[Eg6_analysis] CLASTOOL Issue

[Hovanes Egiyan]

Hi Lamiaa,

The solution for ClasTool is also simple.  Because it compiles and runs 
for both
64bit and 32bit machines all we need to do is to compile and run each 
executable
on the same platform (which would usually be considered the normal way).
This means that after you compile the main CLAS cooking packages 
(user_ana etc) on a
32 bit machine, you then go to the ClasTool directory and compile the 
ClasTool package. 
Now you have 32bit executables for both user_ana and ClasTool.
But then you do not stop there, you need to login to a 64 bit machine, 
and build
the libraries on 64bit machine (with "make lib"). Then without trying to 
compile
the executables ("user_ana" etc) you  only compile ClasTool for  64 bit.

Now when you submit the job you will need to make sure that "user_ana" 
and other
executables run from 32bit FC8 bin directory, while  ClasTool runs from 
either 32bit FC8
or 64 bit RHEL5 depending what farm queue you submitted it to. If we are 
going to use
only 64 bit farm we do not need to build 32bit ClasTool.

I already compiled and tested both executables :
/u/home/claseg6/pass0_nect/ClasTool/bin/LinuxRHFC8/WriteRootDst 
/u/home/claseg6/pass0_nect/ClasTool/bin/Linux64RHEL5/WriteRootDst
but we need to test your scripts. I am not sure what $OS_NAME these 64 bit
farm machines will yield in the jobs. Also the LD_LIBRARY_PATH will
need to be set properly when running WriteRootDst. When I  was testing 
it I had
to add :
/home/claseg6/pass0_nect/ClasTool/slib/LinuxRHFC8   to run on 32 bit
/home/claseg6/pass0_nect/ClasTool/slib/Linux64RHEL5  to run on 64 bit
to the front of "LD_LIBRARY_PATH" shell variable.
I do not quite understand the  way  ClasTool is setup  in your scripts, 
so I did not
want to mess with it. But it seems like you are trying to setup ClasTool 
independent of
the rest of the cooking tree, which is probably not a good idea.  We 
need to  set  it  up
parallel to other packages.

Hovanes.


Lamiaa El Fassi wrote:
>
> Or use "H2ROOT" ROOT file instead of CLASTOOL one :)
>
> Lamiaa
>
> ************************************************************
> *  Lamiaa El Fassi             email: elfassi at jlab.org 
> <mailto:elfassi at jlab.org>
> *  Research Associate @ Rutgers University          
> *  Phone: (757) 269-7011 // Fax: (757) 269-5703                
> *  Jefferson Lab., 12000 Jefferson Ave.               
> *  Suite# 4, MS 12H3
> *  Newport News, VA. 23606
> ************************************************************
>
>
> On Thu, Sep 9, 2010 at 2:02 PM, Stepan Stepanyan <stepanya at jlab.org 
> <mailto:stepanya at jlab.org>> wrote:
>
>     OK then, you all should learn now fortran and paw.
>     No problem with that on 64 bit machines. :-)
>
>     Stepan
>
>
>     On 9/9/10 1:35 PM, Lamiaa El Fassi wrote:
>>     Hi All,
>>
>>     In the cooking test that I did recently, it seems that the
>>     CLASTOOL Root file was not produced because a missed shared
>>     library "libCore.so.5.22" in the 64 bit farm node. This is the
>>     error message that I got in .err ASCII file:
>>     ------------------------------------------------------------------------------------------------------
>>      /home/claseg6/bin/LinuxRHFC8/WriteRootDst: error while loading
>>     shared libraries: libCore.so.5.22: cannot open shared object
>>     file: No such file or directory
>>     ------------------------------------------------------------------------------------------------------
>>
>>     Does anyone know how to fix that?
>>
>>     Thanks,
>>
>>     Lamiaa
>>
>>     ************************************************************
>>     *  Lamiaa El Fassi             email: elfassi at jlab.org
>>     <mailto:elfassi at jlab.org>
>>     *  Research Associate @ Rutgers University          
>>     *  Phone: (757) 269-7011 // Fax: (757) 269-5703                
>>     *  Jefferson Lab., 12000 Jefferson Ave.               
>>     *  Suite# 4, MS 12H3
>>     *  Newport News, VA. 23606
>>     ************************************************************
>>
>>
>>     _______________________________________________
>>     Eg6_analysis mailing list
>>     Eg6_analysis at jlab.org <mailto:Eg6_analysis at jlab.org>
>>     https://mailman.jlab.org/mailman/listinfo/eg6_analysis
>>           
>
>     _______________________________________________
>     Eg6_analysis mailing list
>     Eg6_analysis at jlab.org <mailto:Eg6_analysis at jlab.org>
>     https://mailman.jlab.org/mailman/listinfo/eg6_analysis
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Eg6_analysis mailing list
> Eg6_analysis at jlab.org
> https://mailman.jlab.org/mailman/listinfo/eg6_analysis
>