Core Minutes 4/26/2005  

Present:    Joanne Bogart, Anders Borgland, Toby Burnett, James Chiang, Seth Digel, Richard Dubois, Zach Fewtrell, Riccardo Giannitrapani, Tom Glanzman, Navid Golpayegani, Tony Johnson, Heather Kelly, Michael Kuss, Matt Langston, Chuck Patterson, Leon Rochester, Tracy Usher

• I & T data-taking: (Richard) Test runs did not happen yesterday as planned; likely will happen today. That will be followed by a couple days of calibration runs.
• Data handling: (Tony)
  • Data server Jean-Paul is close to having a complete data server. So far have been using it with DC1 ntuple data. The interface allows you to make a a search to determine how many events satisfy your cuts, run a batch job to fetch them, and save your search criteria for another session. Tom is making slow progress on the event peeler (that facility which will return complete events selected by [run number, event id] ).
  • GUI for Installer It's coming along.
• GlastRelease and weekly tags: (Toby) His understanding was that he would make a tag about once a week (special tag id incorporating the date) and that Sys Tests would be run on these tags. Then, depending on the results, the tag might or might not become a (recommended) release, however Julie is opposed to running Sys Tests on all these tags, and she only has one slot for Sys Test results on HEAD builds. Also if the builds are not to occur automatically, he needs to know how to do it manually. (Richard) The single HEAD slot is not a fundamental restriction. Julie is concerned about the interface: all those builds will tend to obscure the ones a user is likely to want to see. (Leon) has been talking with Julie about ways to separate the HEAD builds, e.g. by means of a submenu. (Richard) We also need a way to delete unneeded HEAD Sys Tests after a while. (Toby) Once a HEAD build is accepted to become a Release, would be preferable to just rename it and associated Sys Tests, rather than redoing everything. (Richard) As for the issue of how to run Sys Tests if you're not Julie: Heather has done it and can advise. Julie has written up instructions in a Confluence page.
• EM releases: (Heather)
  • Current EngineeringModel tag is v4r060302p18, which, since Leon rearranged the dead strips calibration file somewhat, seems to be ok [more on this topic below in the Calibration infrastructure item].
  • There is new CAL code, not yet incorporated in EM HEAD because it doesn't all build on Linux.
  • Since the fix to reconRootData treating the three TObjArrays for TkrRecon as TClonesArrays, memory use has jumped in the system tests. Some increase was expected, but not hundreds of megabytes - Heather continues to look into that problem.
  • Some rearrangement in event id handling to provide just the new 32 bit version is coming soon.
• 2-Tower Calibrations: (Leon) Tracker is all set, to the best of anyone's knowledge. (Zach) Code in calibGenCAL is in beta test. For most calibrations for the near-term at least, procedure will be to calibrate towers individually, then merge the results. He could use some test files. Anders volunteered to provide them.
• Calibration infrastructure: (Joanne) New tag CalibData v0r12p1 will handle bad strip files like the one that Leon had to rearrange, however a more complete solution to the problem of keeping track of types of strip badness will require a change to the dtd (i.e., format of XML file). She and Leon are working out the design details, but will wait to implement until after she returns from vacation (May 4-13).

  rdbGUI documentation has been updated to reflect recent enhancements. It also includes a new Use Cases section, with recipes for a couple of the more common non-atomic operations.

• RM news: (Navid) For Linux RM, he has been working on the web interface. Hot issue at the moment for Windows RM is how to get back correct status. Sometimes perfectly good builds will set a bad return code, so Navid is investigating other ways to determine the status.
• Disks: (Richard) He's been talking to SCS about more disk space for both Windows and Linux.

previous

minutes index

next

J. Bogart, Last Modified: 01-Jun-2010 15:47:18 -0700