[molpro-user] Molpro2009.1 parallel mode problem
Manhui Wang
wangm9 at cardiff.ac.uk
Mon Jul 12 15:18:44 BST 2010
Hi Radosław,
I noticed you are using very old mvapich2 library, which contains some
bugs about MPI window and memory leakage. Please install the latest
MVAPICH2 1.5 and build Molpro with it.
Best wishes,
Manhui
Radoslaw Stachowski wrote:
> Hello
>
> The problem is I cannot run molpro in parallel mode.
>
> Compilation is ok, there are no suspicious logs.
>
> my CONFIG file:
> CONFIGURE_OPTIONS="-x86_64" "-icc" "-ifort" "-mpp"
> "-mppbase" "/usr/mpi/intel/mvapich2-1.0.3/include/"
> "-var" "INSTBIN=/home/stach008/molpro2009"
> "-var" "INSTLIB=/home/stach008/molpro2009"
> "-var" "INSTHTML=/home/stach008/molpro2009"
>
>
> BLASLIB=-L/opt/intel/Compiler/11.1/064/mkl/lib/em64t -lmkl_intel_ilp64 -lmkl_sequential -lmkl_core
> CC=/opt/intel/Compiler/11.1/064/bin/intel64/icc
> CCVERSION=11.1
> CC_FRONT=
> CDEBUG=-g
> CDEFINE=-D
> CFLAGS=-ftz -fPIC -vec-report0 -DMOLPROC_PAR -DINT64 -DZLIB
> CLEARSPEEDLIB=
> CMPPINCLUDE=/usr/mpi/intel/mvapich2-1.0.3/include
> CSFLAGS=-O3 -I. --dynamic
> FC=/opt/intel/Compiler/11.1/064/bin/intel64/ifort
> FCVERSION=11.1
> FTCFLAGS=molpro unix unix-i8 Linux lapack sf eaf mpi-io mpi2 mpp blas1 blas2 blas3
> F_OPT0=
> F_OPT1=nevpt2_optrpc.f explicit_util.f artwo2.f drv2el_l3ext_lmp2g.f drv2el_l3ext_cen.f rmp2_f12_drv2.f90 ri_lmp2g.f df_llmp2.f
> F_OPT2=integrals.f90 RL4gen1.f basis_integral_shells.f
> LAUNCHER=/usr/mpi/intel/mvapich2-1.0.3/bin/mpiexec -machinefile %h -np %n %x
> LD_ENV=/opt/intel/Compiler/11.1/064/lib/intel64:/opt/intel/Compiler/11.1/064/mkl/lib/em64t
> LD_ENVNAME=LD_LIBRARY_PATH
> LAUNCHER=/usr/mpi/intel/mvapich2-1.0.3/bin/mpiexec -machinefile %h -np %n %x
> LD_ENV=/opt/intel/Compiler/11.1/064/lib/intel64:/opt/intel/Compiler/11.1/064/mkl/lib/em64t
> LD_ENVNAME=LD_LIBRARY_PATH
> MPILIB=-i-dynamic -I/usr/mpi/intel/mvapich2-1.0.3/include -L/usr/lib64 -L/usr/mpi/intel/mvapich2-1.0.3/lib -Wl,-rpath -Wl,/usr/mpi/intel/mvapich2-1.0.3/lib -lmpich -L/usr/lib64 -lrdmacm -libverbs -libumad -lpthread -lrt
> MPPLIB=
> OBJECT_SUFFIX=o
> OWNERPOS=3
> PAPER=a4paper
> PARSE=parse-Linux-x86_64-i8.o
> PATCHER=patcher.exe
> PDFLATEX=
> PNAME=molprop_2009_1_Linux_x86_64_i8
> PTSIZE=11
> RANLIB=ranlib
> RM=rm -rf
> SHAREDLIBFLAGS=-shared --whole-archive ../lib/libmolpro.a --no-whole-archive -rpath-link . -L/opt/intel/Compiler/11.1/064/lib/intel64 -lsvml
> SHAREDLIBSUFFIX=so
> SHELL=/bin/sh
>
> .SUFFIXES:
> MAKEFLAGS+=-r
>
>
>
> I use PBS in interactive mode (for testing)[$ qsub -I -l mem=7200mb -l nodes=2L:ppn=2].
> After taping ./molprop_2009_1_Linux_x86_64_i8 ~/1.inp I get:
>
> "mpiexec_nova: cannot connect to local mpd (/tmp/mpd2.console_stach008); possible causes:
> 1. no mpd is running on this host
> 2. an mpd is running but was started without a "console" (-n option)
> In case 1, you can start an mpd on this host with:
> mpd &
> and you will be able to run jobs just on this host.
> For more details on starting mpds on a set of hosts, see
> the MPICH2 Installation Guide."
>
> So I start mpd daemons manually.
> /usr/mpi/intel/mvapich2-1.0.3/bin/mpdboot -n 2 -f mpd_host (I`ve requested from PBS 2 nodes and 2 proc per node)
>
> and run ./molprop_2009_1_Linux_x86_64_i8 ~/1.inp (1.inp - just a test job input file)
>
> and I get this error :
> molprop_2009_1_Linux_x86_64_i8.exe: rdma_iba_1sc.c:472: MPIDI_CH3I_RDMA_win_create: Assertion `MPIDI_CH3I_RDMA_Process.current_win_num <= (16)' failed.
> molprop_2009_1_Linux_x86_64_i8.exe: rdma_iba_1sc.c:472: MPIDI_CH3I_RDMA_win_create: Assertion `MPIDI_CH3I_RDMA_Process.current_win_num <= (16)' failed.
> molprop_2009_1_Linux_x86_64_i8.exe: rdma_iba_1sc.c:472: MPIDI_CH3I_RDMA_win_create: Assertion `MPIDI_CH3I_RDMA_Process.current_win_num <= (16)' failed.
> molprop_2009_1_Linux_x86_64_i8.exe: rdma_iba_1sc.c:472: MPIDI_CH3I_RDMA_win_create: Assertion `MPIDI_CH3I_RDMA_Process.current_win_num <= (16)' failed.
> forrtl: error (76): Abort trap signal
> Image PC Routine Line Source
> libc.so.6 000000349542E21D Unknown Unknown Unknown
> libc.so.6 000000349542FA1E Unknown Unknown Unknown
> libc.so.6 0000003495427AE1 Unknown Unknown Unknown
> libmpich.so 00002AE1422B60B7 Unknown Unknown Unknown
> libmpich.so 00002AE14224AB83 Unknown Unknown Unknown
> libmpich.so 00002AE142299DD8 Unknown Unknown Unknown
> libmpich.so 00002AE1422F1B92 Unknown Unknown Unknown
> molprop_2009_1_Li 00000000021444B6 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000002143DE2 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000002143BB6 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000001258EBB Unknown Unknown Unknown
> molprop_2009_1_Li 00000000012548BF Unknown Unkforrtl: error (76): Abort trap signal
> Image PC Routine Line Source
> libc.so.6 000000349542E21D Unknown Unknown Unknown
> libc.so.6 000000349542FA1E Unknown Unknown Unknown
> libc.so.6 0000003495427AE1 Unknown Unknown Unknown
> libmpich.so 00002AF52EBE00B7 Unknown Unknown Unknown
> libmpich.so 00002AF52EB74B83 Unknown Unknown Unknown
> libmpich.so 00002AF52EBC3DD8 Unknown Unknown Unknown
> libmpich.so 00002AF52EC1BB92 Unknown Unknown Unknown
> molprop_2009_1_Li 00000000021444B6 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000002143DE2 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000002143BB6 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000001258EBB Unknown Unknown Unknown
> molprop_2009_1_Li 00000000012548BF Unknown Unknown Unknown
> molprop_2009_1_Li 00000000004C3219 Unknown Unknown Unknown
> molprop_2009_1_Li 00000000004BDF65 Unknown Unknown Unknown
> molprop_2009_1_Li 000000000043D56C Unknown Unknown Unknown
> libc.so.6 000000349541C3FB Unknown Unknown Unknown
> molprop_2009_1_Li 000000000043D49A Unknown Unknown Unknown
> nown Unknown
> molprop_2009_1_Li 00000000004C3219 Unknown Unknown Unknown
> molprop_2009_1_Li 00000000004BDF65 Unknown Unknown Unknown
> molprop_2009_1_Li 000000000043D56C Unknown Unknown Unknown
> libc.so.6 000000349541C3FB Unknown Unknown Unknown
> molprop_2009_1_Li 000000000043D49A Unknown Unknown Unknown
> forrtl: error (76): Abort trap signal
> Image PC Routine Line Source
> libc.so.6 000000349542E21D Unknown Unknown Unknown
> libc.so.6 000000349542FA1E Unknown Unknown Unknown
> libc.so.6 0000003495427AE1 Unknown Unknown Unknown
> libmpich.so 00002B5261EBA0B7 Unknown Unknown Unknown
> libmpich.so 00002B5261E4EB83 Unknown Unknown Unknown
> libmpich.so 00002B5261E9DDD8 Unknown Unknown Unknown
> libmpich.so 00002B5261EF5B92 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000002144499 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000002143DE2 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000002143BB6 Unknown Unknown Unknown
> molprop_2009_1_Li 0000000001258EBB Unknown Unknown Unknown
> molprop_2009_1_Li 00000000012548BF Unknown Unknown Unknown
>
> Any ideas?
> I will appreciate any help.
>
> Cheers,
>
> Radosław Stachowski
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user
--
-----------
Manhui Wang
School of Chemistry, Cardiff University,
Main Building, Park Place,
Cardiff CF10 3AT, UK
Telephone: +44 (0)29208 76637
More information about the Molpro-user
mailing list