Gilles, On hopper there aren't any psm libraries - its an infiniband/infinipath free system - at least on the compute nodes.
For my own work, I never use things like the platform files, I just do ./configure --prefix=blahblah --enable-mpi-java (and whatever else I want to test this tie) Thanks for the ideas though, Howard 2015-08-14 19:20 GMT-06:00 Gilles Gouaillardet < gilles.gouaillar...@gmail.com>: > Howard, > > I have no infinipath hardware, but the infinipath libraries are installed. > I tried to run with --mca mtl_psm_priority 0 instead of --mca mtl ^psm > but that did not work. > without psm mtl, I was unable to reproduce the persistent communication > issue, > so I concluded there was only one issue here. > > do you configure with --disable-dlopen on hopper ? > I wonder whether --mca mtl ^psm is effective if dlopen is disabled > > Cheers, > > Gilles > > On Saturday, August 15, 2015, Howard Pritchard <hpprit...@gmail.com> > wrote: > >> Hi Jeff, >> >> I don't know why Gilles keeps picking on the persistent request problem >> and mixing >> it up with this user bug. I do think for this user the psm probably is >> the problem. >> >> >> They don't have anything to do with each other. >> >> I can reproduce the persistent request problem on hopper consistently. >> As I said >> on the telecon last week it has something to do with memory corruption >> with the >> receive buffer that is associated with the persistent request. >> >> Howard >> >> >> 2015-08-14 11:21 GMT-06:00 Jeff Squyres (jsquyres) <jsquy...@cisco.com>: >> >>> Hmm. Oscar's not around to ask any more, but I'd be greatly surprised >>> if he had InfiniPath on his systems where he ran into this segv issue...? >>> >>> >>> > On Aug 14, 2015, at 1:08 PM, Howard Pritchard <hpprit...@gmail.com> >>> wrote: >>> > >>> > Hi Gilles, >>> > >>> > Good catch! Nate we hadn't been testing on a infinipath system. >>> > >>> > Howard >>> > >>> > >>> > 2015-08-14 0:20 GMT-06:00 Gilles Gouaillardet <gil...@rist.or.jp>: >>> > Nate, >>> > >>> > i could get rid of the problem by not using the psm mtl. >>> > the infinipath library (used by the psm mtl) sets some signal handlers >>> that conflict with the JVM >>> > that can be seen by running >>> > mpirun -np 1 java -Xcheck:jni MPITestBroke data/ >>> > >>> > so instead of running >>> > mpirun -np 1 java MPITestBroke data/ >>> > please run >>> > mpirun --mca mtl ^psm -np 1 java MPITestBroke data/ >>> > >>> > that solved the issue for me >>> > >>> > Cheers, >>> > >>> > Gilles >>> > >>> > On 8/13/2015 9:19 AM, Nate Chambers wrote: >>> >> I appreciate you trying to help! I put the Java and its compiled >>> .class file on Dropbox. The directory contains the .java and .class files, >>> as well as a data/ directory: >>> >> >>> >> >>> http://www.dropbox.com/sh/pds5c5wecfpb2wk/AAAcz17UTDQErmrUqp2SPjpqa?dl=0 >>> >> >>> >> You can run it with and without MPI: >>> >> >>> >> > java MPITestBroke data/ >>> >> > mpirun -np 1 java MPITestBroke data/ >>> >> >>> >> Attached is a text file of what I see when I run it with mpirun and >>> your debug flag. Lots of debug lines. >>> >> >>> >> >>> >> Nate >>> >> >>> >> >>> >> >>> >> >>> >> >>> >> On Wed, Aug 12, 2015 at 11:09 AM, Howard Pritchard < >>> hpprit...@gmail.com> wrote: >>> >> Hi Nate, >>> >> >>> >> Sorry for the delay in getting back to you. >>> >> We're somewhat stuck on how to help you, but here are two suggestions. >>> >> >>> >> Could you add the following to your launch command line >>> >> >>> >> --mca odls_base_verbose 100 >>> >> >>> >> so we can see exactly what arguments are being feed to java when >>> launching >>> >> your app. >>> >> >>> >> Also, if you could put your MPITestBroke.class file somewhere (like >>> google drive) >>> >> where we could get it and try to run locally or at NERSC, that might >>> help us >>> >> narrow down the problem. Better yet, if you have the class or jar >>> file for >>> >> the entire app plus some data sets, we could try that out as well. >>> >> >>> >> All the config outputs, etc. you've sent so far indicate a correct >>> installation >>> >> of open mpi. >>> >> >>> >> Howard >>> >> >>> >> >>> >> On Aug 6, 2015 1:54 PM, "Nate Chambers" <ncham...@usna.edu> wrote: >>> >> Howard, >>> >> >>> >> I tried the nightly build openmpi-dev-2223-g731cfe3 and it still >>> segfaults as before. I must admit I am new to MPI, so is it possible I'm >>> just configuring or running incorrectly? Let me list my steps for you, and >>> maybe something will jump out? Also attached is my config.log. >>> >> >>> >> >>> >> CONFIGURE >>> >> ./configure --prefix=<install-dir> --enable-mpi-java CC=gcc >>> >> >>> >> MAKE >>> >> make all install >>> >> >>> >> RUN >>> >> <install-dir>/mpirun -np 1 java MPITestBroke twitter/ >>> >> >>> >> >>> >> DEFAULT JAVA AND GCC >>> >> >>> >> $ java -version >>> >> java version "1.7.0_21" >>> >> Java(TM) SE Runtime Environment (build 1.7.0_21-b11) >>> >> Java HotSpot(TM) 64-Bit Server VM (build 23.21-b01, mixed mode) >>> >> >>> >> $ gcc --v >>> >> Using built-in specs. >>> >> Target: x86_64-redhat-linux >>> >> Configured with: ../configure --prefix=/usr --mandir=/usr/share/man >>> --infodir=/usr/share/info --with-bugurl= >>> http://bugzilla.redhat.com/bugzilla --enable-bootstrap --enable-shared >>> --enable-threads=posix --enable-checking=release --with-system-zlib >>> --enable-__cxa_atexit --disable-libunwind-exceptions >>> --enable-gnu-unique-object >>> --enable-languages=c,c++,objc,obj-c++,java,fortran,ada >>> --enable-java-awt=gtk --disable-dssi >>> --with-java-home=/usr/lib/jvm/java-1.5.0-gcj-1.5.0.0/jre >>> --enable-libgcj-multifile --enable-java-maintainer-mode >>> --with-ecj-jar=/usr/share/java/eclipse-ecj.jar --disable-libjava-multilib >>> --with-ppl --with-cloog --with-tune=generic --with-arch_32=i686 >>> --build=x86_64-redhat-linux >>> >> Thread model: posix >>> >> gcc version 4.4.7 20120313 (Red Hat 4.4.7-4) (GCC) >>> >> >>> >> >>> >> >>> >> >>> >> >>> >> On Thu, Aug 6, 2015 at 7:58 AM, Howard Pritchard <hpprit...@gmail.com> >>> wrote: >>> >> HI Nate, >>> >> >>> >> We're trying this out on a mac running mavericks and a cray xc >>> system. the mac has java 8 >>> >> while the cray xc has java 7. >>> >> >>> >> We could not get the code to run just using the java launch command, >>> although we noticed if you add >>> >> >>> >> catch(NoClassDefFoundError e) { >>> >> >>> >> System.out.println("Not using MPI its out to lunch for now"); >>> >> >>> >> } >>> >> >>> >> as one of the catches after the try for firing up MPI, you can get >>> further. >>> >> >>> >> Instead we tried on the two systems using >>> >> >>> >> mpirun -np 1 java MPITestBroke tweets repeat.txt >>> >> >>> >> and, you guessed it, we can't reproduce the error, at least using >>> master. >>> >> >>> >> Would you mind trying to get a copy of nightly master build off of >>> >> >>> >> http://www.open-mpi.org/nightly/master/ >>> >> and install that version and give it a try. >>> >> >>> >> If that works, then I'd suggest using master (or v2.0) for now. >>> >> Howard >>> >> >>> >> >>> >> >>> >> >>> >> 2015-08-05 14:41 GMT-06:00 Nate Chambers <ncham...@usna.edu>: >>> >> Howard, >>> >> >>> >> Thanks for looking at all this. Adding System.gc() did not cause it >>> to segfault. The segfault still comes much later in the processing. >>> >> >>> >> I was able to reduce my code to a single test file without other >>> dependencies. It is attached. This code simply opens a text file and reads >>> its lines, one by one. Once finished, it closes and opens the same file and >>> reads the lines again. On my system, it does this about 4 times until the >>> segfault fires. Obviously this code makes no sense, but it's based on our >>> actual code that reads millions of lines of data and does various >>> processing to it. >>> >> >>> >> Attached is a tweets.tgz file that you can uncompress to have an >>> input directory. The text file is just the same line over and over again. >>> Run it as: >>> >> >>> >> java MPITestBroke tweets/ >>> >> >>> >> >>> >> Nate >>> >> >>> >> >>> >> >>> >> >>> >> >>> >> On Wed, Aug 5, 2015 at 8:29 AM, Howard Pritchard <hpprit...@gmail.com> >>> wrote: >>> >> Hi Nate, >>> >> >>> >> Sorry for the delay in getting back. Thanks for the sanity check. >>> You may have a point about the args string to MPI.init - >>> >> there's nothing the Open MPI is needing from this but that is a >>> difference with your use case - your app has an argument. >>> >> >>> >> Would you mind adding a >>> >> >>> >> System.gc() >>> >> >>> >> call immediately after MPI.init call and see if the gc blows up with >>> a segfault? >>> >> >>> >> Also, may be interesting to add the -verbose:jni to your command line. >>> >> >>> >> We'll do some experiments here with the init string arg. >>> >> >>> >> Is your app open source where we could download it and try to >>> reproduce the problem locally? >>> >> >>> >> thanks, >>> >> >>> >> Howard >>> >> >>> >> >>> >> 2015-08-04 18:52 GMT-06:00 Nate Chambers <ncham...@usna.edu>: >>> >> Sanity checks pass. Both Hello and Ring.java run correctly with the >>> expected program's output. >>> >> >>> >> Does MPI.init(args) expect anything from those command-line args? >>> >> >>> >> >>> >> Nate >>> >> >>> >> >>> >> On Tue, Aug 4, 2015 at 12:26 PM, Howard Pritchard < >>> hpprit...@gmail.com> wrote: >>> >> Hello Nate, >>> >> >>> >> As a sanity check of your installation, could you try to compile the >>> examples/*.java codes using the mpijavac you've installed and see that >>> those run correctly? >>> >> I'd be just interested in the Hello.java and Ring.java? >>> >> >>> >> Howard >>> >> >>> >> >>> >> >>> >> >>> >> >>> >> >>> >> >>> >> 2015-08-04 14:34 GMT-06:00 Nate Chambers <ncham...@usna.edu>: >>> >> Sure, I reran the configure with CC=gcc and then make install. I >>> think that's the proper way to do it. Attached is my config log. The >>> behavior when running our code appears to be the same. The output is the >>> same error I pasted in my email above. It occurs when calling MPI.init(). >>> >> >>> >> I'm not great at debugging this sort of stuff, but happy to try >>> things out if you need me to. >>> >> >>> >> Nate >>> >> >>> >> >>> >> On Tue, Aug 4, 2015 at 5:09 AM, Howard Pritchard <hpprit...@gmail.com> >>> wrote: >>> >> Hello Nate, >>> >> >>> >> As a first step to addressing this, could you please try using gcc >>> rather than the Intel compilers to build Open MPI? >>> >> >>> >> We've been doing a lot of work recently on the java bindings, etc. >>> but have never tried using any compilers other >>> >> than gcc when working with the java bindings. >>> >> >>> >> Thanks, >>> >> >>> >> Howard >>> >> >>> >> >>> >> 2015-08-03 17:36 GMT-06:00 Nate Chambers <ncham...@usna.edu>: >>> >> We've been struggling with this error for a while, so hoping someone >>> more knowledgeable can help! >>> >> >>> >> Our java MPI code exits with a segfault during its normal operation, >>> but the segfault occurs before our code ever uses MPI functionality like >>> sending/receiving. We've removed all message calls and any use of >>> MPI.COMM_WORLD from the code. The segfault occurs if we call MPI.init(args) >>> in our code, and does not if we comment that line out. Further vexing us, >>> the crash doesn't happen at the point of the MPI.init call, but later on in >>> the program. I don't have an easy-to-run example here because our non-MPI >>> code is so large and >>> complicated. We have run simpler test programs with MPI and the segfault >>> does not occur. >>> >> >>> >> We have isolated the line where the segfault occurs. However, if we >>> comment that out, the program will run longer, but then randomly (but >>> deterministically) segfault later on in the code. Does anyone have tips on >>> how to debug this? We have tried several flags with mpirun, but no good >>> clues. >>> >> >>> >> We have also tried several MPI versions, including stable 1.8.7 and >>> the most recent 1.8.8rc1 >>> >> >>> >> >>> >> ATTACHED >>> >> - config.log from installation >>> >> - output from `ompi_info -all` >>> >> >>> >> >>> >> OUTPUT FROM RUNNING >>> >> >>> >> > mpirun -np 2 java -mx4g FeaturizeDay datadir/ days.txt >>> >> ... >>> >> some normal output from our code >>> >> ... >>> >> >>> -------------------------------------------------------------------------- >>> >> mpirun noticed that process rank 0 with PID 29646 on node r9n69 >>> exited on signal 11 (Segmentation fault). >>> >> >>> -------------------------------------------------------------------------- >>> >> >>> >> >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27386.php >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27389.php >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27391.php >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27392.php >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27393.php >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27396.php >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27399.php >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27405.php >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27406.php >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> us...@open-mpi.org >>> >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27446.php >>> >> >>> >> >>> >> >>> >> _______________________________________________ >>> >> users mailing list >>> >> >>> >> us...@open-mpi.org >>> >> >>> >> Subscription: >>> >> http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> >>> >> Link to this post: >>> >> http://www.open-mpi.org/community/lists/users/2015/08/27450.php >>> > >>> > >>> > _______________________________________________ >>> > users mailing list >>> > us...@open-mpi.org >>> > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> > Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27465.php >>> > >>> > _______________________________________________ >>> > users mailing list >>> > us...@open-mpi.org >>> > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> > Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27471.php >>> >>> >>> -- >>> Jeff Squyres >>> jsquy...@cisco.com >>> For corporate legal information go to: >>> http://www.cisco.com/web/about/doing_business/legal/cri/ >>> >>> _______________________________________________ >>> users mailing list >>> us...@open-mpi.org >>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users >>> Link to this post: >>> http://www.open-mpi.org/community/lists/users/2015/08/27472.php >>> >> >> > _______________________________________________ > users mailing list > us...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users > Link to this post: > http://www.open-mpi.org/community/lists/users/2015/08/27479.php >
