LHCb Core Software Meeting

Date and Location

10 January 2007
10:00 - 11::30
CERN (2-R-030)

Attendees

Florence, Gloria, Hubert, Joel, Juan, Marco Ca., Marco Cl. (minutes), Markus, Olivier, Pere, Philippe, Stefan, Thomas

Apologies

Andres, Nick

Subjects

News

No news.

Software Releases

Gaudi (Hubert)

  • Fixed all the unchecked StatusCode instances (except inside the Reflex dictionary)
  • New DataStreamTool and changes to EventSelector have been included
  • Still 40 warnings at compile time
  • Waiting for the bug-fix release of ROOT
  • Patches applied:
    • IMonitoringSvc
    • HistorySvc

Marco Ca.:

  • Dependencies are not computed correctly with the install area. Florence is aware of it and will try to use a trick suggested by David Quarrie.
  • It should be possible, with a command line switch, to have on Linux the same link time behavior of Windows. With the install area, we may have more cases in which we discover only when building on Windows that the link options are not correct.

LHCb, Boole, Brunel (Marco Ca.)

  • Built the full stack using LCG nightlies of Monday. The release should be ready ASAP.
  • Not yet tested on Windows.

Gauss (Gloria)

  • Porting to new Gaudi (with the help of Florence)
    • A lot of straight forward fix for the factories.
    • Giga is more complex because it uses special factories. Migrating to the new plug-in manager.
  • Problems in production with one event type.
  • Need to maintain a DC06 version for new event types etc. (new Gauss means new generators and different physics)

DaVinci (Juan)

  • New DaVinci is ready. (release during the day)
  • Florence is fixing the code to work with the new Gaudi.

Philippe: which is the time estimate to get DaVinci ready with all the pre-selections? End of January.

Panoramix (Thomas for Guy)

  • Development version ready , using OpenScientist 16
    • works with QT
  • 2 issues with the Muon:
    • RAW Muon contains digits linked to MC, but on DST there are only Coordinates (missing link from Coord. to MC). There is a way to get from a coordinate to MC hit, but it should be used to generate Linker tables.
    • No way to get Muon hits from Muon particles.

Round Table

Marco Cl.
  • Since the CondDB extension of the meeting never take place, the core software meeting will start again at 10:30, and the CondDB meeting is canceled.
  • The DataStreamTool is not ready for the release: few changes are needed in the interface.
  • Finally I managed to access the Conditions Database from the GRID. The jobs were not limited to tier-1s and the database structure was not particularly efficient, so it took around 1 and 2 hours tor read the whole detector description (~15 min. CPU time). New tests will be done with a more reasonable structure and limiting the jobs to tier-1s.
  • Starting to work to re-organize the structure of XmlDDDB to fit better in the CondDB. The main problem is to modify the scripts that generate XML files.
Joel
Running productions requested by Olivier.
Thomas
Using a patched version of DIRAC that works on Windows.
Florence
Do we still need all the environment variables "Package"Shr? No. We can remove them from the patterns and the ApplicationMgr.

-- MarcoClemencic - 15 Jan 2007

Edit | Attach | Watch | Print version | History: r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r2 - 2007-01-15 - MarcoClemencic
 
    • 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