MantisBT

View Issue Details Jump to Notes ] Issue History ] Print ]
IDProjectCategoryView StatusDate SubmittedLast Update
0000094Hall D OfflineGeneralpublic2011-06-20 15:592011-06-30 16:48
Reporterdavidl 
Assigned Todavidl 
PrioritynormalSeverityminorReproducibilityalways
StatusresolvedResolutionfixed 
PlatformOSOS Version
Summary0000094: hdgeant crashes when adding EM background
DescriptionWhen running hdgeant with BGGATE and BGRATE enabled so that additional beam photons will be simulated in addition to the events read from an external file, it seems to crash immediately. It is unclear if this is the first event or if it gets through one or two before the crash. The "10 events" message does not get printed prior to the crash.

Problem was not noticed for data set with very weak thrown photon that doesn't escape target.
Additional Information---- Snippet from log file ------

 !!!!! ZFATAL called from MZPUSH

 !!!!! ZFATAL reached from MZPUSH for Case= 3

          IQUEST(11) = 1186631 121B47
          IQUEST(12) = ********* 454E494B KINE
          IQUEST(13) = 64000 FA00
          IQUEST(14) = 64000 FA00
          IQUEST(15) = 1 1
          IQUEST(16) = 0 0
          IQUEST(17) = 10 A
          IQUEST(18) = 0 0

          Current Store number = 0 (JQDIVI= 2)
1ZEBRA SYSTEM Post-Mortem from ZPOSTM.
TagsNo tags attached.
Attached Files

- Relationships

-  Notes
(0000133)
jonesrt (developer)
2011-06-28 21:20

I am unable to reproduce the problem, starting from a clean checkout and build of Rev. 7915. Can you post a complete copy of your control.in file that you used to create the error?
(0000134)
jonesrt (developer)
2011-06-29 05:54

Need more detailed instructions on how to reproduce. Cannot reproduce the error, given the information provided.
(0000137)
davidl (administrator)
2011-06-30 16:28

I am able to successfully run the executable that is in Richard's area of the scratch disk, but still get crashes when running an executable I built.

I have tried building my own version of cernlib using the gridmake script from Richard's area the scratch disk and still get crashes.

I have tried linking to Richard's cernlib binaries on the scratch disk and still get crashes.

It seems the crashes are:
1.) not directly related to the cernlib version
2.) not due to settings in the environment
(0000138)
davidl (administrator)
2011-06-30 16:48

This turned out to be due to a stale hdds version.

The problem was traced by systematically linking against packages in Richard's packages directory on the scratch disk (cernlib, sim-recon, hdds) until a working executable was obtained.

- Issue History
Date Modified Username Field Change
2011-06-20 15:59 davidl New Issue
2011-06-20 15:59 davidl Status new => assigned
2011-06-20 15:59 davidl Assigned To => jonesrt
2011-06-28 21:20 jonesrt Note Added: 0000133
2011-06-29 05:54 jonesrt Note Added: 0000134
2011-06-29 05:54 jonesrt Status assigned => feedback
2011-06-30 16:24 davidl Status feedback => assigned
2011-06-30 16:24 davidl Assigned To jonesrt => davidl
2011-06-30 16:28 davidl Note Added: 0000137
2011-06-30 16:48 davidl Note Added: 0000138
2011-06-30 16:48 davidl Status assigned => resolved
2011-06-30 16:48 davidl Resolution open => fixed


Copyright © 2000 - 2020 MantisBT Team
Powered by Mantis Bugtracker