Slides and Notes 17-February-2010

Which releases do we port? 6.1.6, 6.1.4mc Yes. 6.1.4 probably. There are others! And development!

Donatella needed to help crank out releases. Request time.

Rootd servers

  1. Current one is easily overwhelmed
  2. Instability in xrootd's emulation of rootd? (Which is stable?)
  3. Server need not be coupled tightly to cdfsoft release
  4. Don't want to validate a new root 5 at this time

Stephan knows the list of rootd servers. Maybe we can use the logs to figure what files, how many, etc for last 6 months.

Top Ntuple validation: Rolli can run Perfidia suite on a small period's data (takes 2-3 days), but has no time to generate ntuples herself. Sasha Golossanov (ag) or Hyunsu Lee (hslee) more logical candidates. Takes 3 or more days (with farm! but no SL5 CDF farm yet) to generate ntuples.

After Costas gives preliminary go-ahead after his short test (after his work with TrigSim++--may start later this week), we can announce 6.1.4mc.migrate as a rolling release and request user feedback. Maybe Tuesday?

Final releases will be 6.1.4mc.m instead of .migrate.

6.1.4mc.migrate uses instant_oracle_client, and Stntuple compiled happily.

Target is 1-Aug for upgrade to SL5.

Need list of RPMs for farm

Is SL5.4 OK?

Proposals for release and scheduling need to happen soon. They should include

and the order is to run it by the Offline group, the spokesmen, the physics coordinators, and the executive board.

Adding ifndef __CINT__ around the usages of max_block_size and the function keep CLTD_StorableBank.hh from failing at rootcint level.

Summary:

Modified 17-February-2010 at 12:31

http://hep.physics.wisc.edu/~jnb/cdfcode/17Feb2010
Previous notes Next notes Main slide directory

Please contact jnbt@hep.physics.wisc.edu if you have trouble accessing the information on this page.