IPP Progress Report for the week 2011.02.28 - 2011.03.04

(Up to IPP Progress Reports)

Eugene Magnier

Serge Chastel

• MOPS tool for processing: specs + development
• MySQL databases optimization. Full report here. Conclusions:
  • Nebulous optimization took 2 days.
  • Nebulous huge tables sizes have decreased of 20% / 25%.
  • GPC1 optimization took about 2 hours
  • GPC1 tables have globally shrunk in size by 27%.
  • It would be highly interesting to optimize in one week?/month? to see if optimization lasts so long.
• Other:
  • apache log of ippdb00 has been moved to /export/ippdb00.0
  • ippdb01 MySQL should be cleaned (1648 tables in total while gpc1 is 96 tables).
  • ippdb02 doesn't work as expected for backup.

Heather Flewelling

Roy Henderson

• PSPS
  • added user instructions to ippToPsps documentation (as I realized, during my recent absence, that no one but me could run it)
  • some time pulling various images, stacks and cmfs for Jim for MD04. Time figuring out filter type of new stack cmfs that were missing headers
  • 2 new passes of stacks from Gene. 1st:
    • extensions names changed from demo file
    • all fluxes invalid in psf extension
    • SkyChip.xsrc and SkyChip.xrad extensions empty
  • 2nd:
    • now populating ApFluxStDev from SkyChip.xrad
    • headers empty
    • work on xfit extension: Sersic, deV and Exp models. Plan to load FITS straight into a Db in order that I can cleanly query the data. Alternative is to write lots of code to do what databases already do.
  • walk-through of nice new DXLayer from Conrad. Loading some test batches to test it works. It does!

• IPP
  • added 'interested servers' list to XML config for roboczar
  • added comments to Config class and XML config

• Other
  • sick on Monday
  • general catch-up after two weeks off
  • lots of mailing-list admin

Bill Sweeney

• Spent a day learning to use DVO and analyzing why the PanPlanets project is reporting bad astrometry. We belive that the problem is due to stars getting damaged by burntool correction. The large number of brigth stars in the STS fields makes these exposures more sensitive to this problem. Gene has an idea on how to improve the situation.
• Gathered up detrend images used for STS processing and put them on the data store for MPG to download.
• Fixed problem in dvoImagesAtCoords where it found that a point on the sky touched two chips.
• Wrote a script to parse the host configuration files to report how hosts are being targeted for procesing
• Fixed some problems with destreak cleanup for non-replace mode runs (raw stage).
• Found a bug in psphot that caused some older data to fail updates. Implemented fix suggested by Gene.
• Analyzed several bursts of postage stamp errors. Most are due to user error. There are some cases where transient "file not available" errors are treated as "file gone errors".

Chris Waters