On Tue, Feb 08, 2022, Ilyas Yildirim wrote:
>
>I finally installed AmberTools21 using both the tradition method and cmake,
>but it was not easy. Here are what I have seen:
Can you say what operating system and version you are using. The
"traditional" installation system is no longer maintained, is not documented
in the Reference Manual, and will be disabled in future releases.
> 1. In the cmake version, I had to first do "clean_build" in order to
> properly install the parallel version.
This can happen, but won't be fixed unless we can figure out the problem.
Note that with cmake, all requested versions are installed on each build.
> Nevertheless, mpi4py needs to be
> installed. It is a little bit trickly; I think (but I am not sure as I
> have tried a lot of combinations) amber.conda works.
Volunteers, especially those who use mmpbsa.py are solicited here. Adrian:
is there anyone at your end who could look at this (and other problems with
mmpbsa.py? Do we have a test case for MMPBSA.py.MPI?
Developers: can you see if you can import mpi4py on your machines, assuming
that you downloaded at the build stage? The reason that amber.pip is
missing is because the miniconda download no longer provides the pip file
that amber.pip would be pointing to. Perhaps people are not getting mpi4py
because they are not getting pip.
> 5. Also in both methods, in order to install the code in the cluster,
> I had to install the latest version of openmpi (openmpi-4.0.2) using
> configure_openmpi.
Are you using CentOS or RHEL? See
https://ambermd.org/Installation.php for
comments on MPI for these Linux versions?
Ilyas...thanks for the report...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Feb 08 2022 - 06:00:02 PST
