[Clas_offline] GSIM crash

[Nikolai Pivnyuk]

Hi Krishna,

YES !

===============================================================
[pivnyuk at ifarm1402 gsim_try]$ which gsim_bat
/group/clas/builds/centos7/trunk/build/bin/gsim_bat
===============================================================

/group/clas/builds/centos7/trunk/build/bin/gsim_bat - CRASHES.
gsim_bat - WORKS.

I am an ordinary man. Could somebody explain this miracle ?

Best regards,
Nikolai

> Hi Nikolai,
>
> I also had similar issues with gsim and I haven't resolved all of them
> yet. But, mysteriously, I was able to run it when I didn't run gsim_bat
> without giving it the full path name.
>
> For example, you seem to have run gsim (giving full path) as follows:
>
> /group/clas/builds/centos7/trunk/build/bin/gsim_bat -ffread ffread.g12
> -bosout bos_output.g12 -kine 1 -mcin bos_input.g12 > log.g12 2>&1
>
> Could you try without the full path as follows:
>
> gsim_bat -ffread ffread.g12 -bosout bos_output.g12 -kine 1 -mcin
> bos_input.g12 > log.g12 2>&1
>
> Before that, make sure the location of that executable
> (/group/clas/builds/centos7/trunk/build/bin/) has been added to your PATH
> variable. And, check if 'which gsim_bat' gives you the correct path.
>
> I haven't understood it yet, but it crashed (my log file at
> https://userweb.jlab.org/~adhikari/Tmp2/gsim_run.log) when I gave the full
> path name and ran fine when I didn't give the full path name.
>
> May be it will work for you too.
> Thank you.
>
> Regards,
> Krishna
>
>
> From: "Nikolai Pivnyuk" <pivnyuk at jlab.org>
> To: "Michael C. Kunkel" <mkunkel at jlab.org>
> Cc: pivnyuk at jlab.org, wphelps at jlab.org, "Krishna Adhikari"
> <adhikari at jlab.org>, maurik at physics.unh.edu, gilfoyle at jlab.org
> Sent: Tuesday, May 23, 2017 12:41:45 PM
> Subject: Re: GSIM crash
>
> Greetings,
>
> Actually all my problems with GSIM have been started a few months ago. At
> that time and before, "from the very beginning", I ran in c shell. So the
> problem with GSIM, I have faced first time, appeared under the c shell. I
> have switched to bash shell only a few months ago just to make sure that
> it does not depend on shell.
>
> Nikolai
>
>
>> Greetings,
>>
>> I run in C shell.
>> Would you consider running in C shell?
>>
>> BR
>> MK
>> ----------------------------------------
>> Michael C. Kunkel, USMC, PhD
>> Forschungszentrum Jülich
>> Nuclear Physics Institute and Juelich Center for Hadron Physics
>> Experimental Hadron Structure (IKP-1)
>> www.fz-juelich.de/ikp
>>
>> On 23/05/2017 17:22, Nikolai Pivnyuk wrote:
>>> Hello Michael and William,
>>>
>>> Michael, thank you for your respond . I have found the CLAS note with
>>> g12
>>> procedures. It is very useful for me to read it. I have already
>>> proceed.
>>>
>>> Michael, William, in my directory (open for read)
>>> /lustre/expphy/volatile/clas/clasg10/pivnyuk/g10_may_2014/gsim_try
>>> I put a couple of very simple and transparent scripts : gsim.test.g10a
>>> and
>>> gsim.test.g12. One of them runs gsim_bat with ffread and input bos
>>> files
>>> from my g10a stuff, and another (thanks to William) runs gsim_bat with
>>> corresponding g12 files. In both cases the program crashes in similar
>>> way.
>>> See log.g10a and log.g12. In the same time William runs g12 stuff
>>> perfectly, which is not naturraly a surprise. I am forced to accept the
>>> point that something is wrong with my account setups. I have no any
>>> other
>>> idea. My .bashrc file is also available - see my.bashrc. That is why I
>>> have asked in my previous mail for some "standard" CLAS6 .bashrc file.
>>>
>>> It seems to me that it would take only around 10 minutes maximum to
>>> copy
>>> my stuff and run script gsim.test.g10a in your accounts, inside your
>>> environment. It might be the crucial crosscheck. Could you be so kind
>>> to
>>> do that and help me ?
>>>
>>> Thanks,
>>> Nikolai
>>>
>>
>