LHCb Core Software Meeting

Date and Location

12. July 2006
10:00 - 11:30
CERN (2-R-030)

Attendees

Andres (phone), Florence, Gloria, Greig (phone), Hubert, Joel, Juan, Marco Ca., Marco Cl. (minutes), Manuel, Matt, Nick, Nicolas, Pere, Philippe, Stefan, Thomas, Tim Adye (phone)

Subjects

News

  • New releases in LCG (LCG_46)
    • ROOT 5.12.00 (note that win23 version is built in win32_vc71_dbg and not anymore in win32_vc71)
    • RELAX_1_1_2
    • CORAL_1_5_1
    • COOL_1_3_2c and POOL_2_4_5 are being built
  • Application Area Meeting in the afternoon about Geant4.

Software Releases

Gaudi (Hubert)

  • Worked on LCGCMT package for GFAL
    • Got with Philippe a working minimal run-time environment
    • No need to source the gridenv script
    • Will be in LCGCMT 46
  • Two bug found in the option compiler. Fixed by Vanya.
  • Waiting for POOL to start building.

LHCb, Boole, Brunel (Marco Ca.)

  • Full chain released on Monday for DC06 rDST production.
    The stripping version should be ready in a week time, based on Gaudi v18r6.
  • There is a problem (to be discussed with Markus) with the produced ETC files. I principle they should be linked to the raw data files, but they are actually linked to the rDST file from which they were generated (it is correct, but not what we want).
  • The changes in DetDesc that sped up the transport service are buggy, but in the CVS head. If somebody needs to work on DetDesc, he should know that the head version is not a good one.

Gauss (Gloria)

  • Released Gauss v25r0, now used in production.
    • Few coredumps for time-out observed (in Marseille)
  • Crashes in decays in Pythia (b barions?)
  • Trying to compile Gauss with gcc 3.4
  • Writing documentation

DaVinci (Juan)

  • Phys and DaVinci are being built (with everithing mentioned during the software week).
  • Preparing the algorithms for random selection, but triggering the generation of particles.
  • This version will be the last one using the TrackVertex, PhysVertex etc. The next one will use the new vertex classes.

Round Table

Marco Cl.
Few important changes in DetCond. Added support for different COOL database depending on the path. ICondDBAccessSvc splitted in 3.
Pere
The excercise of building all software projects on the head version of ROOT showed a few problems
  • COOL was built in the "internal" directory and not visible.
  • LHCb project was tested (because Pere did it himself)
  • CMS and Atlas could not build (wrong configurations)
  • It seems that the process cannot be automatized: the configuration files have to be modified by hand
Florence
Fighting with the new CMT, which seems to be developed in the way Atlas is using it, but buggy when we use it our way (select the debug version with --tag=$CMTDEB).

Specific Topics

HyperNews for LHCb (Tim Ayde)

Tim showed us what the HyperNews system provides in terms of usability.

A HyperNews LHCb instance has been set up in the LCG managed server.

To be able to use the system one has to register as user.

It has been agreed to test the new system using the lhcb-core-soft mainling list. All the e-mails posted to HyperNews will be forwarded to the old mailing list to be able to go back to it without loosing the posts. Posts from normal users to lhcb-core-soft will not be allowed. A mail will be sent to the list to inform the users.


-- MarcoClemencic - 12 Jul 2006

Edit | Attach | Watch | Print version | History: r4 < r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r4 - 2006-11-17 - unknown
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    LHCb All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback