Problem 1364 - Not seeing the detector in the visualization tool
Summary: Not seeing the detector in the visualization tool
Status: RESOLVED FIXED
Alias: None
Product: Examples/Basic and Novice
Classification: Unclassified
Component: N02 (show other problems)
Version: 9.5
Hardware: PC Linux
: P5 trivial
Assignee: John.Allison
URL:
Depends on:
Blocks:
 
Reported: 2012-10-17 11:31 CEST by Gersende Prior
Modified: 2012-10-30 22:40 CET (History)
1 user (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this problem.
Description Gersende Prior 2012-10-17 11:31:21 CEST
Hello,

Sorry if the problem may have been duplicated but I don't see the same problem in the already reported one.

I am running N02 and N03 on a CERN SLC5 machine (Linux pcen33383 2.6.18-308.16.1.el5 #1 SMP Thu Oct 4 14:02:28 CEST 2012 x86_64 x86_64 x86_64 GNU/Linux) using Geant4 version 9.5.p01 and the following ENV:

source /afs/cern.ch/sw/lcg/contrib/gcc/4.3.2/x86_64-slc5-gcc43-opt/setup.csh

setenv G4BASE /afs/cern.ch/sw/lcg/external/geant4

source $G4BASE/9.5.p01/x86_64-slc5-gcc43/setup.csh

And with the default vis.mac I don't see the volumes (target and detectors) in the vizualisation manager. 

Same case for N03, here the vizualisation manager with the default vis.mac does not show anything at all.

Maybe I am doing something wrong but I recall having problems with the novice examples and earlier versions of G4 where on SLC4 32x I would see the geometry but not on SCL5 64x.

Would you be able to let me know if I miss something and the geometry is not shown in these examples or if you have knowledge of problems running the visualization manager with these example on SLC5 64x ?

Many thanks

Gersende
Comment 1 John Apostolakis 2012-10-17 20:18:23 CEST
I am reassigning it to someone who is more knowledgeable about visualisation.
Comment 2 John.Allison 2012-10-17 21:30:20 CEST
Dear Gersende

There is a problem with CERN SLC4 and, possibly, SLC5.  See copy from an email below.

The issue seems to be with a version of X11 from x.org.  I recommend you make sure that you have the latest version of X11 on both the remote and local machines.

Please try it, and if this is not the solution, please contact us again.

John.

==================
Back in early October 2011, x.org issued a faulty update
to the X11 packages that causes OPENGL applications
to fail, at least on three SL 4.x systems here at
TRIUMF.  The fault is apparently with the X server
itself (file "Xorg").  The fix I used was to back up to the
previous rpms:
xorg-x11-6.8.2-1.EL.69.i386.rpm
xorg-x11-Xdmx-... etc.
    -x11-Xnest-
    -x11-Xvfb-
I am not sure if this affects any SL 5.X or 6.x systems
but would be curious to find out.
If you have a desktop Linux machine, you can check it
by running "glxgears".  With the faulty X server it will
fail with an X protocol error.  If it works you will see
some spinning gears in a small window.
Also you can check the date on the file Xorg and see if
it was updated on around October 6 or so.
(In reply to comment #1)
> I am reassigning it to someone who is more knowledgeable about visualisation.
Comment 3 John.Allison 2012-10-17 21:43:47 CEST
Dear Gersende

There is a problem with CERN SLC4 and, possibly, SLC5.  See copy from an email below.

The issue seems to be with a version of X11 from x.org.  I recommend you make sure that you have the latest version of X11 on both the remote and local machines.

Please try it, and if this is not the solution, please contact us again.

John.

==================
Back in early October 2011, x.org issued a faulty update
to the X11 packages that causes OPENGL applications
to fail, at least on three SL 4.x systems here at
TRIUMF.  The fault is apparently with the X server
itself (file "Xorg").  The fix I used was to back up to the
previous rpms:
xorg-x11-6.8.2-1.EL.69.i386.rpm
xorg-x11-Xdmx-... etc.
    -x11-Xnest-
    -x11-Xvfb-
I am not sure if this affects any SL 5.X or 6.x systems
but would be curious to find out.
If you have a desktop Linux machine, you can check it
by running "glxgears".  With the faulty X server it will
fail with an X protocol error.  If it works you will see
some spinning gears in a small window.
Also you can check the date on the file Xorg and see if
it was updated on around October 6 or so.
(In reply to comment #1)
> I am reassigning it to someone who is more knowledgeable about visualisation.
Comment 4 Gersende Prior 2012-10-18 09:53:47 CEST
Dear John,

Thanks a lot for your reply and suggestions. I have tried to run glxgears as recommended by the email you forwarded me and it works perfectly. The date of my Xorg file is Oct. 5 2012 and the version of X11 seems to be the most recent one. Do you know about a LXPLUS machine SLC5 64x where the N02/N03 examples visualisation works as expected that I could try out ? I would like to make sure that it is a machine related problem and not a Geant4 version/examples running problem.

Thanks,

Gersende
Comment 5 John.Allison 2012-10-29 14:42:29 CET
The email below from Andrea Dotti resulted in a resolution of this bug.  I hereby close the report.
John

===============================
Hi Gersende,

Please try to follow the solution I have described in this thread in the user forum:

 http://hypernews.slac.stanford.edu/HyperNews/geant4/get/visualization/615.html

It worked for a few people.

If you are on LXPLUS (or have access to CernVM-FS) a precompiled version of MesaGL for SLC5 with gcc4.3 can be found under:

 /cvmfs/geant4.cern.ch/opt/OpenGL/

See included README, to test it, just prepend to your LD_LIBRARY_PATH :
 /cvmfs/geant4.cern.ch/opt/OpenGL/x86_64-slc5-gcc43-opt/lib

CVMFS is a "auto-mount" file system.
Just type:

 ls /cvmfs/geant4.cern.ch and the repository will "magically" appear!

In case you can just copy the library (no need to recompile G4)
Andrea Dotti