MadGraph linking error

[Konrad Tywoniuk]

Hello!

It seems MCatNLO crashes when trying to generate events. The generation crashes whenever gfortran tries to compile things that are related to FastJet -- although, I didn't specify any location for FastJet on my system, so I assume that MCatNLO uses it's internal FastJet packages.

The error message starts like this:

    gfortran -o test_ME symmetry_fks_test_ME.o parton_lum_1.o parton_lum_chooser.o matrix_1.o real_me_chooser.o fks_inc_chooser.o leshouche_inc_chooser.o genps_fks.o setcuts.o setscales.o born.o sborn_sf.o b_sf_001.o b_sf_002.o b_sf_003.o fks_Sij.o trapfpe.o fastjetfortran_madfks_core.o fjcore.o fastjet_wrapper.o fks_singular.o montecarlocounter.o reweight_xsec.o boostwdir2.o configs_and_props_inc_chooser.o initcluster.o cluster.o reweight.o get_color.o FKSParamReader.o MC_integer.o reweight_xsec_events_pdf_dummy.o BinothLHADummy.o cuts.o -L../../lib/ -ldhelas -lgeneric -lmodel -lpdf -lcernlib -lstdc++
    Undefined symbols for architecture x86_64:
      "std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >::compare(char const*) const", referenced from:
          fjcore::ClusterSequence::print_jets_for_root(std::__1::vector<fjcore::PseudoJet, std::__1::allocator<fjcore::PseudoJet> > const&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) const in fjcore.o

and goes on...

I have also enclosed the debug_log file.

I am working on a MacOSX 10.9.2. My gcc is
Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/usr/include/c++/4.2.1
Apple LLVM version 5.1 (clang-503.0.38) (based on LLVM 3.4svn)
Target: x86_64-apple-darwin13.1.0
Thread model: posix

and my gfortran is
GNU Fortran (GCC) 4.8.2

Would appreciate any help!

Best regards,

Konrad

[marco zaro (marco-zaro)]

Dear Konrad,
is your gcc/g++ compiler version the same as gfortran?
Looking at your message it looks like your gcc is 4.2.1, while gfortran is 4.8
If you installed gfortran /gcc 4.8 via macports, you can select the 4.8 version to be the default one with
ort select --list gcc
To select a version use

sudo port select --set gcc gcc4X
Let me know if this solves your issue.
Cheers,

Marco

On 01 Apr 2014, at 23:27, Launchpad Bug Tracker <email address hidden> wrote:

> Olivier Mattelaer (olivier-mattelaer) has assigned this bug to you for MadGraph5_aMC@NLO:
>
> Hello!
>
> It seems MCatNLO crashes when trying to generate events. The generation
> crashes whenever gfortran tries to compile things that are related to
> FastJet -- although, I didn't specify any location for FastJet on my
> system, so I assume that MCatNLO uses it's internal FastJet packages.
>
> The error message starts like this:
>
> gfortran -o test_ME symmetry_fks_test_ME.o parton_lum_1.o parton_lum_chooser.o matrix_1.o real_me_chooser.o fks_inc_chooser.o leshouche_inc_chooser.o genps_fks.o setcuts.o setscales.o born.o sborn_sf.o b_sf_001.o b_sf_002.o b_sf_003.o fks_Sij.o trapfpe.o fastjetfortran_madfks_core.o fjcore.o fastjet_wrapper.o fks_singular.o montecarlocounter.o reweight_xsec.o boostwdir2.o configs_and_props_inc_chooser.o initcluster.o cluster.o reweight.o get_color.o FKSParamReader.o MC_integer.o reweight_xsec_events_pdf_dummy.o BinothLHADummy.o cuts.o -L../../lib/ -ldhelas -lgeneric -lmodel -lpdf -lcernlib -lstdc++
> Undefined symbols for architecture x86_64:
> "std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> >::compare(char const*) const", referenced from:
> fjcore::ClusterSequence::print_jets_for_root(std::__1::vector<fjcore::PseudoJet, std::__1::allocator<fjcore::PseudoJet> > const&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char> > const&) const in fjcore.o
>
> and goes on...
>
> I have also enclosed the debug_log file.
>
> I am working on a MacOSX 10.9.2. My gcc is
> Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/usr/include/c++/4.2.1
> Apple LLVM version 5.1 (clang-503.0.38) (based on LLVM 3.4svn)
> Target: x86_64-apple-darwin13.1.0
> Thread model: posix
>
> and my gfortran is
> GNU Fortran (GCC) 4.8.2
>
> Would appreciate any help!
>
> Best regards,
>
> Konrad
>
> ** Affects: mg5amcnlo
> Importance: Undecided
> Assignee: marco zaro (marco-zaro)
> Status: New
>
> --
> MadGraph linking error
> https://bugs.launchpad.net/bugs/1300888
> You received this bug notification because you are a bug assignee.

[Konrad Tywoniuk (konrad-tywoniuk)]

Dear Marco,

I changed the compiler to be gcc-4.8.2 and gfortran-4.8.2 from Homebrew
both. The fjcore problem seems to be avoided, but now at launch the code
produces the following error:

INFO: Starting run

INFO: Compiling the code

INFO: Using built-in libraries for PDFs

INFO: Compiling source...

INFO: ...done, continuing with P* directories

INFO: Compiling directories...

INFO: Compiling on 4 cores

INFO: Compiling P0_epem_bbx...

INFO: P0_epem_bbx done.

INFO: Checking test output:

INFO: P0_epem_bbx

INFO: Result for test_ME:

INFO: Passed.

INFO: Result for test_MC:

INFO: Passed.

INFO: Result for check_poles:

INFO: Poles successfully cancel for 20 points over 20
(tolerance=1.0e-05)

INFO: Starting run

INFO: Using 4 cores

INFO: Cleaning previous results

INFO: Generating events without running the shower.

INFO: Setting up grid

WARNING: program
/Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/ajob1
2 F 0 launch ends with non zero status: 1. Stop all computation

WARNING: Last 15 lines of logfile
/Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/*/log.txt:

 doing the all of this channel

 Normal integration (Sfunction != 1)

 Not subdividing B.W.

 about to integrate 7 -1 12 1

 imode is 0

STOP 1

 ------- iteration 1

 Update # PS points (even): 560 --> 512

Using random seed offsets: 1 , 1 , 0

  with seed 33

 Ranmar initialization seeds 11949 9408

 PDF not supported for pdg -11

 For lepton colliders, please set the lpp* variables to 0 in the run_card

Time in seconds: 0

INFO: remove job currently running

WARNING: program
/Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/ajob2
2 F 0 launch ends with non zero status: 1. Stop all computation

WARNING: Last 15 lines of logfile
/Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/*/log.txt:

 doing the all of this channel

 Normal integration (Sfunction != 1)

 Not subdividing B.W.

 about to integrate 7 -1 12 1

 imode is 0

STOP 1

 ------- iteration 1

 Update # PS points (even): 560 --> 512

Using random seed offsets: 1 , 1 , 0

  with seed 33

 Ranmar initialization seeds 11949 9408

 PDF not supported for pdg -11

 For lepton colliders, please set the lpp* variables to 0 in the run_card

Time in seconds: 0

INFO: remove job currently running

INFO: remove job currently running

INFO: remove job currently running

Command "launch" interrupted with error:

Exception : program
/Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/ajob2
2 F 0 launch ends with non zero status: 1. Stop all computation

Please report this bug on https://bugs.launchpad.net/madgraph5

More information is found in
'/Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/run_01_tag_1_debug.log'.

Please attach this file to your report.

I also attach the debug file here.

Thanks so far! ;)

...

[marco zaro (marco-zaro)]

Dear Konrad,
so the problem is this
PDF not supported for pdg -11

For lepton colliders, please set the lpp* variables to 0 in the
run_card

as you are doing lepton-lepton collisions, you should tell the code not to use any pdf. This can be done by setting lpp1 and lpp2 to 0 in the run_card.
This should solve your problem.

Cheers,

Marco

On 02 Apr 2014, at 12:11, Konrad Tywoniuk <email address hidden> wrote:

> Dear Marco,
>
> I changed the compiler to be gcc-4.8.2 and gfortran-4.8.2 from Homebrew
> both. The fjcore problem seems to be avoided, but now at launch the code
> produces the following error:
>
> INFO: Starting run
>
> INFO: Compiling the code
>
> INFO: Using built-in libraries for PDFs
>
> INFO: Compiling source...
>
> INFO: ...done, continuing with P* directories
>
> INFO: Compiling directories...
>
> INFO: Compiling on 4 cores
>
> INFO: Compiling P0_epem_bbx...
>
> INFO: P0_epem_bbx done.
>
> INFO: Checking test output:
>
> INFO: P0_epem_bbx
>
> INFO: Result for test_ME:
>
> INFO: Passed.
>
> INFO: Result for test_MC:
>
> INFO: Passed.
>
> INFO: Result for check_poles:
>
> INFO: Poles successfully cancel for 20 points over 20
> (tolerance=1.0e-05)
>
> INFO: Starting run
>
> INFO: Using 4 cores
>
> INFO: Cleaning previous results
>
> INFO: Generating events without running the shower.
>
> INFO: Setting up grid
>
> WARNING: program
> /Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/ajob1
> 2 F 0 launch ends with non zero status: 1. Stop all computation
>
> WARNING: Last 15 lines of logfile
> /Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/*/log.txt:
>
> doing the all of this channel
>
> Normal integration (Sfunction != 1)
>
> Not subdividing B.W.
>
> about to integrate 7 -1 12 1
>
> imode is 0
>
> STOP 1
>
> ------- iteration 1
>
> Update # PS points (even): 560 --> 512
>
> Using random seed offsets: 1 , 1 , 0
>
> with seed 33
>
> Ranmar initialization seeds 11949 9408
>
> PDF not supported for pdg -11
>
> For lepton colliders, please set the lpp* variables to 0 in the
> run_card
>
> Time in seconds: 0
>
>
>
> INFO: remove job currently running
>
> WARNING: program
> /Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/ajob2
> 2 F 0 launch ends with non zero status: 1. Stop all computation
>
> WARNING: Last 15 lines of logfile
> /Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/*/log.txt:
>
> doing the all of this channel
>
> Normal integration (Sfunction != 1)
>
> Not subdividing B.W.
>
> about to integrate 7 -1 12 1
>
> imode is 0
>
> STOP 1
>
> ------- iteration 1
>
> Update # PS points (even): 560 --> 512
>
> Using random seed offsets: 1 , 1 , 0
>
> with seed 33
>
> Ranmar initialization seeds 11949 9408
>
> PDF not supported for pdg -11
>
> For lepton colliders, plea...

[Konrad Tywoniuk (konrad-tywoniuk)]

Dear Marco,

sorry for coming back late to you. Thanks for the help! I tried to run for
pp collisions and it works, so it's the pdf's.

Just as a side remark: I guess if I enter the command "generate e+ e- >
..." the pdf option 0 should automatically be chosen, so I'm still a little
confused why MadGraph produced this "erroneous" run card.

Thanks again, best regards,

Konrad

On Wed, Apr 2, 2014 at 12:25 PM, marco zaro <email address hidden>wrote:

> Dear Konrad,
> so the problem is this
> PDF not supported for pdg -11
>
> For lepton colliders, please set the lpp* variables to 0 in the
> run_card
>
> as you are doing lepton-lepton collisions, you should tell the code not to
> use any pdf. This can be done by setting lpp1 and lpp2 to 0 in the run_card.
> This should solve your problem.
>
> Cheers,
>
> Marco
>
>
> On 02 Apr 2014, at 12:11, Konrad Tywoniuk <email address hidden>
> wrote:
>
> > Dear Marco,
> >
> > I changed the compiler to be gcc-4.8.2 and gfortran-4.8.2 from Homebrew
> > both. The fjcore problem seems to be avoided, but now at launch the code
> > produces the following error:
> >
> > INFO: Starting run
> >
> > INFO: Compiling the code
> >
> > INFO: Using built-in libraries for PDFs
> >
> > INFO: Compiling source...
> >
> > INFO: ...done, continuing with P* directories
> >
> > INFO: Compiling directories...
> >
> > INFO: Compiling on 4 cores
> >
> > INFO: Compiling P0_epem_bbx...
> >
> > INFO: P0_epem_bbx done.
> >
> > INFO: Checking test output:
> >
> > INFO: P0_epem_bbx
> >
> > INFO: Result for test_ME:
> >
> > INFO: Passed.
> >
> > INFO: Result for test_MC:
> >
> > INFO: Passed.
> >
> > INFO: Result for check_poles:
> >
> > INFO: Poles successfully cancel for 20 points over 20
> > (tolerance=1.0e-05)
> >
> > INFO: Starting run
> >
> > INFO: Using 4 cores
> >
> > INFO: Cleaning previous results
> >
> > INFO: Generating events without running the shower.
> >
> > INFO: Setting up grid
> >
> > WARNING: program
> >
> /Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/ajob1
> > 2 F 0 launch ends with non zero status: 1. Stop all computation
> >
> > WARNING: Last 15 lines of logfile
> >
> /Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/*/log.txt:
> >
> > doing the all of this channel
> >
> > Normal integration (Sfunction != 1)
> >
> > Not subdividing B.W.
> >
> > about to integrate 7 -1 12 1
> >
> > imode is 0
> >
> > STOP 1
> >
> > ------- iteration 1
> >
> > Update # PS points (even): 560 --> 512
> >
> > Using random seed offsets: 1 , 1 , 0
> >
> > with seed 33
> >
> > Ranmar initialization seeds 11949 9408
> >
> > PDF not supported for pdg -11
> >
> > For lepton colliders, please set the lpp* variables to 0 in the
> > run_card
> >
> > Time in seconds: 0
> >
> >
> >
> > INFO: remove job currently running
> >
> > WARNING: program
> >
> /Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/ajob2
> > 2 F 0 launch ends with non zero status: 1. Stop all computation
> >
> > WARNING: Last ...

[marco zaro (marco-zaro)]

Dear Konrad,
the option 0 is not automatically chosen, it’s up to the user to change it in the run_card, so the code worked as expected.
I will discuss within the collaboration wether to change the behaviour in order to set it to 0 automatically for ee initiated processes.

Cheers,

Marco

On 10 Apr 2014, at 19:52, Konrad Tywoniuk <email address hidden> wrote:

> Dear Marco,
>
> sorry for coming back late to you. Thanks for the help! I tried to run for
> pp collisions and it works, so it's the pdf's.
>
> Just as a side remark: I guess if I enter the command "generate e+ e- >
> ..." the pdf option 0 should automatically be chosen, so I'm still a little
> confused why MadGraph produced this "erroneous" run card.
>
> Thanks again, best regards,
>
> Konrad
>
>
> On Wed, Apr 2, 2014 at 12:25 PM, marco zaro
> <email address hidden>wrote:
>
>> Dear Konrad,
>> so the problem is this
>> PDF not supported for pdg -11
>>
>> For lepton colliders, please set the lpp* variables to 0 in the
>> run_card
>>
>> as you are doing lepton-lepton collisions, you should tell the code not to
>> use any pdf. This can be done by setting lpp1 and lpp2 to 0 in the run_card.
>> This should solve your problem.
>>
>> Cheers,
>>
>> Marco
>>
>>
>> On 02 Apr 2014, at 12:11, Konrad Tywoniuk <email address hidden>
>> wrote:
>>
>>> Dear Marco,
>>>
>>> I changed the compiler to be gcc-4.8.2 and gfortran-4.8.2 from Homebrew
>>> both. The fjcore problem seems to be avoided, but now at launch the code
>>> produces the following error:
>>>
>>> INFO: Starting run
>>>
>>> INFO: Compiling the code
>>>
>>> INFO: Using built-in libraries for PDFs
>>>
>>> INFO: Compiling source...
>>>
>>> INFO: ...done, continuing with P* directories
>>>
>>> INFO: Compiling directories...
>>>
>>> INFO: Compiling on 4 cores
>>>
>>> INFO: Compiling P0_epem_bbx...
>>>
>>> INFO: P0_epem_bbx done.
>>>
>>> INFO: Checking test output:
>>>
>>> INFO: P0_epem_bbx
>>>
>>> INFO: Result for test_ME:
>>>
>>> INFO: Passed.
>>>
>>> INFO: Result for test_MC:
>>>
>>> INFO: Passed.
>>>
>>> INFO: Result for check_poles:
>>>
>>> INFO: Poles successfully cancel for 20 points over 20
>>> (tolerance=1.0e-05)
>>>
>>> INFO: Starting run
>>>
>>> INFO: Using 4 cores
>>>
>>> INFO: Cleaning previous results
>>>
>>> INFO: Generating events without running the shower.
>>>
>>> INFO: Setting up grid
>>>
>>> WARNING: program
>>>
>> /Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/ajob1
>>> 2 F 0 launch ends with non zero status: 1. Stop all computation
>>>
>>> WARNING: Last 15 lines of logfile
>>>
>> /Users/konradtywoniuk/phys/MG5_aMC_v2_1_1/test_1/SubProcesses/P0_epem_bbx/*/log.txt:
>>>
>>> doing the all of this channel
>>>
>>> Normal integration (Sfunction != 1)
>>>
>>> Not subdividing B.W.
>>>
>>> about to integrate 7 -1 12 1
>>>
>>> imode is 0
>>>
>>> STOP 1
>>>
>>> ------- iteration 1
>>>
>>> Update # PS points (even): 560 --> 512
>>>
>>> Using random seed offsets: 1 , 1 , 0
>>>
>>> with seed 33
>>>
...

[Olivier Mattelaer (olivier-mattelaer)]

Hi,

I don't want to start to game to automatically create a valid run_card.
This is more dangerous than anything else. We have default for some standard process of LHC and this is it.
Any prospective detector should be careful about this file. It is difficult to know which default is the correct one for the pt of a particle or stuff like that.

Cheers,

Olivier