[maker-devel] MAKER and OpenMPI

[Carson Holt]

When you install MAKER, you have to provide the location to mpi.h and mpicc. You may have to reinstall, and give the correct location for one or both (don’t just trust the location that shows up, it may be the wrong one). So you will need to track down both of those for MPICH (depending on where you installed it to).

Then do ‘which -a mpiexec’ to get all locations for every mpiexec found in your path. You will need to ensure the one at the top is the one you want. If it’s not, then you need to reconfigure your PATH or provide the full path to the mpiexec you want each time you launch it.

—Carson



> On Feb 8, 2017, at 10:27 AM, Seth Munholland <munholl at uwindsor.ca> wrote:
> 
> That's for sure what it is, but I never installed a second flavour.  I was commited to MPICH from the get go and am trying to figure out where OpenMPI came from.
> 
> Seth Munholland, B.Sc.
> Department of Biological Sciences
> Rm. 304 Biology Building
> University of Windsor
> 401 Sunset Ave. N9B 3P4
> T: (519) 253-3000 Ext: 4755 <>
> On Wed, Feb 8, 2017 at 12:22 PM, Carson Holt <carsonhh at gmail.com <mailto:carsonhh at gmail.com>> wrote:
> You might have two different MPI flavors installed. Since the library and executable files have the same name, this kind of mismatch can happen in that case. Check the locations of MPI libraries given to MAKER during install, and the location of the mpiexec being used when you run. There is likely a mismatch with parts of one MPI flavor being used for one but not the other.
> 
> —Carson
> 
> 
> 
>> On Feb 8, 2017, at 10:18 AM, Seth Munholland <munholl at uwindsor.ca <mailto:munholl at uwindsor.ca>> wrote:
>> 
>> Hello everyone,
>> 
>> I'm setting up a cluster and configured MPICH on it, but after installing MAKER my mpiexec gave me an error that comes from an improperly configured OpenMPI.  The weird thing is I haven't installed anything else associated to MPI, particularly not OpenMPI.  Did I get server gremlins or is there any way MAKER's installation somehow is looking for OpenMPI?
>> 
>> Seth Munholland, B.Sc.
>> Department of Biological Sciences
>> Rm. 304 Biology Building
>> University of Windsor
>> 401 Sunset Ave. N9B 3P4
>> T: (519) 253-3000 Ext: 4755 <>_______________________________________________
>> maker-devel mailing list
>> maker-devel at box290.bluehost.com <mailto:maker-devel at box290.bluehost.com>
>> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org <http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org>
> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20170208/8041d544/attachment-0003.html>