MadGraph5_aMC@NLO

Fortran error during reweighting on four top quark production

Asked by Vichayanun Wachirapusitanand

Hi,

I am currently generating events with four top quark production using MadGraph5 v3.1.1 and the following cards: https://www.dropbox.com/s/pg0tuzg2ijrh2wv/TTTT_5f_NLO.zip?dl=0

Here are the contents of proc_card.dat:

import model loop_sm-no_b_mass

define p = p b b~
define j = j b b~

generate p p > t t~ t t~ [QCD] @0

output TTTT_5f_NLO -nojpeg
launch
MadSpin=ON
/user/vwachira/LCG_98python3/TTTT_5f/TTTT_5f_NLO_run_card.dat
/user/vwachira/LCG_98python3/TTTT_5f/TTTT_5f_NLO_madspin_card.dat
set param_card mass 6 172.5
set param_card yukawa 6 172.5
set param_card mass 25 125.0

After I started running the card with the command:

MG5_aMC_v3_1_1/bin/mg5_aMC TTTT_5f/TTTT_5f_NLO_proc_card.dat

I noticed that MadGraph had completed the compilation, grid setup, upper envelope calculation, and event generation in LHE format. However, it got stuck during reweighting with the following error:

At line 408 of file reweight_xsec_events.f
Fortran runtime error: Bad integer for item 15 in list input

Could you please let me know what could possibly be the cause of this problem? Please let me know if you would like to see more details in my private production.

Regards,
Vichayanun

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
Rikkert Frederix Edit question
Solved by:
Vichayanun Wachirapusitanand
Solved:
Last query:
Last reply:
Revision history for this message
Rikkert Frederix (frederix) said : 		#1

Dear Vichayanun,

It probably means that one of your intermediate LHE event files was corrupted.
You could check the <YourProcess>/SubProcesses/P*/G*/events.lhe files to check if they are all okay. They should all end with a line consisting of"</LesHouchesEvents>" (or something like that). Are there event files that do not have this line?
Are you running on a single machine, or on a computer cluster?

Best regards,
Rikkert

Revision history for this message
Vichayanun Wachirapusitanand (vicha-w) said : 		#2

I'm sorry, I started rerunning the executable as I was waiting your response. Are these files overwritten? I have checked right after the next run finished (with another error) and I saw every LHE file ended with </LesHouchesEvents>.

Revision history for this message
Rikkert Frederix (frederix) said : 		#3

Dear Vichayanun,

Yes, they get overwritten when you start a new run.

You got the same error, or a different error the second time around?

best,
Rikkert

Revision history for this message
Vichayanun Wachirapusitanand (vicha-w) said : 		#4

When I reinitiated a run, I got the following error:

FileNotFoundError: [Errno 2] No such file or directory: '/storage_mnt/storage/user/vwachira/LCG_98python3/TTTT_5f_NLO_noloop/Events/run_04/alllogs_0.html'

By the way, I have just reproduced an error with a separate production and checked for </LesHouchesEvents> at the end of all LHE files. I found that all files ended with the same </LesHouchesEvents>. What should I look for next?

Revision history for this message
Rikkert Frederix (frederix) said : 		#5

Are you running on a single machine, or on a computer cluster?

Revision history for this message
Vichayanun Wachirapusitanand (vicha-w) said : 		#6

I am running this as a multicore process on a computer cluster. I submitted the following in my executable script:

cd /user/vwachira/LCG_98python3

pwd
source /user/vwachira/LCG_98python3/setupLCGview.sh
export LHAPDF_DATA_PATH=/cvmfs/sft.cern.ch/lcg/external/lhapdfsets/current/:/cvmfs/sft.cern.ch/lcg/releases/MCGenerators/lhapdf/6.3.0-73cda/x86_64-centos7-gcc10-opt/share/LHAPDF/

MG5_aMC_v3_1_1/bin/mg5_aMC TTTT_5f/TTTT_5f_NLO_proc_card.dat

Revision history for this message
Rikkert Frederix (frederix) said : 		#7

Dear Vichayanun,

Given that you get a different error every time you run, it looks like that it is related to some instability in the system that you are running on, rather than with the MG5_aMC code itself. In particular, it looks like some jobs might not have finished copying back the relevant files and/or did not finish correctly.

I've got no idea what the commands that you posted to the screen mean, since I'm not familiar with the system that you are working on. I'm afraid I can't help you any further. You might want to talk to your system administrator to check what the reason is that things go amiss.

Best regards,
Rikkert

Revision history for this message
Vichayanun Wachirapusitanand (vicha-w) said : 		#8

Sorry for the radio silence. I have discussed with my colleague about this issue, and it turns out that the datacards can be used in version 2.6.7 of MadGraph without any issues during reweighting. We would need some time to figure out the changes between MadGraph 3.1.1 vs 2.6.7 and see why they do not work on v3.1.1. Thank you for you help anyway. Your answers here are appreciated.