Re: [OMPI users] Java-OpenMPI returns with SIGSEGV

2016-07-07 Thread Saliya Ekanayake
I've received SIGSEGV a few times for different reasons with OpenMPI Java and one of the most common reasons was the ulimit settings. You might want to look at -l (max lock memory) -u (max user processes), -n (open files). Here's a snapshot of what we use in our clusters running OpenMPI and Java

Re: [OMPI users] Java-OpenMPI returns with SIGSEGV

2016-07-07 Thread Gundram Leifert
Hello Gilles, I tried you code and it crashes after 3-15 iterations (see (1)). It is always the same error (only the "94" varies). Meanwhile I think Java and MPI use the same memory because when I delete the hash-call, the program runs sometimes more than 9k iterations. When it crashes, there

Re: [OMPI users] Java-OpenMPI returns with SIGSEGV

2016-07-07 Thread Gilles Gouaillardet
Gundram, can you please provide more information on your environment : - configure command line - OS - memory available - ulimit -a - number of nodes - number of tasks used - interconnect used (if any) - batch manager (if any) Cheers, Gilles On 7/7/2016 4:17 PM, Gundram Leifert wrot

[OMPI users] Class information in OpenMPI

2016-07-07 Thread Emani, Murali
Hi all, I want to know if there is “class diagram” for OpenMPI code base that shows existing classes and dependencies/associations. Are there any available tools to extract and visualize this information. — Murali

Re: [OMPI users] Java-OpenMPI returns with SIGSEGV

2016-07-07 Thread Nathaniel Graham
Hello Gundram, I was also not able to reproduce the issue on my computer (OS X El Capitan). I ran both your code and the one provided by Gilles with no issues. I can try it on my Ubuntu machine when I get home. -Nathan On Thu, Jul 7, 2016 at 2:05 AM, Gilles Gouaillardet wrote: > Gundram, > >

[OMPI users] mpirun has exited due to process rank N

2016-07-07 Thread Alberti, Andrea
Hi, my name is Andrea and I am a new openMPI user. I have a code compiled with: intel/16.0.3 openmpi/1.6.5 --> When I try to run my code with: mpirun -n N ./code.exe a) the code correctly runs and gives results if N<=25 b) the code gives the following error if N>25: mpirun

Re: [OMPI users] mpirun has exited due to process rank N

2016-07-07 Thread Ralph Castain
Try running one of the OMPI example codes and verify that things run correctly if N > 25. I suspect you have an error in your code that causes it to fail if its rank is > 25. > On Jul 7, 2016, at 2:49 PM, Alberti, Andrea wrote: > > Hi, > > my name is Andrea and I am a new openMPI user. > >

Re: [OMPI users] Class information in OpenMPI

2016-07-07 Thread Ralph Castain
We used to have Doxygen support that would create what you are asking for, but I don’t think anyone has maintained it in a long time. I ran “doxygen” at the top-level directory and it did indeed generate a bunch of html, but I’m not sure it is all that helpful. You might take a look and see if

Re: [OMPI users] mpirun has exited due to process rank N

2016-07-07 Thread Gilles Gouaillardet
Andrea, On top of what Ralph just wrote, you might want to upgrade OpenMPI to the latest stable version (1.10.3) 1.6.5 is pretty antique and is no more maintained. the message indicates that one process died, and so many things could cause a process crash. (since the crash occurs only wi