Begin forwarded message:
From: "Garrison, David" Garrison@uhcl.edu To: Erik Schnetter schnetter@gmail.com Subject: Re: [Users] PUGH issues with upgrade to Cactus 4.2.3 Date: March 18, 2015 at 18:10:20 EDT
Thanks,
It looks like I was compiling with openmpi-1.6.5 but running rocks-openmpi.
-DG
-- David Garrison, Ph.D. Director of Graduate Programs for the School of Science and Computer Engineering, Associate Professor and Chair of Physics University of Houston-Clear Lake Bayou 3531 Houston, TX 77058
Tel: 281-283-3796 Fax: 281-283-3709 http://sce.uhcl.edu/garrison
For more information about the UHCL Physics Program go to http://www.uhcl.edu/physics or sign up for UHCL newsletters at https://ecommunications.houston.edu/uhcl/join.html
"If we knew what it was we were doing, it would not be called research, would it?" ‹ Albert Einstein.
On Mar 18, 2015, at 4:37 PM, Erik Schnetter schnetter@gmail.com wrote:
David
This is likely an issue with your MPI configuration. Either Cactus didn't recognize that you are using MPI, or you are building Cactus with one MPI implementation and are then using mpirun from another MPI implementation.
If you want help diagnosing this, then you can open a ticket on trac.einsteintoolkit.org and attach your configure options, your thorn list, as well as the screen output from configure there.
-erik
On Mar 18, 2015, at 17:34 , Garrison, David Garrison@uhcl.edu wrote:
Hello,
Ever since I upgraded to Cactus 4.2.3, my code seems to have issues with MPI and PUGH. For example, when I perform an 8 processor run, I get 8 copies of full grid output instead of the 8 processors splitting up the grid and creating 1 output. This works with both the arrangements I wrote and WaveToy. Any ideas how to fix this?
-DG
-- David Garrison, Ph.D. Director of Graduate Programs for the School of Science and Computer Engineering, Associate Professor and Chair of Physics University of Houston-Clear Lake Bayou 3531 Houston, TX 77058
Tel: 281-283-3796 Fax: 281-283-3709 http://sce.uhcl.edu/garrison
For more information about the UHCL Physics Program go to http://www.uhcl.edu/physics or sign up for UHCL newsletters at https://ecommunications.houston.edu/uhcl/join.html
"If we knew what it was we were doing, it would not be called research, would it?" ‹ Albert Einstein.
Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/
My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from https://sks-keyservers.net.
Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/
My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from https://sks-keyservers.net.