TGeoManager hanging in TBrowser

Hello,

I'm running into a problem opening TGeoManager objects using TBrowser. For instance, I save a closed TGeoManager object to an output file using TGeoManager::Export()

[build the geometry]
gGeoManager->CloseGeometry("di");
gGeoManager->Export("ND280Geometry.root");

and then open the exported file from the command line and create a browser

$ root ND280Geometry.root
root [1] new TBrowser

If I click through the "All Folders" panel on the left side of the browser, I can browse the geometry without any problems (i.e. double click "ROOT Files", double click "ND280Geometry.root", single click "ND280Geometry") . I get the following messages:

> Info in <TGeoManager::CloseGeometry>: Geometry loaded from file...
> Warning in <TGeoHMatrix::RegisterYourself>: cannot register without geometry
> Info in <TGeoManager::SetTopVolume>: Top volume is t2k. Master volume is t2k
> Info in <TGeoManager::CloseGeometry>: Voxelization retrieved from file
> Info in <TGeoManager::BuildCache>: --- Maximum geometry depth is 11
> --- node ID tracking disabled
> Info in <TGeoManager::CloseGeometry>: 47785 nodes/ 318 volume UID's in ND280 Detector Geometry
> Info in <TGeoManager::CloseGeometry>: ----------------modeler ready----------------

The warning is a bit disturbing, but seems to be harmless. I can then browse the geometry.

However, if I use the "Contents of" window on the right side of the browser, ROOT hangs while using 100% of the CPU, and apparently trying to allocate huge chunks of memory (i.e. double click "ROOT FIles", double click "ND280Geometry.root", double click "ND280Geometry" in the "Contents of /ROOT Files/ND280Geometry.root" panel). There are no error messages or warnings, and the only way I've found to recover is to leave X windows and kill the root.exe process by hand.

If anybody has insight on how to track down the problem, or if this is a root bug, help will be very much appreciated!

I'm using root 5.10/00c compiled from source on Debian Linux and have placed a ROOT file with a TGeoManager object demonstrating the problem at

<http://nngroup.physics.sunysb.edu/~mcgrew/ND280Geometry.root>

Thanks you very much in advance for any help.

Best Regards,

-- 
Clark McGrew                    Univ. at Stony Brook, Physics and Astronomy
<clark.mcgrew_at_stonybrook.edu>   631-632-8299