IPP Progress Report for the week 2013.03.04 - 2013.03.09

(Up to IPP Progress Reports)

Eugene Magnier

Serge Chastel

  • Thu AM: Training session on Sexual Harassment and Violence at the Workplace
  • IPP
    • ippdb02 back on Friday. Started rsync from db04. Caught up on Monday.
    • mysql servers that crashed during the week on the storage nodes. Monday: stsci08); Tuesday: stsci09, ipp020; Wednesday: stsci03, ipp064, stsci08, stsci01, stsci06; Thursday: stsci02; Friday: ipp062, stsci01, stsci02, stsci06, stsci08, stsci03, ipp021; Saturday: ipp049, ipp012, ipp055; Sunday: ipp021
    • IASC outreach for nights 2013-03-08 and 2013-03-09. Updated documentation and wrote a few lines of code to make sure smfs of exposures were published
    • Problem with MegaCam? images in psastro which are actually related to a ppImage failure: only part of the cmf file is written. Gene explained me how to "fix" that by manually running psphot. Spent some time on the problem with Heather on the problem though
    • I can run the IPP without the gpc1 database and without pantasks till the warp stage. I need to figure out how to correctly call ppSub now.

Heather Flewelling

  • horrible discovery: we have duplicate addstar processing (for a given add_id, run 1-2 times according to the log file)
    • addstar does not block this
    • addtool doesn't complain when it attempts to add a processedexp and one is already there
    • why is this happening in the first place?
  • blew away an init batch for conrad
  • various work with CA/ROTSE, and MH/JT
  • started gene's delstar to clean up duplicate mess - done on slice 08 and slice 10
  • checked (with Gene's help) the minis to make sure the duplicates were gone (they are)
  • psps wg meeting
  • friday vacation

Mark Huber

Bill Sweeney

Chris Waters

  • Thilker galaxies are still running. Defining the stacking queues was more challenging than expected to reduce the amount of stacking/searching for correct skycells for a galaxy. I think this is now effectively complete, other than the processing time.
  • Three different stack related issues that are likely not actually three different issues. It looks like the excess detection/Kirsten Larsen stack coverage/Eddie's stack photometry issues are all related to an as yet not fully understood issue with warp weighting during the stacking. This is the main wiki page on this issue, with links to other wiki pages describing the individual issues.
  • Trail test data reprocessing.