Segmentation Fault (core dumped) with text2workspace.py

[AnYpku]

I am using CMSSW_14_1_0_pre4 with Combine v10 . I met the "Segmentation fault (core dumped)" when I tried to run text2worksapce.py for a combined card with channels from different years. It doesn't have any error information wait a while and "Segmentation fault (core dumped)" appears and the generation stops. I put my data cards in the public directory /afs/cern.ch/user/y/yian/public/data_card

To repeat the issue, the commands are: combineCards.py muon_16pre.txt muon_16post.txt muon_17.txt muon_18.txt > all_muon_run2.txt

text2workspace.py -P HiggsAnalysis.CombinedLimit.PhysicsModel:multiSignalModel --PO verbose --PO 'map=.*/tgammaQ_1:r_1[1,0,5]' --PO 'map=.*/tgammaQ_2:r_2[1,0,5]' --PO 'map=.*/tgammaQ_3:r_3[1,0,5]' --PO 'map=.*/tgammaQ_4:r_4[1,0,5]' --PO 'map=.*/ttg_1:r_ttg1[1,0,5]' --PO 'map=.*/ttg_2:r_ttg2[1,0,5]' --PO 'map=.*/ttg_3:r_ttg3[1,0,5]' --PO 'map=.*/ttg_4:r_ttg4[1,0,5]' all_muon_run2.txt -o ws_all_muon_run2.root

BTW, every single muon_1*.txt card can be transferred to the workspace smoothly. The combine -M MultiDimFit for every singlemuon_1*.txt workspace also looks fine.

Many thanks if you can give me some hints for that. Looking forward to your reply.

[adewit]

Hi, did you run ulimit -s unlimited before executing text2workspace? If not, could you try that and see if that solves the problem?

[AnYpku]

Hi, many thanks! It works if I run ulimit -s unlimited first.