ue_enabled bug

[Michele Selvaggi]

Hello,

I am using MG5_aMC_2_2_3, and generate p p > t t~ [QCD], but then put the flag to LO.
I am simply changing the shower_card to set "ue_enabled = true" and I get the follwing error (*):
I have attached the debug.log file.

Thanks in advance for help.

Michele

(*)

      Summary:
      Process p p > t t~ [QCD]
      Run at p-p collider (6500 + 6500 GeV)
      Total cross-section: 4.610e+02 +- 1.5e+00 pb
      Ren. and fac. scale uncertainty: +29.8% -22.3%
      Number of events generated: 10000
      Parton shower to be used: HERWIG6
      Fraction of negative weights: 0.00
      Total running time : 2m 59s

INFO: The /nfs/user/mselvaggi/MG5_aMC_v2_2_2/test/Events/run_01_LO/events.lhe.gz file has been generated.

INFO: Events generated
decay_events -from_cards
INFO: Preparing MCatNLO run
INFO: Compiling MCatNLO for HERWIG6...
INFO: ... done
INFO: Showering events...
INFO: (Running in /nfs/user/mselvaggi/MG5_aMC_v2_2_2/test/MCatNLO/RUN_HERWIG6_1)
INFO: Idle: 0, Running: 1, Completed: 0 [ current time: 12h57 ]
mv: cannot stat `events.lhe.hep': No such file or directory
gzip: events.hep: No such file or directory
INFO: Idle: 0, Running: 0, Completed: 1 [ 0.2s ]
INFO: Idle: 0, Running: 0, Completed: 0 [ current time: 12h57 ]
Command "launch auto " interrupted with error:
IOError : [Errno 2] No such file or directory: '/nfs/user/mselvaggi/MG5_aMC_v2_2_2/test/MCatNLO/RUN_HERWIG6_1/events.hep.gz'
Please report this bug on https://bugs.launchpad.net/madgraph5
More information is found in '/nfs/user/mselvaggi/MG5_aMC_v2_2_2/test/run_01_LO_tag_1_debug.log'.
Please attach this file to your report.
INFO:

quit
INFO:

[Paolo Torrielli (paolo-torrielli)]

Ciao Michele,
this is due to the fact that when using the UE, Herwig needs particular PDF sets,
like for example 20651.

Your file RUN_HERWIG6_1/mcatnlo_run.log should end with

 Underlying event requires external PDF sets
 Please set PDFCODE != 0 in shower_card.dat

Now, I guess the reason why I forced the code to stop when PDFCODE=0 is that
I wanted the user to think about the fact that Herwig needs such PDFs with the
UE, but actually the correct PDF set is anyway forced in the code in this case, so
I'll remove this stop statement right away in version 2.3.

As far as you are concerned, just set PDFCODE = 1 in the card, and it will go smooth
(with PDF set = 20651).

Cheers.
Paolo

[Paolo Torrielli (paolo-torrielli)]

Hi Michele,
I have thought a bit more about your bug report.

The solution is still what I wrote in the last lines of my previous comment, but I've
realised that there was a deeper reason why the code should stop if you use HERWIG
+UE+ PDFCODE=0: namely in order to force the use of some specific PDF set, like
20651, the interface needs to know before showering that you're gonna use the
'external pdf' mode. You instruct the code in this way precisely by setting PDFCODE
!=0 in the shower card.
Thus it is necessary that the code stops if the user chosen UE+PDFCODE=0 (for HERWIG)
and that the user reads the error message printed in mcatnlo_run.log, and resets
PDFCODE in the card.

Cheers!
Paolo