- Table of contents
- Simulation trouble-shooting
Seg fault during sim startup when using visualisation (e.g. writing HepRepXML for paraview files)¶
If running the simulation with artg4 enableVisualization set to true (e.g. writing HepRepXML for paraview files) and you have a seg fault, e.g.
Hadronic Processes for <proton> ----------------------------------- hadElastic Models: hElasticCHIPS: Emin(GeV)= 0 Emax(GeV)= 100000 hadElastic Crs sctns: ChipsProtonElasticXS: Emin(GeV)= 0 Emax(GeV)= 100000 GheishaElastic: Emin(GeV)= 0 Emax(GeV)= 100000 ProtonInelastic Models: FTFP: Emin(GeV)= 4 Emax(GeV)= 100000 BertiniCascade: Emin(GeV)= 0 Emax(GeV)= 5 ProtonInelastic Crs sctns: Barashenkov-Glauber: Emin(GeV)= 0 Emax(GeV)= 100000 GheishaInelastic: Emin(GeV)= 0 Emax(GeV)= 100000 ============================================================================================ Segmentation fault (core dumped)
This may be because the paths in macroPath don't find your vicMaco (e.g. vis.mac). By default they are relative to the location of your fcl file, so if you are running from somewhere else (e.g. a data directory) then you get a seg fault if it can't find the file.
To fix it change the macroPath or run from the same directory as your fcl file.
TODO We should make artg4 report this error better (e.g. fill not found rather than seg fault).
Missing particle attributes in HepRep file¶
An error has been seen before where the output HepRep file contains trajectory points (x,z,y) for particles in events but all other attributes (e.g. name, PDG code, IMag, etc) are missing (or set to default values, e.g. -1 etc).
This turns out to be due to using our own Trajectory class, which was not writing HepRep attributes. This has now been fixed.