Re: [OMPI users] OpenIB problems

Hi Andrew, Brock, and everyone else, Andrew Friedley wrote: If this is what I think it is, try using this MCA parameter: -mca btl_openib_ib_timeout 20 Just FYI, in addition to the above, I retried using the gigabit links ('--mca btl tcp,self', right?) and that failed too, so at least in /m

Re: [OMPI users] OpenIB problems

Thanks, We have asked the user to try that and let us know if it fails. I will let the list know if this works. Brock Palen Center for Advanced Computing bro...@umich.edu (734)936-1985 On Nov 21, 2007, at 3:39 PM, Andrew Friedley wrote: If this is what I think it is, try using this MCA par

Re: [OMPI users] OpenIB problems

Hi Brock We have a user whos code keep failing at a similar point in the code. The errors (below) would make me think its a fabric problem, but ibcheckerrors is not returning any issues. He is using openmpi-1.2.0 With OFED on RHEL4, Strangely enough, I hit this exact problem about half an

Re: [OMPI users] OpenIB problems

If this is what I think it is, try using this MCA parameter: -mca btl_openib_ib_timeout 20 If this fixes it -- I don't fully understand what's going on, but it's an issue in the IB fabrics itself. Someone else might be able to explain in more detail.. Andrew Brian Dobbins wrote: Hi Brock

Re: [OMPI users] OpenIB problems

Hi Brock We have a user whos code keep failing at a similar point in the code. The errors (below) would make me think its a fabric problem, but ibcheckerrors is not returning any issues. He is using openmpi-1.2.0 With OFED on RHEL4, Strangely enough, I hit this exact problem about ha

Re: [OMPI users] Problems compiling 1.2.4 using Intel Compiler 10.1.006 on Leopard

Thanks for the suggestion Jeff. Unfortunately, that didn't fix the issue. -Mark On Nov 21, 2007, at 7:55 AM, Jeff Squyres wrote: Can you try also adding CCASFLAGS=-no-multibyte-chars? On Nov 20, 2007, at 2:45 PM, Mark Dobossy wrote: I am running into some roadblocks in compiling OpenMPI 1

[OMPI users] OpenIB problems

We have a user whos code keep failing at a similar point in the code. The errors (below) would make me think its a fabric problem, but ibcheckerrors is not returning any issues. He is using openmpi-1.2.0 With OFED on RHEL4, Far field AIM propagators require(MB):1.441955566406250 Arra

Re: [OMPI users] openmpi-sessions-userid directories

Jeff, Thanks for the info. I will find out what $TMPDIR value they are using at these other sites. Pat J.W. (Pat) O'Bryant,Jr. Business Line Infrastructure Technical Systems, HPC "Jeff Squyres (j

Re: [OMPI users] openmpi-sessions-userid directories

Either that or they were put in a different location (eg, ompi obeys $tmpdir). -jms Sent from my PDA -Original Message- From: pat.o'bry...@exxonmobil.com [mailto:pat.o'bry...@exxonmobil.com] Sent: Wednesday, November 21, 2007 08:07 AM Eastern Standard Time To: Open MPI Users Cc:

Re: [OMPI users] openmpi-sessions-userid directories

Jeff, We have some users that have noticed the files and they made the comment that at other locations where they ran OpenMPI they didn't see these files. So, in their mind, OpenMPI was not installed correctly. Thanks, Pat J.W. (Pat) O'Bryant,Jr. Business Line Infras

Re: [OMPI users] Problems compiling 1.2.4 using Intel Compiler 10.1.006 on Leopard

Can you try also adding CCASFLAGS=-no-multibyte-chars? On Nov 20, 2007, at 2:45 PM, Mark Dobossy wrote: I am running into some roadblocks in compiling OpenMPI 1.2.4 using the latest Intel Compilers (icc, icpc, ifort) version 10.1.006 on OS X Leopard (10.5.1). Using the last revision of the

Re: [OMPI users] reg. heterogeneous cluster confguration

On Nov 20, 2007, at 11:26 PM, Chadalavada Kalyana Krishna wrote: 1) Endian conversion - Open MPI does automatic endian conversion as long as the size of the data types on the communicating nodes is equal. This support is fully qualified from v1.1. Is this a correct statement? It should wor