Problem 315

Summary: Malfunction of terminal output
Product: Geant4 Reporter: hase
Component: intercomsAssignee: Makoto.Asai
Status: RESOLVED FIXED    
Severity: major    
Priority: P2    
Version: 3.2   
Hardware: PC   
OS: Linux   

Description hase 2001-10-22 09:29:53 CEST
When I run the exampleN01 I get the following output:
Step#    X(mm)    Y(mm)    Z(mm) KinE(MeV)  dE(MeV) StepLeng TrackLeng
NextVolume ProcName
    0     1 -1.6e+03     2     3     4     5     6     7     8     9    10
11    12    13    14    15    16    17    18    19    20    21    22    23
1.01e+03    24 1.09e+03    25 1.11e+03    26 1.19e+03    27 1.21e+03    28
1.29e+03
29 1.31e+03    30 1.39e+03    31 1.41e+03    32 1.49e+03    33 1.51e+03    34
1.59e+03    35 1.61e+03    36 1.69e+03    37 1.71e+03    38 1.79e+03    39
1.81e+03    40 1.89e+03    41 1.91e+03    42    43 Track (trackID 1, parentID 0)
With no information about KinE, dE, StepLength, etc, like in the examples
outputs. The same happens when I try to dump the materials table and the Cuts in
energy for the materials. I worry that the programa may not be constructing the
data tables that it needs at initialization.
This happen with all the examples. I've tried many possibilities and none seems
to work.
Comment 1 Makoto.Asai 2001-10-22 18:26:59 CEST
Could you let me know the versions of your compiler and CLHEP?
Comment 2 Makoto.Asai 2001-10-24 15:28:59 CEST
I guess the bug is fixed.

I contacted V. Hejny who reported a similar problem previously and he said
the problem lies in one of the libraries of the g++. So I downloaded
geant4.3.2 again at home and compile it there. At home I have Suse 6.4
instead of 7.1
At home I had no problems with the output !!!
(I had other problems though because this old compiler apparently does not
support -lXm the flag).
 So I came back to my working machine and there I ONLY changed the libraries:
libstdc++-libc6.2-2.so.3 for the OLD ONE  libstdc++-libc6.1-2.so.3 and did
the same for the archive files.
  Now it works fine ! Also with OpenGLXm and Xm UI. I will continue testing
geant for a while hoping that this change does not interfere with the rest of
the code.
  I hope this will help to get more robust scrypts that detect the problem
in advance! Thanks a lot...