Minutes of WFAU VDFS Science Archive meeting:  15th December 2006
-------------------------------------------------------------------------
-------------------------------------------------------------------------

Present:       RSC, MAR, NJC, ETWS, LGR
Apologies:     JB, AL, NCH (mid-life crisis in Hawaii), 
               JPE, RGM, JMS, JDT, BCW, MSH, PMW

DONM: Provisionally 10am, Friday 5th January 2007 in the Plate Library 

Congrats to Nige who despite the stresses of herding cats aka WFAU/VDFS
has reached the ripe old age of 40. Unfortunately the Eclipse Malt Whiskey
has failed to arrive yet so no formal toast was made.

Actions discharged:
-------------------

There were no actions from last week's meeting.


Actions partly discharged but continuing:
-----------------------------------------

The following actions from previous weeks are partly done but continue:

ACTION: JB to rearrange external catalogues onto batch catalogue
         server.
Continues.

ACTION: RSC to profile CU4 and investigate possible optimisation.
Continues.


Specific points and new actions:
--------------------------------

Comments and issues arising from CASU fortnightly minutes:

On the subject of UKLight, Andy Lawrence has already been prodding the 
various parties on behalf of WFAU. 

Networking:

No news on UKLight.

ETWS noted that transfer of BestDR5 from Johns Hopkins is done (2.7MB/s).
The installation of the files to Thutmose will happen in the new year. 

WSA Operations:

JB reported that a WSA backup had been started. Quality Flags for the GCS 
and GPS have also now been done. Some tidying of the windows servers has 
been done to gain a little more space on them.  Ahmose's backup software
is now back in working order.

Replacing the broken databases on Thutmose caused by the disk failures is
now in progress with IRAS, FIRST and ROSAT having been done, UKIDSSDR1
being in progress (as of Tuesday night) and UKIDSSDR1PLUS next in line.
USNOB will have to be recovered from tape as we do not have a copy of this
database on Amenhotep.

MAR noted that the WSA backup had just finished and that UKIDSSDR1 on Thutmose
(used by Astrogrid and bestDR3 joins) was put back online on Wednesday 
morning. UKIDSSDR1plus might be copied over to Thutmose over the weekend.

Hardware:

Thumose's disks have now, in theory been fixed and there are now two
spares for Ahmose and Amenhotep; a spare disk for Thutmose should arrive
on Friday from Eclipse.


Software:

JB reported that he had tested multi-passband non-surveys with CU7, 
they work, and that he had found a minor bug in CU19 (dropping indexes?).

RSC and NJC reported that they had been concentrating on the re-calibration
work and had performed a series of tests to check everything was as it should
be. It was estimated that the update (FITS files and database) could
take around two weeks. The main overhead being the database querying to
match up filenames (as these minutes are being written, an improvement in 
performance is thought to be in the pipeline). 

Staff movements/absences over the next few weeks meant that monitoring
the re-calibration update and re-starting it in the event of hardware
or network issues might prove challenging but all efforts would be made.

RSC raised the issue that some SQL/CU updates might be needed to ensure 
non-survey releases contain the relevant tables related to any re-calibration.
[ACTION RSC]

ETWS reported that he had been continuing work on the wrappers to 
parallelise ingest CUs.

MAR had resumed QC at the start of the week but had been delayed by the
WSA being offline while the back-up proceeded.

Non-survey Data Release:

U/SERV/1678 Quality Flagged, Source Merged, Indexed, Built, Copied to
Amenhotep and Attached and Released.

Astrogrid deployment:

No news.

Miscellaneous:

Merry Christmas