|
|
[Rivet] Fwd: Running RAPGAP with RIVET
Tim Martin
tim.martin at cern.ch
Thu Apr 15 14:06:31 BST 2010
Thanks Andy,
That did it. I'm down to a runtime error! I can run Pythia with AGILe
('agile-runmc Pythia6:422' works) but am still getting a python error
running RAPGAP though the error seems to be coming from Pythia?! I've
asked Holger about it as well.
Cheers, Tim.
tamartin at lxplus253.cern.ch$ agile-runmc Rapgap:32 -n 100000 -o
hepmc.fifo < rapgap/share/steer-pp-rapidity-gap
Generating 100000 events
Generator is Rapgap:32
Setting random seed = 31415926
Setting initial state: 2212 @ 7000.0 GeV -> <- 2212 @ 7000.0 GeV
AGILe.Rapgap: INFO Setting initial state...
AGILe.Rapgap: INFO Setting ... (KBE1) = 2212
AGILe.Rapgap: INFO Setting ... (KBE2) = 2212
AGILe.Rapgap: INFO Setting ... (PBE1) = 7000
AGILe.Rapgap: INFO Setting ... (PBE2) = 7000
AGILe.Rapgap: INFO Now calling GRAINI...
NEW BLOCK DATA for RAPGAP initialised
HERACLES 4.6.6 BLOCK DATA initialized
python: symbol lookup error:
/afs/cern.ch/sw/lcg/external/MCGenerators/pythia6/421.2/slc4_amd64_gcc34/lib/libpythia6.so:
undefined symbol: pytime_
On 14 April 2010 18:25, Andy Buckley <andy.buckley at ed.ac.uk> wrote:
> Hi Tim,
>
> Try "autoreconf -i". This is the normal way to bootstrap an autotools
> setup. Should also work on lxplus (although the automake installed there
> is not as nice as on our release build machine, so data file
> installation might take a while!)
>
> Andy
>
>
> On 14/04/10 17:22, Tim Martin wrote:
>> Thanks Andy,
>>
>> I think I'm almost there now, main problem before was I was not using
>> the head but rather AGILe-1.1.4.
>>
>> When I installed from the tarball I used the lines...
>>
>> ./configure --prefix=/afs/cern.ch/user/t/tamartin/scratch0/rivet/local
>> --with-lcgtag=slc4_amd64_gcc34
>> --with-hepmc=/afs/cern.ch/sw/lcg/external/HepMC/2.05.00/slc4_amd64_gcc34
>> --with-boost=/afs/cern.ch/user/t/tamartin/scratch0/rivet/local
>> --enable-pyext
>>
>> make -j2 && make -j2 install
>>
>> ...to install the package but I see that there is no configure file
>> saved in the repository, I've tried
>>
>> autoconf configure.ac > configure
>>
>> on LXPLUS but I get the errors....
>>
>> configure.ac:8: error: possibly undefined macro: AM_INIT_AUTOMAKE
>> If this token and others are legitimate, please use m4_pattern_allow.
>> See the Autoconf documentation.
>> configure.ac:32: error: possibly undefined macro: AC_CEDAR_OSX
>> configure.ac:35: error: possibly undefined macro: AC_LCG_TAG
>> configure.ac:44: error: possibly undefined macro: AC_DISABLE_STATIC
>> configure.ac:49: error: possibly undefined macro: AM_FFLAGS
>> configure.ac:49: error: possibly undefined macro: AM_FCFLAGS
>> configure.ac:53: error: possibly undefined macro: AC_LIBTOOL_DLOPEN
>> configure.ac:54: error: possibly undefined macro: AC_PROG_LIBTOOL
>> configure.ac:61: error: possibly undefined macro: AC_CEDAR_HEADERS
>> configure.ac:61: error: possibly undefined macro: AC_MSG_ERROR
>> configure.ac:77: error: possibly undefined macro: AC_CEDAR_LIBRARYANDHEADERS
>> configure.ac:81: error: possibly undefined macro: AM_CPPFLAGS
>> configure.ac:84: error: possibly undefined macro: AM_LDFLAGS
>> configure.ac:85: error: possibly undefined macro: AM_CXXFLAGS
>> configure.ac:86: error: possibly undefined macro: AC_CEDAR_CHECKCXXFLAG
>> configure.ac:116: error: possibly undefined macro: AM_CONDITIONAL
>> configure.ac:123: error: possibly undefined macro: AC_EMPTY_SUBST
>>
>> Then when I try and run configure...
>>
>> ./configure: line 1348: syntax error near unexpected token `dist-bzip2'
>> ./configure: line 1348: `AM_INIT_AUTOMAKE(dist-bzip2)'
>>
>> ..among others. Does anyone happen to know how to properly generate
>> the configure or have a copy suitable for head version of AGILe?
>>
>> Thanks, Tim.
>>
>> On 13 April 2010 19:02, Andy Buckley <andy.buckley at ed.ac.uk> wrote:
>>> Hi Tim,
>>>
>>> RapGap and Cascade in AGILe are currently only known about by Holger, I
>>> think: the issue with static libraries etc. is really something that
>>> Hannes & co can hopefully fix. If the shared lib is in place, and can be
>>> found in a Genser-like structure in the path pointed at by the
>>> AGILE_GEN_PATH, then I *think* it should just work. The -lTRACE argument
>>> to agile-runmc should dump out a big list of all the paths being
>>> searched to find the generator libraries.
>>>
>>> As Holger has said, Rivet doesn't write ROOT tuples... well, you could
>>> add a private analysis that works that way, but it's not a built-in
>>> feature. But Rivet does read in HepMC text streams, so it may be that
>>> you don't need AGILe at all if you already have RapGap writing HepMC
>>> ASCII to stdout: just pipe the stdout into Rivet! AGILe is a convenience
>>> rather than a necessity, so if you have a more convenient way to steer
>>> it and produce HepMC, then use that instead!
>>>
>>> Feel free to ask more questions, but send them to the Rivet list,
>>> please... we'll hopefully have the technical stuff sorted fairly soon.
>>>
>>> Andy
>>>
>>>
>>> On 13/04/10 17:45, Tim Martin wrote:
>>>> Hi Andy,
>>>>
>>>> I'm trying to run the RAPGAP generator with AGILe and RIVET and I was
>>>> hoping you could help with some of the technicalities I've detailed
>>>> below.
>>>>
>>>> Many thanks,
>>>> Tim.
>>>>
>>>> ---------- Forwarded message ----------
>>>> From: Judith.Katzy at desy.de <judith.katzy at desy.de>
>>>> Date: 13 April 2010 17:24
>>>> Subject: Re: Running RAPGAP with RIVET
>>>> To: Tim Martin <tim.martin at cern.ch>
>>>> Cc: Judith.Katzy at cern.ch, Dr Paul R Newman <prn at hep.ph.bham.ac.uk>
>>>>
>>>>
>>>> Hi Tim,
>>>>
>>>> I'm sorry but I cannot help you with Agile. I was about to point you
>>>> to Holger - but it seems
>>>> this was already done. I can only suggest to ask Andy Buckley - he
>>>> should know these technical things.
>>>>
>>>> Please let me know if you don't succeed this way - I could than point
>>>> you to some other tools.
>>>>
>>>> cheers,
>>>> Judith
>>>>
>>>> On 13/04/2010 17:08, Tim Martin wrote:
>>>>>
>>>>> Hi Judith,
>>>>>
>>>>> I'm having MC problems and I was hoping you could help me or point me
>>>>> towards someone who can.
>>>>>
>>>>> Essentially, I'm looking to run RAPGAP to get the included pion
>>>>> exchange channel in diffraction and look at the forward neutron
>>>>> spectrum for ZDC and diffractive purposes. I have RAPGAP from Hannes
>>>>> Jung compiled on my AFS though I can only get it to output to STDOUT.
>>>>>
>>>>> I'm hoping to use AGILe to run RAPGAP and pass the output through
>>>>> RIVET to get (hopefully) a root ntuple.
>>>>>
>>>>> My problem is interfacing RAPGAP with AGILe, I have AGILe and RIVET
>>>>> installed and have made shared libraries for RAPGAP (as suggested in
>>>>> Fw:, below) but have now hit a brick wall regarding what needs to be
>>>>> done next to run RAPGAP in AGILe.
>>>>>
>>>>> Any help would be greatly appreciated,
>>>>> Thanks, Tim.
>>>>>
>>>>>
>>>>> ---------- Forwarded message ----------
>>>>> From: Holger Schulz<holger.schulz at physik.hu-berlin.de>
>>>>> Date: 9 April 2010 14:39
>>>>> Subject: Re: Running RAPGAP with RIVET
>>>>> To: Tim Martin<tim.martin at cern.ch>
>>>>>
>>>>>
>>>>> Tim Martin wrote:
>>>>> Hi Tim,
>>>>>
>>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> I hear from Hannes Jung that you have previously managed to run the
>>>>>> RAPGAP generator with RIVET to make a root ntuple.
>>>>>>
>>>>>>
>>>>>
>>>>> this is not entirely correct. I managed to interface Rapgap with
>>>>> AGILe, a generator interface for FORTRAN generators that
>>>>> produces output in the common HepMC event record format.
>>>>>
>>>>> Rivet is entirely independent on any generator code now. It uses
>>>>> this HepMC output as only input. Furthermore the output of
>>>>> Rivet is currently limited to AIDA xml format although there
>>>>> are some tools that can do a conversion from AIDA to root but
>>>>> I have no idea how well they work.
>>>>>
>>>>> So what you want to to in order to run Rapgap with Rivet is
>>>>> to install the latest HEAD version of AGILe from the SVN repository.
>>>>> AGILe looks for generator libraries in some common places
>>>>> on your hard drive.
>>>>>
>>>>> There are some more stumbling blocks on the way, however.
>>>>> AGILe of course only works with shared libraries, but the
>>>>> Rapgap installation only provides you with the static library.
>>>>> So you will need to make a shared library libragpag32.so
>>>>> from the object files yourself.
>>>>>
>>>>> Another thing that might go away in the future, provided Hannes
>>>>> can something with the parameter steering, is that you cannot
>>>>> change parameters in the AGILe way, i.e. specifying a text files
>>>>> with parameters and the values you want to set them to.
>>>>> Rapgap expects the steering file to be supplied as such:
>>>>>
>>>>> rapgap< steer-ep
>>>>>
>>>>> or so. So you have to run agile accordingly, e.g.
>>>>>
>>>>> agile-runmc Rapgap:32 -n 100000 -o hepmc.fifo< steerp_ep& rivet
>>>>> hepmc.fifo -a H1_2000_S4129130 -H rapgap.aida
>>>>>
>>>>> This would give you a file called rapgap.aida with the histograms
>>>>> filled according to this particular H1 analysis with the Spires ID 4129130.
>>>>> With the Rivet tools compare-histos and make-plots you can turn them
>>>>> into output like this:
>>>>>
>>>>> http://users.hepforge.org/~holsch/Rapgap/plots/H1_2000_S4129130_d18-x01-y01.pdf
>>>>>
>>>>> Hope that helps.
>>>>> Holger
>>>>>
>>>>>
>>>>>>
>>>>>> Is this correct? And if so, do you happen to have any example code
>>>>>> lying around which I could have a look at?
>>>>>>
>>>>>> Many thanks,
>>>>>> Tim.
>>>>>>
>>>>>> ---------- Forwarded message ----------
>>>>>> From: Hannes Jung<hannes.jung at desy.de>
>>>>>> Date: 9 April 2010 13:28
>>>>>> Subject: Re: rapgap
>>>>>> To: Tim Martin<tim.martin at cern.ch>
>>>>>>
>>>>>>
>>>>>> Hi Tim
>>>>>>
>>>>>> very good.
>>>>>> yes, sorry, the particle code was just from tevatron...
>>>>>> Now, for analysing the output, it depends:
>>>>>> - if you wnat it in the ATLAS frame, then you need Judith to help you.
>>>>>> - H. Schultz from ATLAS did run RAPGAP also with RIVET, so you could ask him...
>>>>>> - i only have a fortran version... but this might be not that
>>>>>> usefulfor you, or ?
>>>>>>
>>>>>> Cheers
>>>>>>
>>>>>> Hannes
>>>>>>
>>>>>> On 09.04.2010, at 13:01, Tim Martin wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>
>>>>>>> Hi Hannes,
>>>>>>>
>>>>>>> No problem, copied lhapdf-config into ~tamartin/bin/ and it's working :)
>>>>>>>
>>>>>>> Getting output, but only to stdout at the moment, how do you go about
>>>>>>> analysing the output? Ideally it would be nice to have an ntuple, but
>>>>>>> may have to involve the MC group for that?
>>>>>>>
>>>>>>> One other quick question, in your 7TeV pp steering file you have beam
>>>>>>> one set as -2212 and beam two set as 2212. Is this just a Tevatron
>>>>>>> typo?
>>>>>>>
>>>>>>> Thanks for your continuing help! Tim.
>>>>>>>
>>>>>>> On 8 April 2010 19:30, Hannes Jung<hannes.jung at desy.de> wrote:
>>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>> Dear Tim
>>>>>>>>
>>>>>>>> sorry for my late reply.
>>>>>>>> I have now copied the lhapdf-config to the public/rapgap/share folder.
>>>>>>>> the lhapdf-config is also available from
>>>>>>>> /afs/cern.ch/sw/lcg/external/MCGenerators/lhapdf/5.3.1/slc4_amd64_gcc34/bin/
>>>>>>>>
>>>>>>>> LEt me know, if this works
>>>>>>>>
>>>>>>>> cheers
>>>>>>>> Hannes
>>>>>>>>
>>>>>>>> On 08.04.2010, at 17:28, Tim Martin wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>>
>>>>>>>>> Hi Hannes,
>>>>>>>>>
>>>>>>>>> I see you have an entry for
>>>>>>>>> /afs/cern.ch/user/j/jung/.lhapdf/pdfsets-path
>>>>>>>>> But I can't see where it's set sadly :)
>>>>>>>>>
>>>>>>>>> Is it just a soft link? rapgap is refusing to run without it.
>>>>>>>>>
>>>>>>>>> Cheers, im.
>>>>>>>>>
>>>>>>>>> On 6 April 2010 07:00, Hannes Jung<hannes.jung at desy.de> wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Hi Tim
>>>>>>>>>>
>>>>>>>>>> I have now put rapgap into my public area:
>>>>>>>>>>
>>>>>>>>>> /afs/cern.ch/user/j/jung/public/rapgap/share
>>>>>>>>>>
>>>>>>>>>> run it with (for example)
>>>>>>>>>> ../bin/rapgap32< steer-pp-rapidity-gap
>>>>>>>>>>
>>>>>>>>>> Good luck
>>>>>>>>>>
>>>>>>>>>> hannes
>>>>>>>>>>
>>>>>>>>>> On 06.04.2010, at 03:45, Tim Martin wrote:
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Hi Sebastian,
>>>>>>>>>>>
>>>>>>>>>>> Hope you had a good Easter.
>>>>>>>>>>>
>>>>>>>>>>> I'll have a go at running rapgap this week and see how it is
>>>>>>>>>>> comparing. Can't say what I'll get out of it by the 8th though.
>>>>>>>>>>> Hannes, could you mod the permission on the rapgap dir in your
>>>>>>>>>>> userspace or pop it in public as I can't get at it at the moment.
>>>>>>>>>>>
>>>>>>>>>>> Thanks, Tim.
>>>>>>>>>>>
>>>>>>>>>>> On 30 March 2010 22:24, Hannes Jung<hannes.jung at desy.de> wrote:
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> Dear Sebastian& Tim
>>>>>>>>>>>>
>>>>>>>>>>>> I have already installed RAPGAP on lxplus.
>>>>>>>>>>>>
>>>>>>>>>>>> You can find it under
>>>>>>>>>>>>
>>>>>>>>>>>> /afs/cern.ch/user/j/jung/scratch0/rapgap/
>>>>>>>>>>>>
>>>>>>>>>>>> and you can run it with:
>>>>>>>>>>>> cd /afs/cern.ch/user/j/jung/scratch0/rapgap/share/
>>>>>>>>>>>> ../bin/rapgap32< steer-pp
>>>>>>>>>>>>
>>>>>>>>>>>> in this dir you also find the manual.
>>>>>>>>>>>>
>>>>>>>>>>>> LEt me know, if you need some help
>>>>>>>>>>>>
>>>>>>>>>>>> Cheers
>>>>>>>>>>>> Hannes
>>>>>>>>>>>>
>>>>>>>>>>>> On 30.03.2010, at 22:41, Sebastian White wrote:
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>> Dear Tim,
>>>>>>>>>>>>> Hannes and I just discussed what we might do to introduce forward
>>>>>>>>>>>>> physics into Emily and Marj's meeting on April 8.
>>>>>>>>>>>>> The idea is that with fast forward neutrons of probability
>>>>>>>>>>>>> 0.45 in each beam direction and the very large xF coverage ATLAS
>>>>>>>>>>>>> has for neutrons, correlations with ZDC and barrel data are bound to
>>>>>>>>>>>>> be important for their purposes.
>>>>>>>>>>>>> We propose to start with RAPGAP because we will have more
>>>>>>>>>>>>> insight into what the code is doing with forward neutrons.
>>>>>>>>>>>>> As far as I can tell from Judith it is not in ATHENA. However
>>>>>>>>>>>>> there must be possible tests we could do now. We can easily provide
>>>>>>>>>>>>> a simple effective collimator aperture due to the beamline, since we now
>>>>>>>>>>>>> study this with HECTOR and FPtracker. Charged particles are basically
>>>>>>>>>>>>> all lost in this region.
>>>>>>>>>>>>> I think I can find a theory PHD interested in simulation to help
>>>>>>>>>>>>> eventually.
>>>>>>>>>>>>> Hannes proposes that he could install RAPGAP on LXPLUS at CERN
>>>>>>>>>>>>> for example to get things running. What do you think?
>>>>>>>>>>>>> Sebastian
>>>>>>>>>>>>>
>>>>>>>>>>>>> ps. please forward if you also have colleagues who would be
>>>>>>>>>>>>> interested.
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> ***********************************************************************
>>>>>>>>>>>> Hannes Jung
>>>>>>>>>>>> Email: Hannes.Jung at desy.de
>>>>>>>>>>>> mobile :+49 40 8998 93741
>>>>>>>>>>>> http://www.desy.de/~jung
>>>>>>>>>>>> Tel: +49 (0) 40 8998 3741
>>>>>>>>>>>> Fax: +49 (0) 40 8998 3093
>>>>>>>>>>>> DESY, CMS/FH1, Bldg. 1b, 331, Notkestr.85, 22603 Hamburg, FRG
>>>>>>>>>>>> ***********************************************************************
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> --
>>>>>>>>>>> -+-+-+-+ Tim Martin +-+ ATLAS +-+-+-+-
>>>>>>>>>>> Physics West, Room W313
>>>>>>>>>>> University of Birmingham, B152TT
>>>>>>>>>>> Tel: (+44)121 414 4631
>>>>>>>>>>> Web: http://cern.ch/Tim.Martin
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> ***********************************************************************
>>>>>>>>>> Hannes Jung
>>>>>>>>>> Email: Hannes.Jung at desy.de
>>>>>>>>>> mobile :+49 40 8998 93741
>>>>>>>>>> http://www.desy.de/~jung
>>>>>>>>>> Tel: +49 (0) 40 8998 3741
>>>>>>>>>> Fax: +49 (0) 40 8998 3093
>>>>>>>>>> DESY, CMS/FH1, Bldg. 1b, 331, Notkestr.85, 22603 Hamburg, FRG
>>>>>>>>>> ***********************************************************************
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> -+-+-+-+ Tim Martin +-+ ATLAS +-+-+-+-
>>>>>>>>> Physics West, Room W313
>>>>>>>>> University of Birmingham, B152TT
>>>>>>>>> Tel: (+44)121 414 4631
>>>>>>>>> Web: http://cern.ch/Tim.Martin
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> ***********************************************************************
>>>>>>>> Hannes Jung
>>>>>>>> Email: Hannes.Jung at desy.de
>>>>>>>> mobile :+49 40 8998 93741
>>>>>>>> http://www.desy.de/~jung
>>>>>>>> Tel: +49 (0) 40 8998 3741
>>>>>>>> Fax: +49 (0) 40 8998 3093
>>>>>>>> DESY, CMS/FH1, Bldg. 1b, 331, Notkestr.85, 22603 Hamburg, FRG
>>>>>>>> ***********************************************************************
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> -+-+-+-+ Tim Martin +-+ ATLAS +-+-+-+-
>>>>>>> Physics West, Room W313
>>>>>>> University of Birmingham, B152TT
>>>>>>> Tel: (+44)121 414 4631
>>>>>>> Web: http://cern.ch/Tim.Martin
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> ***********************************************************************
>>>>>> Hannes Jung
>>>>>> Email: Hannes.Jung at desy.de
>>>>>> mobile :+49 40 8998 93741
>>>>>> http://www.desy.de/~jung
>>>>>> Tel: +49 (0) 40 8998 3741
>>>>>> Fax: +49 (0) 40 8998 3093
>>>>>> DESY, CMS/FH1, Bldg. 1b, 331, Notkestr.85, 22603 Hamburg, FRG
>>>>>> ***********************************************************************
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>> --
>>> Dr Andy Buckley
>>> SUPA Advanced Research Fellow
>>> Particle Physics Experiment Group, University of Edinburgh
>>>
>>
>>
>>
>
>
> --
> Dr Andy Buckley
> SUPA Advanced Research Fellow
> Particle Physics Experiment Group, University of Edinburgh
>
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
>
>
--
-+-+-+-+ Tim Martin +-+ ATLAS +-+-+-+-
Physics West, Room W313
University of Birmingham, B152TT
Tel: (+44)121 414 4631
Web: http://cern.ch/Tim.Martin
More information about the Rivet
mailing list
|