or, you could
install the conda package if that meets your needs.
It seems that very few Amber developers (besides me!) use OSX now, but I
think a lot of users do. We haven't previously seen any such reports since
update.11 was released last October. But maybe someone on the list will
chime in with ideas here.
....dac
------------------------------
Message: 11
Date: Thu, 29 Apr 2021 15:59:40 +0000
From: "Azurmendi, Hugo" <Hugo.Azurmendi.fda.hhs.gov<mailto:Hugo.Azurmendi.fda.hhs.gov>>
Subject: [AMBER] Adventures installing AMBER 20 in Centos 7
To: "amber.ambermd.org<mailto:amber.ambermd.org>" <amber.ambermd.org<mailto:amber.ambermd.org>>
Message-ID: <D63291C4-E657-4E5F-A13A-59603C5398BE.fda.gov<mailto:D63291C4-E657-4E5F-A13A-59603C5398BE.fda.gov>>
Content-Type: text/plain; charset="utf-8"
Hello there.
I just finished installing AMBER 20 on a new Linux and found a few problems that I think developers should be aware of. I did a fresh install of Centos 7 installing the recommended packages for centos with yum. The problems were:
1. For parallel I followed the instructions for Fedora to install openmpi as follows (not sure all these are required, but it worked):
? yum install dnf
? dnf install openmpi openmpi-devel environment-modules
? source /etc/profile.d/modules.sh (to load modules)
? Add line 'module load mpi/openmpi-x86_64' to .bashrc/.cshrc.
At this point I could start ?make test.parallel?, however it would stall (without messages) after a few tests. I finally could get it to go all the way through by taking out (renaming the corresponding $AMBERHOME/test/testname) the tests for dhfr and rism3d. I haven?t investigated further this issue.
1. The initial installation and test I did it in bash, deciding to switch to csh later. I sourced to amber.csh (located in the amber20 folder) in my .cshrc but didn?t work:
* It warned that the shell was not csh, which it was. I fixed replacing ( "$0" != '-tcsh' && "$0" != '-csh' ) for ( "$0" !~ '*csh' ) in the ?if? line of amber.csh.
* AMBERHOME was set to the path of the .cshrc instead of the path to amber.csh. I commented the setenv line in amber.csh and set AMBERHOME in my .cshrc in the old fashion.
Hope that helps to improve future releases. Looking forward to comments if I did something wrong.
Hugo
------------------------------
Message: 12
Date: Thu, 29 Apr 2021 12:01:07 -0600
From: "German P. Barletta" <pbarletta.gmail.com<mailto:pbarletta.gmail.com>>
Subject: [AMBER] cpptraj energy using 2 masks
To: ?amber.ambermd.org<mailto:?amber.ambermd.org>? <amber.ambermd.org<mailto:amber.ambermd.org>>
Message-ID:
<CAJo28majJB5FPaOQ6Uikved8zSphPoGCBCJvosNw0oAbYMhygg.mail.gmail.com<mailto:CAJo28majJB5FPaOQ6Uikved8zSphPoGCBCJvosNw0oAbYMhygg.mail.gmail.com>>
Content-Type: text/plain; charset="UTF-8"
Dear amber users and developers,
I'd like to calculate the nonbonded (LJ + Coulombic) energy terms for the
interaction between a ligand (residue 164) and a protein (residues 1-163)
but if I use the mask ":1-164", cpptraj will also include the interactions
of the protein and the ligand with themselves.
gromacs tutorials
<http://www.mdtutorials.com/gmx/complex/09_analysis.html<http://www.mdtutorials.com/gmx/complex/09_analysis.html>> show
a convenient way to do this, but how can I do the same with cpptraj?
Thanks!
------------------------------
Message: 13
Date: Thu, 29 Apr 2021 14:29:49 -0400
From: David A Case <david.case.rutgers.edu<mailto:david.case.rutgers.edu>>
Subject: Re: [AMBER] Adventures installing AMBER 20 in Centos 7
To: AMBER Mailing List <amber.ambermd.org<mailto:amber.ambermd.org>>
Message-ID:
<20210429182949.fea2p6kkmhu7opjr.ufa33kbb210dl.rad.rutgers.edu<mailto:20210429182949.fea2p6kkmhu7opjr.ufa33kbb210dl.rad.rutgers.edu>>
Content-Type: text/plain; charset=us-ascii; format=flowed
On Thu, Apr 29, 2021, Azurmendi, Hugo wrote:
>I just finished installing AMBER 20 on a new Linux and found a few problems
>that I think developers should be aware of. I did a fresh install of Centos
>7 installing the recommended packages for centos with yum. The problems
>were:
>
> 1. For parallel I followed the instructions for Fedora
Can you say a little more about exactly what you did? Did you make any
changes to run_cmake other than setting -DMPI_TRUE? What value did you set
for DO_PARALLEL, and what was the exact command you used to run the tests?
* It warned that the shell was not csh
I can believe that the amber.csh may not work in all cases, but it would be
more helpful if you could quote the exact error message, since that would
tell us what the value of the "$0" variable is in your case.
...thx...dac
------------------------------
_______________________________________________
AMBER mailing list
AMBER.ambermd.org<mailto:AMBER.ambermd.org>
http://lists.ambermd.org/mailman/listinfo/amber<http://lists.ambermd.org/mailman/listinfo/amber>
End of AMBER Digest, Vol 3348, Issue 1
**************************************
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber