Tera-meeting of 19th December Present: Fred, Gerard, Gilles, Henry, Emmanuel, Laurent, Mireille, Yannick

This is the last 2003 Tera-meeting. Merry X-mas to all!

--> Next Tera-meeting: Friday, Jan. 9, 2004, 15:30.

Panorapix (Fred) Cfitsio library almost replaced by the version C++; Should soon be soon. 8 FITS bit mode not supported yet; Demo before 27th December. MEF handling running soon.

VO-tables (Gerard) Gerard demonstrated how an XML VO-table could be exported from an image in the panier. Important: UCD descriptions have to be homogenised with ours this has to be investigated (maybe by Yannick and Emmanuel using the informations available at CDS), in and UCD keywords corresponding to MEGACAM keywords must be identified. TERAPIX should propose UCD list definitions to CADC and CFHT by January as a first tentative.

Scamp (Emmanuel) Main problem currently is that the astrometric solution does not work: general approach to finding astrometric solution; difficult to debug. Advances slowly. Some added visualisation debugging utilities added. Most basic plot tools developed and implemented in Scamp by Emmanuel. Uses the PLPLOT library.

Swarp (Emmanuel) GOODs people (Dickinson et al) found projection bug in the latest version of swarp. Report sent to Emmanuel.

Filename problem (Laurent, Gerard) Files arriving from the CFHT reprocessed by Elixir have .phu extension. For now the FILENAME will be the *actual name of the FILE*.

Dbclient bug (everyone) Some images crash the database; others do not. We don't understand why this happens. Headers of FITS images (ones which crash the DB and ones which don't) examined. Bertrand will try to help to fix these issues. Critical issue that sticks SPICA and the MEGACAM processing at TERAPIX.

Checksums (Fred, everyone): Datatransfer makes the checksums for the incoming images. If the images are uncompressed (not all images arriving from the CFHT compressed) they will be compressed.

CVS (all): Please remember that all software packages must be loaded on the TERAPIX CVS (we forgot to discuss this; this is a reminder).

Data Estelle Moreaux (Gilles) Gilles made table showing the astrometric accuracy output from Mario's program for all input image. The coadded FITS images and the control files are now available to Estelle and the TERAPIX staff using a securised web access.

Hardware summary (Emmanuel, Fred) Opteron will be 'loaned' to Henry (in 32 bit mode) for 15 January CFHTLS meeting.

3/4 of the nodes upgraded to MP2800 processors.

Memory for node#3 installed.

'testtix' will be upgraded to MP2800 (using AVO money)

Still awaiting other components for the CFHTLS science cluster.

Photometric calibrations (Mireille, Henry) Mireille will check new ZP and photometric data on the new AF19 released data. Will check whether MAG_APER with a 14" diameter aperture returns similar values as Landolt original calibration data.

Henry will cross check Mireille's results to validate TERAPIX photometric procedure that will be implemented in SPICA.

ADASS proceedings (all)

We are all late... waiting for strong warnings from the editors....

15th January CFHTLS meeting at IAP *important* (everybne)

Already 47 personns registered.

Priority 1 deg^2 (5 filters)+ 4 deg^2 (2 filters) for the CFHTLS/IAP meeting. Preferred: W1 and D1 targets. Need to check if D4 could be better.

How to reduce these images? The Opteron will be installed in 32 bit mode (Emmanuel, Fred). Henry will use this machine to reduce the data. Astrometrix/Photometrix must be installed on the node. (Henry?). If Opteron not working, "stetix" will be installed as node 10.

Also it's extremely important to check the photometric calibration which has been carried out on the images.

QualityFits will also be run on *all* survey images to provide quality assessments of all available images (not just the images in D4).

For now, QualityFITS will be run without attempting to load the images into the database: the processing will be repeated after the meeting, once the problems with dbclient are resolved.