Tera-minutes of the October 2, 2006 meeting

Present: Mireille, Chiara, Mathias (by videoconf), Anis (for the start of the meeting), Emmanuel, Fred, Laurent, Yannick

Absent: Anthony, Henry

PI data : WIRCAM (Chiara)

 In progress: Bouvier' data. serious photometric calibration problems with Y-band data. Mag. ZP in Y band based on standard stars not provided in FITS header. CFHT only provides ZP from 2MASS, but in J band. Need color correction, but unclear how an absolute Y magnitude can be derived. Chiara will ask Jerome Bouvier whether he got already from past runs some Y-calibrated data in his fields.

 McHardy data on the way to Terapix from CFHT; almost fully downloaded.

 The reprocessing of Lin Lihwhai data using new Loic pre-procesing tools does not seem critical and is not an urgent task.

 Pending data: L. Deharveng, C. Willot, L. Albert

 It would be useful to have an idea of the amount of data of each PI program. Fred suggests the wget command may have options to return this information. We could also ask Loic to provide this information.

PI data : Megacam (Mireille, Laurent)

 All Cowie data received at Terapix have been processed. Output data and metadata put on ftpix and retrieved by Len Cowie by Sept. 29. Still waiting for missing images Elixir-ed by Kanoa, but the 1/2 Megacam files.
 Sanders (06AH18) has now been downloaded to the TERAPIX cluster. Laurent will run QFITS-in on these data this week.
 Kneib (03AF01) : Laurent will run QFITS-in on these data today.
 Robin (04AF09 and 05AF14): pending, waiting for a new Scamp.

Panorapix (Anis)

 Anis Valgrin-ed Panorapix and found several bugs. The way MEF are handled by the current Panorapix version is not satisfactory and makes Panorapix unstable. Anis is re-writing the FITS/MEF handling module prior to move ahead. Goal: the FITS module ready before the end of the week.

Scamp (Emmanuel)

 Several bugs fixed in the pattern matching. Better contrast obtained.

 Emmanuel will work on the XML output this afternoon in order to have this option ready for the next version of QFITS. No impact on SPICA developement.

 In progress: removing the "scalop"-shape derived from the astrometric solution using a better weighting scheme. We (all) try to find typical/extreme cases with scalop-shape images for Emmanuel.

QFITS (Fred)

 Fred implemented the selection of sub-areas on each CCD that are used to get full resolution sub-images as quality assessment. CCD corners and centers are selected using WCS information. The current version uses the PGPLOT lib., but the final version will use PLPLOT. Expected delivery date: Wednesday October 4.

SPICA: (Laurent)

 Few bugs corrected (PI/Wide RunID identification mismatch, access to files/metadata with errors, etc..)

 In order to fill completely the public T0003 table, several links must be added for each stack that cannot be stabilised until all T0003 metadata be archived on ftpix.

Hardware:

 Pix3 RAID rebuilt with 4 disks only. No need to have more since this node is for SPICA testing only.

 The Migration process of mix8 to other nodes got stuck by the unexpected cosmix problems last week. The migration is posponed to later this week, once Fred will have finished other urgent matters. In the meantime, Laurent can run QFITS in Sanders and Kneib PI data.

 Cosmix had a major failure: 1 disk dead, filesystem corrupted, lost+found dirrectory contained several files from Liping. Fred backuped the /home on fcix2. Very long : this /home contains 1TB of data. Fred installed a new OS by cloning pix. Fred proposes we buy a new cosmix for the system, all servers, etc... and move the old cosmix as a cosmix1 with simialr usage as other cosmix*

dotproject: (All)

 W1 u-band for willot: pending, re-scheduled for October 12.

 Replication of mix8 onto the cluster : delayed due to the failure of cosmix.

 SPIP-1.9: waiting for 2.0 (December)

 QFITS: expected delivery Wednesday October 4.

 Power point presentation 2005 CFHTLS meeting on the web: Yannick's task, not done yet.

 SExtractor: simultaneous ASCII+Binary files ,from Laurent request, not done yet. Pending bug on background subtraction on giant image put to high priority for T0004.

 SCAMP: in progress

 T0003 documentation: Yannick's task, in progress, 85% done.

 W3 advisory: Henry's task, nothing done yet.

 EYE retina: pending.

 UPS: posponed to Nov. 1.

 Update Web page on software news: will be removed and repalce by the Forum.

 Archive T0003 on ftpix: this week.

Next Tera-Meeting: Monday Oct. 9: 10-12 hrs