by
YME
- Updated December 22nd, 2006
Tera-minutes of the September 8, 2006 meeting
Present: Laurent, Mathias (in person!), Fred, Yannick, Henry, Mireille, Anis (for the start),
Absent: Chiara, Emmanuel, Anthony
Panorapix (Anis)
Anis has made used several modules from panorapix for the new program
Problems displaying very large images with OpenGL.
The images are split up and divided in sub-images; still problems making this sub-division automatically
Schedule for Anis: Anis leaves on the 8th of November; the final version of the software should be delivered on the 1st of October
By the 15th September the image display problems should fixed
21st September basic image manipulation tools (colour map, zooming, etc...)
1st of October export tool working, astro utils, eg, external catalogues
15th October beta version (with doc and installation)
8th November Anis' farewell Pot!
Points from las week
Fred shows a photograph of the outstanding problems with the ongoing release discussed at the last terameeting
0: SNOOPPIX: runing, 1841 images already transferred to Terapix.
1: problems with fc3/fc4: It turns out the problems are not from the controller but the disks from fc4; fc3 is okay, but fc4 is inaccessible. QualityFITS-out data is stored on the fc3 disks; Fred started of copy of fc3 on fcix2. But we are not sure what is there on fc4.
2: L02/L05 last squares of wides are still processing (mix4 data; all mix5 L02/L05 are ready). Yannick has to check all mix5 L02/L05 QFITs-outr are graded. Mireille will grade QFITSout from mix4 once ready. Laurent remind us the error message sent by [teraop] is no longer valid
3: Laurent started few QFITS-in , still running. still need to run stacks on mix8.
4: Atlas V.3.7.15 still posing problems: it is not possible to compile the LAPACK version on mix9 until a specific GENTOO system tool be stabilised.
5: Fred tested rcp from/to mix8 to/from mix9. seems to work. The rcp problems seems on clix
6: Willot 17 september W1-uband data: no progress
7: Quality-fits no progress, becomes urgent, deadline is Sept. 14
8: Problems with USNOB local now understood and fixed by Emmanuel and Fred. It seems that there is still something not understood/obscure. Likely a problem with the parsing in the input command line when RA,DEC position are expressed in decimal. Fred will sort this out. Note that this different that the error found before with negative co-ordinates
9: backup of release T0003 is around 200GB; we will keep a backup of the data here. The metadata will be copied to CPPN
10: L02/L05, mix4: 69 images to be graded. Still 10 to process. Likely finished tomorrow. Mireille will grade them next week
11a: reprocessing QFITS-in z-band to modifiy z-count plots; not done yet.
11b: cowie data: no progress, problems recovering all the data from CADC; fred will send a message to the CADC about this missing data, cc'ed to Cowie. Still 80 images missing. Fred stopped his QFITS-in and Laurent will run QFITS-in again, in order to get the meta-data in the data base and allow Mireille and Yannick to grade them.
12 power supplies / ups: IAP has made available to Terapix 3 SNMP UPS cards for free. Fred ordered the last one.
Other matters
Yannick and Mathias will save and edit the T0003 table produced by laurent to add the missing information; this happens after Laurent has added the missing completeness information
New position for Terapix to replace Mireille: it seems that the concours was not well advertised, and it turns out that there will be no-one in the jury from Terapix. Note: progress done during the afternoon: Laurent will by part of the Jury on sept. 12 and 15.
06AH18 data from Sanders; Fred will download this data - Laurnet will then QFITs-in them, then Mireille will grade them and will do the processing. Yannick must ask len Cowie what is wants Terapix to do.
mix2 should be cleaned up if you have the stuff there -
SMART disk monotoring tool: nothing done
SPIP 1.9: nothing done. Better to wait for SPIP 2.0 expected by december according to Laurent
Mail sent to root from mix8 and mix9 does not work. Fred is looking at this problem
Mix1: mother board tests: nothing done
next terameeting: Friday September 15, 10:30, Salle du Conseil.