Hi Robert,

I have attached the log files I made while debugging things and one of the xml files I was testing.  I did some more tests today, though, and had the same issue with gxspl-t2k-v2.6.2.xml and gxspl-freenuc-v2.6.2.xml. To answer your questions I was running the command

gspl2root -f xsec_splines_14_1000010010.xml -p 14 -t 1000010010 -w

with 

GENIE: 2.6.4 and 2.8.0
GCC: 4.4.7 and 4.4.8
ROOT: 5.34.18, 5.34.19, and 5.34.20

I have GENIE both on my personal machine (GCC 4.4.8) and our cluster (GCC 4.4.7).  On my machine 2.6.4 and 2.8.0 are running with ROOT 5.34.19 while on our cluster I built GENIE 2.6.4 with ROOT 5.34.20 and GENIE 2.8.0 with ROOT 5.34.18.  

Now that Costas has given me the username/pw I can still make a bug report if you like, but have never done so and am not sure how much is too much (or too little) information.  

Thanks for the help!
Joshua 

________________________________________
From: Robert W Hatcher <[log in to unmask]>
Sent: Friday, October 17, 2014 10:23 AM
To: Joshua Hignight
Cc: GENIE Support
Subject: Re: Problems with gspl2root

Hi Joshua,
  Thank you for reporting this issue; I'm not sure of the procedure for non-hepforge registered developers to make trac entries, but I'm happy to collect all the information you could send me on the problem.

The exact command you issue would be helpful.  Also which version of the compiler and ROOT that you're using.  I'm hoping that it isn't too dependent on SLF version nor these auxiliary issues, because matching them might be problematic.  Another potential issue the exact collection of isotopes in your spline file -- I might ask for that if I can't reproduce the problem.

-robert

On Oct 16, 2014, at 5:32 PM, Joshua Hignight <[log in to unmask]> wrote:

> Dear GENIE users,
>
> I have been playing around with making plots using gspl2root.  While doing so I have encountered a bug when the -w (print to PS) option is enabled.  Using gdb and valgrind I can trace the bug back to when AlgFactory tries to clean up with a seg fault occurring when the COHKinematicsGenerator destructor is called.  I have been able to replicate the problem with both GENIE version 2.6.4 and 2.8.0 running on both SL 6.5 and RHEL 6.3.  I would be happy to report the bug at http://projects.hepforge.org/genie/trac/report/ if I could get the pw for general_user.  I can also provide the backtrace and valgrind log if anyone would find it useful.  Unfortunately, even through I can find what is causing the bug I do not know how to fix it.
>
> Thanks,
> Joshua Hignight
>