MadGraph5_aMC@NLO

Survey ceases working after many generations

Asked by Jared Evans

While parallelizing MG5 to generate many different samples, I find that it will eventually cease working entirely. It generates many, many processes and events successfully, but then it suddenly stops working. Processes that began running survey before this cutoff will continue generating events successfully. However, all new attempts to generate anything meet with the errors below, always in survey.

My kludge to solve this problem so far is that I simply unzip a new version of MG5, swap in the model files, scripts and event folders. The new version will run without any issues for a few hundred processes before it craps out as well. I first encountered this in MG5_v1.4.3, it has happened thrice now in v1.4.5 as well. The kludge is time consuming and will grow more tedious as time rolls on. Any idea how to prevent this from happening? If not, do you have any idea how to do a kludge to move a fix (repaired files, or whatever) into the old MG5 directory as that would be much less tedious than moving all of my model files, scripts, data, etc. into a new MG5 folder?

My apologies if someone has asked about this before. I did not find it.

Thank you!
Jared

Screen output for a sample generation:

************************************************************
* *
* W E L C O M E to M A D G R A P H 5 *
* *
* *
* * * *
* * * * * *
* * * * * 5 * * * * *
* * * * * *
* * * *
* *
* VERSION 1.4.5 2012-04-20 *
* *
* The MadGraph Development Team - Please visit us at *
* https://server06.fynu.ucl.ac.be/projects/madgraph *
* *
* Type 'help' for in-line help. *
* Type 'tutorial' to learn how MG5 works *
* *
************************************************************
load MG5 configuration from input/mg5_configuration.txt
Using default text editor "vi". Set another one in ./input/mg5_configuration.txt
Using default eps viewer "gv". Set another one in ./input/mg5_configuration.txt
Using default web browser "firefox". Set another one in ./input/mg5_configuration.txt
Loading default model: sm
INFO: Restrict model sm with file models/sm/restrict_default.dat .
INFO: Run "set stdout_level DEBUG" before import for more information.
INFO: Change particles name to pass to MG5 convention
Defined multiparticle p = g u c d s u~ c~ d~ s~
Defined multiparticle j = g u c d s u~ c~ d~ s~
Defined multiparticle l+ = e+ mu+
Defined multiparticle l- = e- mu-
Defined multiparticle vl = ve vm vt
Defined multiparticle vl~ = ve~ vm~ vt~
mg5>generate p p > t t~
INFO: Checking for minimal orders which gives processes.
INFO: Please specify coupling orders to bypass this step.
INFO: Trying coupling order WEIGHTED=2
INFO: Trying process: g g > t t~ WEIGHTED=2
INFO: Process has 3 diagrams
INFO: Trying process: u u~ > t t~ WEIGHTED=2
INFO: Process has 1 diagrams
INFO: Trying process: u c~ > t t~ WEIGHTED=2
INFO: Trying process: c u~ > t t~ WEIGHTED=2
INFO: Trying process: c c~ > t t~ WEIGHTED=2
INFO: Process has 1 diagrams
INFO: Trying process: d d~ > t t~ WEIGHTED=2
INFO: Process has 1 diagrams
INFO: Trying process: d s~ > t t~ WEIGHTED=2
INFO: Trying process: s d~ > t t~ WEIGHTED=2
INFO: Trying process: s s~ > t t~ WEIGHTED=2
INFO: Process has 1 diagrams
INFO: Process u~ u > t t~ added to mirror process u u~ > t t~
INFO: Process c~ c > t t~ added to mirror process c c~ > t t~
INFO: Process d~ d > t t~ added to mirror process d d~ > t t~
INFO: Process s~ s > t t~ added to mirror process s s~ > t t~
5 processes with 7 diagrams generated in 0.087 s
Total: 5 processes with 7 diagrams
mg5>output test123456
INFO: initialize a new directory: test123456
INFO: remove old information in test123456
INFO: Organizing processes into subprocess groups
INFO: Generating Helas calls for process: g g > t t~ WEIGHTED=2
INFO: Processing color information for process: g g > t t~ WEIGHTED=2
INFO: Generating Helas calls for process: u u~ > t t~ WEIGHTED=2
INFO: Processing color information for process: u u~ > t t~ WEIGHTED=2
INFO: Generating Helas calls for process: c c~ > t t~ WEIGHTED=2
INFO: Combining process with u u~ > t t~ WEIGHTED=2
INFO: Generating Helas calls for process: d d~ > t t~ WEIGHTED=2
INFO: Combining process with u u~ > t t~ WEIGHTED=2
INFO: Generating Helas calls for process: s s~ > t t~ WEIGHTED=2
INFO: Combining process with u u~ > t t~ WEIGHTED=2
INFO: Creating files in directory P0_gg_ttx
INFO: Generating Feynman diagrams for Process: g g > t t~ WEIGHTED=2
INFO: Finding symmetric diagrams for subprocess group gg_ttx
INFO: Creating files in directory P0_qq_ttx
INFO: Generating Feynman diagrams for Process: u u~ > t t~ WEIGHTED=2
INFO: Finding symmetric diagrams for subprocess group qq_ttx
History written to /cms/data27/jaevans/MG5_v1_4_5/test123456/Cards/proc_card_mg5.dat
Generated helas calls for 2 subprocesses (4 diagrams) in 0.030 s
Wrote files for 16 helas calls in 1.924 s
Export UFO model to MG4 format
ALOHA: aloha creates FFV1 routines
ALOHA: aloha creates VVV1 routines
save configuration file to /cms/data27/jaevans/MG5_v1_4_5/test123456/Cards/me5_configuration.txt
INFO: Use Fortran compiler gfortran
INFO: Generate jpeg diagrams
INFO: Generate web pages
Output to directory /cms/data27/jaevans/MG5_v1_4_5/test123456 done.
Type "launch" to generate events from this process, or see
/cms/data27/jaevans/MG5_v1_4_5/test123456/README
Run "open index.html" to see more information about this process.
mg5>launch
************************************************************
* *
* W E L C O M E to M A D G R A P H 5 *
* M A D E V E N T *
* *
* * * *
* * * * * *
* * * * * 5 * * * * *
* * * * * *
* * * *
* *
* VERSION 1.4.5 2012-04-20 *
* *
* The MadGraph Development Team - Please visit us at *
* https://server06.fynu.ucl.ac.be/projects/madgraph *
* *
* Type 'help' for in-line help. *
* *
************************************************************
load configuration from test123456/Cards/me5_configuration.txt
Using default text editor "vi". Set another one in ./input/mg5_configuration.txt
generate_events run_01
Will run in mode parton
Do you want to edit one cards (press enter to bypass editing)?
  1 / param : param_card.dat (be carefull about parameter consistency, especially widths)
  2 / run : run_card.dat
  Path to a valid card.
 [0, done, 1, param, 2, run, enter path][20s to answer]
0
Generating 10000 events with run name run_01
survey run_01
compile directory
gen_ximprove.f: In program `gen_ximprove':
gen_ximprove.f:84:
            call get_integer(npara,param,value," gevents " ,nreq ,2000
                 1 2
Invalid token at (2) in expression or subexpression at (1)
gen_ximprove.f:89:
            write(*,*) "Running on Grid to generate ",nreq," additional eve
                                                           1 2
Character constant at (1) has no closing quote at (2) [info -f g77 M LEX]
gen_ximprove.f: In subroutine `write_bash':
gen_ximprove.f:239:
         double precision xsec(lmaxconfigs), xerru(lmaxconfigs),xerrc(lmaxc
                                                                      ^
Invalid declaration of or reference to symbol `lmaxc' at (^) [initially seen at (^)]
gen_ximprove.f:239:
         double precision xsec(lmaxconfigs), xerru(lmaxconfigs),xerrc(lmaxc
                                                                           ^
Invalid form for type-declaration statement at (^)
gen_ximprove.f:217:
         subroutine write_bash(xsec,xerru,xerrc,xtot,
                                          1
gen_ximprove.f:259: (continued):
            xt(j)= sqrt((xerru(j)+xerrc(j)**2)*mfact(j)) !sort by error
                                  2
Invalid declaration of or reference to symbol `xerrc' at (2) [initially seen at (1)]
gen_ximprove.f:197: warning:
            call write_bash(xsec,xerru,xerrc,xtot,mfact,err_goal,
                 1
gen_ximprove.f:217: (continued):
         subroutine write_bash(xsec,xerru,xerrc,xtot,
                    2
Argument #3 (named `xerrc') of `write_bash' is one type at (2) but is some other type at (1) [info -f g77 M GLOBALS]
gen_ximprove.f: In subroutine `write_gen':
gen_ximprove.f:488:
         double precision goal_lum, xlum(lmaxconfigs), xsec(lmaxconfigs),xt
                                                                         1
gen_ximprove.f:497: (continued):
         double precision xt(lmaxconfigs+1),elimit
                          2
Invalid declaration of or reference to symbol `xt' at (2) [initially seen at (1)]
gen_ximprove.f:514:
         goal_lum = goal_lum/(xtot*1000) !Goal luminosity in fb^-1
         ^
Invalid form for GO TO statement at (^)
gen_ximprove.f:270: warning:
         call sort2(xt,io,ng)
              1
gen_ximprove.f:537: (continued):
         call sort2(xt,io,ng)
              2
Argument #1 of `sort2' is one type at (2) but is some other type at (1) [info -f g77 M GLOBALS]
gen_ximprove.f:549:
               write(*,*) 'Improving for error ',kk,gn(io(kk)),xt(kk),xsec(
                                                                      1 2
Null element at (1) for array reference at (2)
gen_ximprove.f:549:
               write(*,*) 'Improving for error ',kk,gn(io(kk)),xt(kk),xsec(
                                                                      1 2
Invalid token at (2) in expression or subexpression at (1)
gen_ximprove.f:572:
         npfile = 0
         ^
Invalid declaration of or reference to symbol `npfile' at (^) [initially seen at (^)]
gen_ximprove.f:592:
               write(*,*) 'Error in gen_ximprove.f, too many events request
                          1 2
Character constant at (1) has no closing apostrophe at (2)
gen_ximprove.f:466:
         subroutine write_gen(goal_lum,ng,jpoints,gn,xlum,xtot,mfact,xsec,
                                                          1
gen_ximprove.f:727: (continued):
         end
         2
Invalid declaration of or reference to symbol `xtot' at (2) [initially seen at (1)]
gen_ximprove.f:208: warning:
               call write_gen(err_goal,i,nevents,gname,xlum,xtot,mfact,
                    1
gen_ximprove.f:466: (continued):
         subroutine write_gen(goal_lum,ng,jpoints,gn,xlum,xtot,mfact,xsec,
                    2
Argument #6 (named `xtot') of `write_gen' is one type at (2) but is some other type at (1) [info -f g77 M GLOBALS]
gen_ximprove.f:549: confused by earlier errors, bailing out
make: *** [gen_ximprove.o] Error 1
Error make ../bin/internal/gen_ximprove in Source not successful
Command "generate_events run_01" interrupted with error:
MadEventError : Impossible to compile
Please report this bug on https://bugs.launchpad.net/madgraph5
More information is found in '/cms/data27/jaevans/MG5_v1_4_5/test123456/run_01_tag_1_debug.log'.
Please attach this file to your report.

quit

mg5>Xlib: connection to "localhost:12.0" refused by server
Xlib: Invalid MIT-MAGIC-COOKIE-1 key
Xlib: connection to "localhost:12.0" refused by server
Xlib: Invalid MIT-MAGIC-COOKIE-1 key
Error: cannot open display: localhost:12.0
Xlib: connection to "localhost:12.0" refused by server
Xlib: Invalid MIT-MAGIC-COOKIE-1 key
Traceback (most recent call last):
writting history and quit on KeyboardInterrupt
  File "bin/mg5", line 58, in <module>
    subprocess.call([sys.executable] + ['-O'] + sys.argv)
  File "/cms/base/python-2.7.1/lib/python2.7/subprocess.py", line 486, in call

    return Popen(*popenargs, **kwargs).wait()
  File "/cms/base/python-2.7.1/lib/python2.7/subprocess.py", line 1243, in wait
    pid, sts = _eintr_retry_call(os.waitpid, self.pid, 0)
  File "/cms/base/python-2.7.1/lib/python2.7/subprocess.py", line 471, in _eintr_retry_call
    return func(*args)
KeyboardInterrupt

Here is the debug log

[jaevans@hexcms test123456]$ cat run_01_tag_1_debug.log
#************************************************************
#* MadGraph/MadEvent 5 *
#* *
#* * * *
#* * * * * *
#* * * * * 5 * * * * *
#* * * * * *
#* * * *
#* *
#* *
#* VERSION 1.4.5 2012-04-20 *
#* *
#* The MadGraph Development Team - Please visit us at *
#* https://server06.fynu.ucl.ac.be/projects/madgraph *
#* *
#************************************************************
#* *
#* Command File for MadEvent *
#* *
#* run as ./bin/madevent.py filename *
#* *
#************************************************************
generate_events run_01
Traceback (most recent call last):
  File "/cms/data27/jaevans/MG5_v1_4_5/madgraph/interface/extended_cmd.py", line 548, in onecmd
    return cmd.Cmd.onecmd(self, line)
  File "/cms/base/python-2.7.1/lib/python2.7/cmd.py", line 219, in onecmd
    return func(arg)
  File "/cms/data27/jaevans/MG5_v1_4_5/madgraph/interface/madevent_interface.py", line 1712, in do_generate_events
    postcmd=False)
  File "/cms/data27/jaevans/MG5_v1_4_5/madgraph/interface/extended_cmd.py", line 586, in exec_cmd
    stop = cmd.Cmd.onecmd(current_interface, line)
  File "/cms/base/python-2.7.1/lib/python2.7/cmd.py", line 219, in onecmd
    return func(arg)
  File "/cms/data27/jaevans/MG5_v1_4_5/madgraph/interface/madevent_interface.py", line 1965, in do_survey
    self.configure_directory()
  File "/cms/data27/jaevans/MG5_v1_4_5/madgraph/interface/madevent_interface.py", line 3067, in configure_directory
    raise MadEventError, 'Impossible to compile'
MadEventError: Impossible to compile
Value of current Options:
              web_browser : None
              text_editor : None
          pythia-pgs_path :
                  td_path :
             delphes_path :
             cluster_type : condor
         madanalysis_path :
            cluster_queue : madgraph
       group_subprocesses : Auto
         fortran_compiler : None
                  nb_core : 8
      exrootanalysis_path :
               eps_viewer : None
                  timeout : 20
   automatic_html_opening : False
             cluster_mode : 0
             pythia8_path :
ignore_six_quark_processes : False
                 run_mode : 0

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Johan Alwall
Solved:
Last query:
Last reply:
Revision history for this message
Johan Alwall (johan-alwall) said : 		#1

Hello Jared,

I haven't seen anything like this, but from your log I have the following observations:
1. it should not have anything to do with the number of times you have generated processes before
2. it looks like it is some problem with the fortran compiler

Is it possible that the fortran compiler somehow changed between the runs? Note that the compiler is set by the environment variable FC - if this variable is not set, then the default compiler is defined in either the setup script (input/mg5_configuration.txt or Cards/me5_configuration.txt) or in the file Template/Source/make_opts (and then Source/make_opts in each process directory). By default we use gfortran. It looks like your compiler has problems with the use of " instead of ' in string definitions, which indicates some very old compiler, perhaps some old version of g77.

Also note that if you want to run the same process with different model cards and/or cuts, you don't need to regenerate the process directory, but can reuse the same process directory to generate processes in serial. It is only if you want to actually change the process that you need to rerun bin/mg5.

All the best,
Johan

Revision history for this message
Jared Evans (jaevans) said : 		#2

Hello Johan,

Thanks for your reply.

FC is undefined. Running diff mg5_configuration.txt OLDmg5_configuration.txt shows the files are identical (the two files being the one in the functional directory and the one in the non-functional directory). The relevant lines are likely:

# Prefered Fortran Compiler
# If None: try to find g77 or gfortran on the system
fortran_compiler = None

I tried changing "None" to "gfortran" and got the same errors.

However, Template/Source/make_opts is empty! Somehow in generation this appear to be overwritten by something. My scripts do not access this folder directly and no other folder in Source becomes empty. Might the be overwritten for some reason? I'll try removing my write permissions to that file and see if it prevents it.

Just to be clear, what I am doing is parallelizing the event generation for entirely different processes. I throw O(hundred) different processes at a server at once, it is very hands off for me. At for quite a while everything runs fine, eventually, every process will start failing with the same error (I guess that is when make_opts is vacated). If I try a "hands-on" generation after this, that will fail as well. When I move everything over to a new MG5 directory, and start things up again it works perfectly fine (until it fails again). In particular, it runs the processes that had failed before without any issue.

Thanks,
Jared

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#3

Hi Jared,

Which compiler is used for the process which are working?

> However, Template/Source/make_opts is empty! Somehow in generation
> this
> appear to be overwritten by something.

Do you have a huge number of mg5 session working at the same time?
mg5 modifies indeed this file and if two session of mg5 tries to
modify it in the same times,
This might (maybe) creates some troubles and fully erase it (that's
just a crazy hyppothesis to try to explain the problem)

Could you send to us the script/command that you use to generate all
those processes.
This might give us a hint of what the problem is and will help us to
see if this is due to concurent version of mg5 trying to modifies the
same file.

Cheers,

Olivier

On May 3, 2012, at 2:30 PM, Jared Evans wrote:

> Question #195769 on MadGraph5 changed:
> https://answers.launchpad.net/madgraph5/+question/195769
>
> Status: Answered => Open
>
> Jared Evans is still having a problem:
> Hello Johan,
>
> Thanks for your reply.
>
> FC is undefined. Running diff mg5_configuration.txt
> OLDmg5_configuration.txt shows the files are identical (the two files
> being the one in the functional directory and the one in the non-
> functional directory). The relevant lines are likely:
>
> # Prefered Fortran Compiler
> # If None: try to find g77 or gfortran on the system
> fortran_compiler = None
>
> I tried changing "None" to "gfortran" and got the same errors.
>
> However, Template/Source/make_opts is empty! Somehow in generation
> this
> appear to be overwritten by something. My scripts do not access this
> folder directly and no other folder in Source becomes empty. Might
> the
> be overwritten for some reason? I'll try removing my write
> permissions
> to that file and see if it prevents it.
>
> Just to be clear, what I am doing is parallelizing the event
> generation
> for entirely different processes. I throw O(hundred) different
> processes at a server at once, it is very hands off for me. At for
> quite a while everything runs fine, eventually, every process will
> start
> failing with the same error (I guess that is when make_opts is
> vacated).
> If I try a "hands-on" generation after this, that will fail as well.
> When I move everything over to a new MG5 directory, and start things
> up
> again it works perfectly fine (until it fails again). In
> particular, it
> runs the processes that had failed before without any issue.
>
> Thanks,
> Jared
>
> --
> You received this question notification because you are a member of
> MadTeam, which is an answer contact for MadGraph5.

Revision history for this message
Jared Evans (jaevans) said : 		#4

Hi Olivier,

Sorry for the delayed reply! If by compiler, you mean python, then it is python-2.7.1. For fortran I don't set it explicitly, so I assume it uses gfortran.

 gfortran -v
Using built-in specs.
Target: x86_64-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --disable-plugin --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=x86_64-redhat-linux
Thread model: posix
gcc version 4.1.2 20080704 (Red Hat 4.1.2-50)

Yes, I may have O(100) mg5 sessions running at once. As I mentioned, this usually works fine... until it doesn't!

I'll give a truncated version of the scripts (because I do a lot of things to the files after using MG5 that have nothing to do with running MG5 or the Template folder). The most relevant thing is probably the input file handed to MG5 (the last of the 4 files). I am using a FeynRules generated RPVMSSM_UFO, but I'd be rather surprised if this is causing any problems though.

#
# SCRIPT 1:
#

#!/bin/bash
#
# Pass form to GREP, passes all to a jdl file.

BASEJDL=runMG5command.jdl.base
MG5DIR=`pwd`

# used to select particular input files to use
if [ "$2" != "" ]
then
  filelist=`ls | grep $1 | grep $2`
else
  filelist=`ls | grep $1`
fi
for filename in $filelist
do
echo $filename
filebase=${filename##input}
NEWJDL=runMG5${filebase}.jdl
sed -e "s/XXX/${filebase}/" <$BASEJDL >$NEWJDL
if [[ -d ${filebase} ]]; then
mv $filebase RPV/$filebase
fi
condor_submit $NEWJDL
done

#
# Example JDL file
#
universe = vanilla
+AccountingGroup = "jaevans"
Executable = /cms/data27/jaevans/MG5_v1_4_5/MG5RUNNER.sh
should_transfer_files = NO
Arguments = winoLLE233_M500
Output = /cms/data27/jaevans/condor/winoLLE233_M500.out
Error = /cms/data27/jaevans/condor/winoLLE233_M500.err
Log = /cms/data27/jaevans/condor/script.condor
Queue 1

#
# MG5RUNNER.sh
#

#!/bin/bash

export VO_CMS_SW_DIR="/cms/base/cmssoft"
export COIN_FULL_INDIRECT_RENDERING=1
export SCRAM_ARCH="slc5_amd64_gcc434"
source $VO_CMS_SW_DIR/cmsset_default.sh

export PYTHIA8DATA="/cms/data27/jaevans/pythia8160/xmldoc"
export LD_LIBRARY_PATH="/cms/data27/jaevans/LHAPDF/lib"

PYTHONPATH=/cms/base/python-2.7.1/bin/python
MG5PATH=/cms/data27/jaevans/MG5_v1_4_5
RUNMG5PATH=$MG5PATH/bin/mg5
RPVDIR=$MG5PATH/RPV
FOLDERNAME=$1
INPUTFILENAME=input$FOLDERNAME
MASSVAL=`echo $FOLDERNAME | sed 's/[a-zA-Z0-9]*_M\([0-9]*\)[A-Za-z]*$/\1/'`

echo "Moving to $MG5PATH"
cd $MG5PATH

if [ ! -d $RPVDIR/$FOLDERNAME ]; then
echo "Running MG5"
$PYTHONPATH $RUNMG5PATH $MG5PATH/$INPUTFILENAME

echo "Moving $FOLDERNAME to RPV/${FOLDERNAME}"
if [ -d $MG5PATH/$FOLDERNAME ]; then
mv $MG5PATH/$INPUTFILENAME $MG5PATH/$FOLDERNAME/$INPUTFILENAME
mv $MG5PATH/$FOLDERNAME $RPVDIR/$FOLDERNAME
fi
fi

#
#.... a bunch of other stuff enters here, but this neither calls MG5 nor would ever access the Template folder.
#

#
# INPUT FILE
# this file is the one passed to MG5 for running
#

import model RPVMSSM_UFO
generate ul > t mu+ vt~ ta- / go n1 n3 n4 x2+ x2-
add process ul > t vm~ ta+ ta- / go n1 n3 n4 x2+ x2-
add process ul > t mu- vt ta+ / go n1 n3 n4 x2+ x2-
add process ul > t vm ta- ta+ / go n1 n3 n4 x2+ x2-
add process ul > b mu+ ta+ ta- / go n1 n3 n4 x2+ x2-
add process ul > b vm vt ta+ / go n1 n3 n4 x2+ x2-
output winoLLE233_M500
launch -i
!sed -e 's/1000002 3.00/1000002 5.00/' -e '/LLE233/ s/0.000000e-09/1.000000e-01/' -e '/LLE323/ s/0.000000e-09/1.000000e-01/' <param_card_base.dat >winoLLE233_M500/Cards/param_card.dat
survey --accuracy=0.005 --points=5000 --iterations=15
refine 100000
combine_events
store_events
exit

#
# end
#

Survey and refine were modified so that I could generate more than just a handful of events that tend to come with the 4-body phase space. (On that note, these numbers were mostly chosen through trial and error, I am not sure if there is a better way to optimize this for sampling phase space accurately for the number of events I want.)

Thank you for your help!
Jared

Revision history for this message
Best Johan Alwall (johan-alwall) said : 		#5

Hello Jared,

The problem is clearly that several instances are trying to write to the make_opts file in Template/Source at the same time. I would suggest that you simply comment out lines 753-754 (using "#" at the start of the lines) in madgraph/iolibs/export_v4.py:

        # Replace also for Template but not for cluster
        if not os.environ.has_key('MADGRAPH_DATA'):
            self.replace_make_opt_compiler(compiler, os.path.join(MG5DIR, 'Template'))

All the best,
Johan

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#6

HI Jared,

Johan was faster than me on this. :-)

But I am worry about another point in your script.
I don't see any point where you change the seed.
This might induce a correlation between all your run since they will
all use the same random number.
I would suggest to update the seed as well to be 100% sure that you
don't have any correlation between all your run.

Cheers,

Olivier

On 07-mai-12, at 23:40, Johan Alwall wrote:

> Question #195769 on MadGraph5 changed:
> https://answers.launchpad.net/madgraph5/+question/195769
>
> Johan Alwall proposed the following answer:
> Hello Jared,
>
> The problem is clearly that several instances are trying to write to
> the
> make_opts file in Template/Source at the same time. I would suggest
> that
> you simply comment out lines 753-754 (using "#" at the start of the
> lines) in madgraph/iolibs/export_v4.py:
>
> # Replace also for Template but not for cluster
> if not os.environ.has_key('MADGRAPH_DATA'):
> self.replace_make_opt_compiler(compiler,
> os.path.join(MG5DIR, 'Template'))
>
> All the best,
> Johan
>
> --
> You received this question notification because you are a member of
> MadTeam, which is an answer contact for MadGraph5.

Revision history for this message
Jared Evans (jaevans) said : 		#7

Thanks Johan Alwall, that solved my question.

Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#8

Hi Tim,

> 1.I install the new MadAnalysis in MadGraph Home page ( Is it
> MadAnalysis5? )

No its not.

I put Benjamin (and all the MA5 author in cc) such that they can
communicate with you.
[ma5team__A_T__iphc.cnrs.fr]
Maybe you can contact them directly on their email.

Cheers,

Olivier

Begin forwarded message:

> From: Tim Lu <email address hidden>
> Date: May 16, 2012 4:05:52 AM CDT
> To: <email address hidden>
> Subject: Re: [Question #197045]: problem about MG5 output
> Reply-To: <email address hidden>
>
> Question #197045 on MadGraph5 changed:
> https://answers.launchpad.net/madgraph5/+question/197045
>
> Status: Answered => Open
>
> Tim Lu is still having a problem:
> Dear Professor Olivier,
> Thank for your suggestions ! ! !
> 1.I install the new MadAnalysis in MadGraph Home page ( Is it
> MadAnalysis5? )
> But I don't find "Benj " .Where can I find it ?
>
> 2.If I want to do analysis by myself using fortran,I need to read
> the data of .lhe file
> But the form of .lhe file is difficult to separate character and
> number
> </init>
> <event>
> 12 0 0.1298238E-03 0.1958641E+03 0.7957747E-01 0.1159402E+00
> 1 -1 0 0 502 0 0.00000000000E+00 0.00000000000E
> +00 0.42608335559E+03 0.42608335559E+03 ...
> How can I do to get rid of </init> , <event> and separate
> 12 0 0.1298238E-03 0.1958641E+03 0.7957747E-01 0.1159402E+00
> and
> 1 -1 0 0 502 0 0.00000000000E+00 0.00000000000E+00
> 0.42608335559E+03 0.42608335559E+03 ...
>
> 3.If I want to combine two .top file by using MadAnalysis,we can use
> combine-pl ,but it can just deal with the same quantity ,if I
> want to
> combine two .top file with different quantities (like y(1) and y(2))
> ,how can I do ?
>
> Thank you very much ! ! !
>
> Best,
> Tim Lu
>
> --
> You received this question notification because you are a member of
> MadTeam, which is an answer contact for MadGraph5.