#2 closed defect (worksforme)
Sherpa 1.0.9 crashes when including multiple interactions
| Reported by: | Markus.Merschmeyer@physik.rwth-aachen.de | Owned by: | support@sherpa-mc.de |
|---|---|---|---|
| Priority: | minor | Milestone: | |
| Component: | Unknown | Version: | 1.0.9 |
| Keywords: | Multiple Interactions | Cc: | Markus.Merschmeyer@physik.rwth-aachen.de |
Description
I start from Sherpa 1.0.9 with all available patches included. My Environment is 'scl4_ia32_gcc345' and CLHEP was included in the compilation to generate output readable by the CMS simulation and analysis software. I tried to generate jet+jet-> mu+nu events with multiple interactions switched on and Sherpa crashes AFTER generating the 100 events I wanted. This does not occur when multiple interactions are switched off. (The Sherpa output and Status directory are attached to this ticket).
Attachments (2)
Change History (7)
Changed 17 years ago by
| Attachment: | Sherpa_output.txt added |
|---|
Changed 17 years ago by
| Attachment: | Status__Fri_Jul_20_12-26-44_2007.tgz added |
|---|
comment:1 Changed 17 years ago by
comment:2 Changed 17 years ago by
| Resolution: | → worksforme |
|---|---|
| Status: | new → closed |
Markus Merschmeyer, Friday 20 July 2007:
good point. Unfortunately this was not the cause for the crash. I recompiled with the original Main.C, everything else was left as before. The same error occured, the new output and status files are attached to this mail.
I cannot reproduce this crash. With your setup I can generate 100 events and Sherpa finishes properly (this is on an x86_64 machine, gcc 4.0.2). Since it only crashes after the event generation has finished, I'd think that you can simply ignore the crash. The HepMC files should have been written out correctly.
Something else I noticed when testing your setup: The given charge conservation warnings are valid, two of our hadron decay channels have the wrong final state (K_S -> pi+ nu_e e+ and K_S -> pi- anti-nu_e e-). I'd suggest turning off all decayers but the tau in our Decaydata/HadronDecays.dat (if already installed, also in share/SHERPA-MC/Decaydata/HadronDecays.dat). This has the effect, that everything but the tau will be decayed by Pythia. Our own hadron decay module is still work-in-progress, and only the tau decays are really tested well.
Cheers, Frank
comment:3 Changed 17 years ago by
Maybe I should have added, that the charge conservation problem will be fixed in the next release.
comment:8 Changed 12 years ago by
| Milestone: | → rel-old |
|---|
Hi Markus,
I see that you are using a modified Main.C. Have you tried the same run with a non-modified Sherpa? It seems like some pointers get messed up or deleted twice, and I'd like to make sure, that this doesn't come from your changes.
Cheers, Frank