Endless Loop in TFluka

Dear ROOT/VMC experts,
I have a working VMC application.

When I use the fluka executable sometimes an endless loop appears.

I obtain the list of operation executed in the endless loop
by means of the use of the gdb debugger:

In particular it seems that in the TFlukaMCGeometry::lkwr() function
a part of the code that is described by this comment
"//this should not happen!! ??? Ernesto"
begins to be executed.

What is the possible source of this problem?
Is there something wrong with my geometry?



What version of FLUKA and fluka_vmc are you using. Make sure you run the latest advertised on the VMC page. This can indeed point to a geometry problem. Is it something that you can ship to me so that I can try to reproduce the problem ?


I’m using the very last fluka_vmc (0.4) and fluka 2008 respin 3 (1 year ago).

In attachment you can find the TGeometry I use.

Many thanks,

Hi Alberto,

Your AMS geometry has about 180 illegal volume overlaps and some of them are ~1cm. No wonder TFluka transportation finds a lot of inconsistencies. You should check and fix these overlaps before retrying the exercise. To do that, just:
gGeoManager->CheckOverlaps(0.001); // down to 10 microns
gGeoManager->PrintOverlaps(); // prints the full list
new TBrowser(); // go to “Illegal overlaps” and double click items inside
Here is just a picture of one overlap to get the idea:

Hope this helps.

Thanks for your reply,

I have solved most of all the superpositions in the geometry
(see the attachment), but the problem is always there.

I have noticed two features:

The problem occurs always when an hadronic interaction
occurs for a primary (process 13).

The problem occurs in normal volumes (38, 202, 234,
235, 279, 314, 346, 349, 378, 603, 672, 870, 873,
1017, 1116, 1177, 1178, 1194…), no one of them is
related with the overlapping region.

I hope this can help
many thanks,

Is it too difficult to send me your vmc application so that I can reproduce ?