Core Minutes 5/17/2005  

Present:    Joanne Bogart, Anders Borgland, Toby Burnett, James Chiang, Johann Cohen-Tanugi, Seth Digel, Richard Dubois, Riccardo Giannitrapani, Navid Golpayegani, Nicola Giglietto, Tom Glanzman, Tony Johnson, Heather Kelly, Michael Kuss, Matt Langston, Julie McEnery, Chuck Patterson, Igor Pavlin, Leon Rochester, Tracy Usher

• Data handling: (Tony)
  • Data server Jean-Paul has finished first phase of the new data server. There is one remaining bug which should be fixed soon. While working on the Data Server, Tony & Co. noticed the last five days of DC1 data look rather different from the first day. For example, there are spikes in distributions of angles for last five days whereas for the first day the distribution is much smoother. Is this understood?
  • Pipeline Matt is working on Pipeline front-end. Dan has been sick -- no word.
  • Data catalog Working on a more generic version such that the web interface, at least, might be used for other applications.
  • Timeline Richard discussed his estimate of what needed to get done when. After a first quick look, Tony believes it's possible.
• RM: (Navid) He is unable to reliably parse error codes from test phase of Windows builds. A similar problem occurs for compile phase for which he has a work-around. The new problem is quite random. When the same build is rerun it can go away. Help from Windows gurus is sought.

  (Toby) could RM say something like RC=5 rather than just Failed? (Navid) The information is already saved in the database; this one should be easy. (Toby) Red-dot page, something he uses a lot, seems to be broken. (Navid) Made many changes last week which could have introduced a bug. He'll look into it.

  Richard requested a status update from Navid next week on RM.

• Data-taking: (Anders) I & T is quiet this week. By the end of the week, should have 4 towers installed.

  (Navid) We're using up the disk space for Pipeline output, so he's started archiving. We learned from SCS that there is a special, industrial-strength method of archiving for large datasets (we have 800 gigabytes) to 200 gigabyte tapes which will be more efficient for us.

  (Navid) There was a problem somewhere, probably an infinite loop (not ever tracked down) causing the MySQL log to fill up. It was large enough that he couldn't look at it with an editor; he just had to delete it.

• Miscellany
  • Absence Richard will be away Thursday through Monday. Most likely he will be electronically reachable.
  • IT & Tea talk on GLAST last week went well, if a little long.
  • Looking for a logo for DC2. Entries will be judged by an august panel: Jim, Julie, Heather and Richard.

previous

minutes index

next

J. Bogart, Last Modified: 01-Jun-2010 15:45:41 -0700