by
HJMCC
- Updated February 27th, 2005
Terameeting for the 25th of February 2005: Absent: Gilles, Gerard Present: Laurent, Yannick, Mireille, Henry, Fred, Emmanuel, Jean-Christophe
Release T00002: problems (Laurent, Henry, Yannick)
50% of stacks do not have any scamp plots because there are NaNs in the output from scamp, which prevents them being loaded into the database. png images of scamp diagnostic plots do not appear on these web pages.
Emmanuel suggests that is because there are instrument groups where there is data in a certain filter which are not contiguous except through a second (different) filter. This means that although the astrometric solution is constrained, the photometric solution is not. Emmanuel proposes to modify scamp so that the grouping is computed differently
Also, if scamp crashes in some cases the swarp continues, WITHOUT the header files. This can produce a final image which is completely wrong (no astrometric solution / no photometric solution).
Yannick asks that the list of images processed by spica is displayed on the web pages using the pre-registration panel created by Laurent that turns out to be very synthetic and useful, with access to QualityFITSITS information of each image used for SWarp or SCamp. Yannick asks Laurent to add information on data status for each targets: done, running, pending.
Yannick asks that comments on images are initialed (HJMCC, MD, YM)
Version control: Laurent will allow SPICA to run with software which can changes their version number DURING the release. This will allow improvements and bug-fixes to be made during the release without having to re-start all the stacks.
Processing of the release will continue so as to find any other potential bugs.
Data Transfer to CADC (Jean-Christophe)
Data transfer not started, JC waiting for go-ahead from Yannick. Likewise for the release of data from very-wide.
PI data (Annie Robin, Florence Durret)
This data will be transferred to the nodes and QualityFITS will be run.
QualityFits
JCC has been supplied with QualityFITS for tests (and associated software), under the understanding that it (and associated software) be absolutely not be distributed. QualityFITS documentation underway.
Skywatcher
several upgrades; CFHT targets lists added; work been done on the web interface; cuts in the dust maps updated.
Defectix (Anthony)
Neural net recreated. AB will tell us when defectix will be ready for testing. Furthermore, defectix will provided under the "subversion" version control system.
Documentation provided on the TERAPIX intranet of the "subversion" documentation control system.
SAC document
Yannick is currently preparing the CFHTLS report for the SAC; awaiting input from the CFHTLS community.
installation of SPIP2
can SPIP be upgraded? Perhaps we can back up the existing web server on another machine which can then be upgraded so as not to interfere with TERAPIX web server.
Gerard points out that there are many pages which will have to be modified
Gerard no longer works for TERAPIX after the 1st of March.
Gerard will be asked to provide the latest version of the "panier" software tool attached to DBTERAPIX before leaving.
Other matters, hardware
Installation of wiki: yannick asks that the number of different types of groupware (wiki, SPIP, dotproject) is limited otherwise it can be very difficult to find information on the TERAPIX intranet.
fc7 disks are full, and Laurent and JC need space for incoming and outgoing data. We should buy another fibre channel for the PI data which currently stored on fc7.
Problems with mix2: mix2 still crashes about once per day. We still have a replacement motherboard; we will wait until after the release is finished before making this change to minimize disruptions.