Computing Services Team Activity for the Period 6/02 to 10/02 - MEC - 10/14/02

HET

Spiesman, Odoms, and Cornell continued work on the moving baffle, and had
an initial science commissioning run with MacQueen and Hill.  Other than a
few bugs, the major remaining issues are calibration of the baffle
geometry, and allowing for the fact that the tracker can move to places
where the baffle cannot follow. Zhang resolved a bug in the baffle-enabled
version of HET TCS relating to setting down the telescope at the proper
azimuth.  Odoms modified the baffle computer software to permit operation
of the Hartmann test fixture.  Zhang began learning LabView programming,
and studying the SAMS LabView software.  Wilkinson wrote an initial version
of a night report stack data formatter, and the HET night staff evolved
that into a more integrated solution that is now currently in service.
Odoms and Cornell worked with Ward to define an ICD for the motion control
code for MARS2.  Cornell helped Wolf select a camera and frame-grabber for
the instrument, as well.  Wilkinson continued to work on the automation of
DIMM data reduction, including integration into the weather
system. Wilkinson began work on a new hardware architecture for the DIMM
system which will allow remote control and setup when the system is
installed on the new tower.

Mt. Locke Operations

Zhang continued to fix bugs in the 107TCS software, completed the initial
port of 107TCS from LynxOS to Linux, and tested this successfully on the
telescope with Spiesman and Odoms.  A few bugs were identified and fixed.
The remaining barrier to deployment is our current serial port card, which
appears not be compatible with standard linux drivers.  If Wilkinson and
Odoms cannot adapt existing drivers to work, we'll purchase a new card and
test in a future daytime engineering run.  Spiesman tuned the 107TCS PMAC
software for the new power amps.  Cornell and Wilkinson installed a new
guide camera for the 107-inch, and a computer to control it.  The power
budget was a significant issue.  Odoms and Cornell created new autoguider
software for the 107-inch and the Apogee camera.  Odoms and Zhang resolved
a communications problem between the new apogee software and the LynxOS
TCS.  Wilkinson worked on the TAEC heliostat to improve tracking offsets
induced by PC-TCS software inconsistencies. He has also started to design
an alignment system for that instrument, which will likely double as an
exhibit display.  Cornell assisted in planning for the ROTSE and MONET
telescopes, spending considerable time on the issue of CCDs for MONET.

Austin Operations

Landazuri took over the empty slot created when Vargas left us in January.
The focus of his position is installing new computers, and day-to-day
maintenance.  Landazuri worked on 15 systems since he started with us five
weeks ago.  Anderson and Cornell continued to make improvements to the
Department and Observatory web pages, adding, for example, ADS and astro-ph
links for all faculty, researchers, and grad students.  Anderson updated
the faculty and staff listings, as well as the committee and research group
memberships, made the initial postings for the Fall 2002 courses, and
continued weekly postings for Wheeler/Bash/Harvey, etc., courses. Umbarger
took over maintenance of the Department and Observatory web pages when
Anderson departed in September.  Umbarger began work to improve navigation
on our web pages, and to update the online material in the areas needing
the most work, namely the HET and research group pages.  Wilkinson
developed a new methodology for booting the grad PCs over the network using
DHCP and TFTP. The new system was integrated into the latest generation of
13 Redhat 7.2 grad workstations which were deployed by Schaefer over the
summer.  Wilkinson and Landazuri will migrate the older Redhat-6.2-based
grad PCs to the new 7.2-based system as time permits.  Schaefer assisted in
10,000 office moves over the summer, and helped protect computers during
the sprinkler installation.  Wilkinson and Cornell migrated the linux
software collection from zantac to preakness.  Cornell updated much of the
science software available on the Suns and Linux PCs, including IRAF
v2.12.1 and all external IRAF packages, IDL v5.5, the Goddard IDL library,
and atv, calfuse-2.1.6, ciao-2.2.1, and xvista-6.01.  Wilkinson brought the
graduate student terrabyte storage array online after a large amount of
hardware and software debugging. The new RAID system is stable and
operating at 90% of the theoretical network throughput.  Wilkinson is in
the progress of developing a new DHCP/NAT system for installation on the
local Austin network. It will provide dynamically allocated addresses for
all laptops, visitor systems, and some specialized computer labs.  This
project is necessary to address our network address crunch, and will make
laptops easier to carry from one network to another.  Wilkinson wrote a new
web-based interface to the "skycalendar" program.  Cornell updated the
server software on hyperion, improving the response times of the department
web server.  Wilkinson set up computers for the ApJ Letters office, and
began building a new server closet in the back room in RLM 15.320.  Collins
and Cornell renewed the Sun hardware and software maintenance contracts.
Collins renewed the IDL and Maple software maintenance contracts, and well
as the TACC accounts.  Cornell finally did some performance evaluations.
Collins purchased 2 Sun workstations, 24 PCs, 7 Macs, 5 laptops, a CCD
camera, a couple of printers, a UPS, plus lots of memory, computer cards,
cables, software, and supplies, and coordinated numerous repairs during the
period.  Cornell dealt with a round of hacker attacks on the department and
Mt. Locke Suns.  Schaefer left the group for a research position at CSR.
Recruiting continues for replacements for Spiesman and Schaefer.