LHCb Core Software Meeting

Date and Location

14. June 2006
10:30 - 12:00
CERN (2-R-030)

Attendees

Florence, Gloria, Hubert, Joel, Juan, Marco Ca., Marco Cl. (minutes), Markus, Manuel, Matt, Nick, Nicolas, Olivier, Pere, Philippe

Apologies

Thomas

Subjects

News

  • Application Area Meeting in the afternoon about MacOSX platform support and POOL/CORAL.
  • Next week is LHCb Software Week

Software Releases

Gaudi (Hubert)

  • Applied a patch to G.O.D. to fix a problem with gcc 3.2.3 introduced with the previous patch (Marco Cl.)
    Marco Cl. will provied an example to test G.O.D. in Gaudi.
  • Ported StagerSvc to Castor2 (not yet tested because of a bug in Castor2)

Marco Ca.: Why G.O.D. is in Gaudi? It is LHCb specific.
We should check with other possible users and may be move it to LHCb.

Two email threads were circulating:

  • Copy constructor needed for classes inheriting from KeyedObject.
    G.O.D. will be modified to implement when needed (as soon as Stefan is back).
  • Pretty print out of enums does not work with std::ostream.
    Two things should be done:
    • change the current implementation to use std::ostream instead of MsgStream.
      It can be done now.
    • add a function to convert the enum value to a string.
      Better to wait for Stefan.
All the changes will be in a future version.

LHCb, Boole, Brunel (Marco Ca.)

  • LHCb v21r1 almost ready for release (out by Friday or Monday) not for production
    • fixes on tracking
    • compiles on SLC4
We would like to have the new version of ROOT to solve the problems we have with dictionaries.

Gauss (Gloria)

  • Gauss v24r8 released with many fixes (geometry, units, etc.)
  • Now going through all pending bugs (null momentum, particles on the moon)
  • With the new generators, the script to generate option files needs an update.
    Manuel is working on it.
Marco Ca.: new versions of XmlDDDB (v30r8) and XmlConditions (v1r7) for the production, with fixes and also improvements for alignments.
It is becoming more and more important to move to the CondDB.

Marco Cl.: prepared a first prototype of a "real" CondDB that can be used as a replacement for XmlConditions (based on v1r7).

There are still problems of touching volumes.

DaVinci (Juan)

  • VertexFitting finished.
  • Association of MC particles almost ready.
    Preparing "realistic" particles in DaVinci to test the tool.

Round Table

Florence
Prepared a new version of the login script CMT.csh that is able to set up the correct environment on SLC4 for both IA32 and AMD64.
It will be announced to the LHCb mailing list.
Joel
Production: problems with the storage for logs (filled my monitoring files), do we need all those monitoring files?
Marco Ca.: We do not need all of them, it is enough to have just high statistics histograms (10kEvts).
[... discussion ...]
There should be a way of having a sort of dynamic workflow. It should generate monitoring files only for the first ~200 jobs, for the others it should be turned off.
Nicolas
Started a new mailing list for CondDB: lhcb-conddb@cernNOSPAMPLEASE.ch. Updated Tools/CondDBUI (not yet tagged), it requires LGCCMT_43b.
Philippe: we need to establish regular meeting for the CondDB Challenge due in October.
We will extend the core software meeting to incorporate CondDB discussions. The next core software meeting (28/6) will start at 10:00 and will be followed by the CondDB one.
Juan
Contacted by RICH software people about their problems with the alignment framework. They cannot use it as it is and they need a way of simulating displacements for volumes that are not DetectorElement.
Juan will organize a meeting to discuss the issues with all the relevant people.
Pere
Asked if there is somebody in LHCb willing to set up HyperNews. There is somebody coming from BaBar.
GENSER needs to be built on the new version of HepMC (fixes on heavy ions). This should not affect us.
Florence mentioned a problem with error messages coming from ROOT concerning dictionaries and that seem related a crash in Gauss (in production). They are not related, but they appear at the end of the log file because of the way the merge of stdout and stderr is done (they go to two different files and then they are concatenated, thus loosing any meaning from a problem-finding point of view).
Marco Ca.
The tutorial in Bologna was very succesful and well attended. It have been asked for a session next year.


-- MarcoClemencic - 15 Jun 2006

Edit | Attach | Watch | Print version | History: r6 < r5 < r4 < r3 < r2 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r6 - 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