Core Minutes 6/10/2008
Core Minutes 6/10/2008Remaining before launch: (Richard) See the list of what's left in Confluence. [And, for more detail on OBF/Config, a new page Heather just made.]
Interleave (Richard) ..is still broken. Deadtime is not being filled in properly. (Toby) is trying to track it down but no luck so far.
Simulations (Richard) Requested sampled day of pointed observations is in progress
LPM effect (Richard) Is it ready to go? (Philippe) Has been tested in BeamtestRelease and looks good, but haven't tested everything; for example ACD response. Suggest the code should go on a new branch. (Richard) If we put it on the main branch it is easy to make standard test runs.
Filter bits (Richard) Heather's part of handling real filter bits is done. What about the merit tuple? (Leon) Haven't pursued it. AnalysisNtuple is way ahead of releases; would like it to get caught up. (Heather and/or Tracy) This has to happen anyway for new OnboardFilter. (Heather) Need to think about exactly what gets put in; not all filters have information as complete as gamma has.
Disentangling FluxSvc (Leon) Work is ongoing to
(Toby) astro, the home of gps, now builds a shared rather than static library. This takes care of the old problem of multiple copies of gps in different component libraries.
(Richard) What about stellar aberration? (Toby) Will be easy to finish off once it's decided where PointingInfo will go.
On-the-fly Filter config. (Tracy) OnboardFilter can get information from Moot; that's last week's news. He has also tried it with the new package organization (MootSvc class moved out of CalibSvc into its own package) and that's fine. He was able to retrieve acquisition mode from the data, the last piece of information needed for on-the-fly configuration.
It turns out we can't just use the SLAC Flight Software installation for Linux since there are remote Linux users, so he will package up an obf external for Linux as well as Windows.
(Joanne) Concerning MootSvc: on beyond the version in its own package that Tracy is now using there will be another tag in the near future, hopefully by the end of the day, including services Eric C. has requested for getting at filter configuration and GEM configuration. Code is written and builds but hasn't been tested.
Documentation: (Chuck) Right now he's buried in pictures from Florida. He believes main items for How-to-fix are all in place but there are likely to be some holes. Review and comment from experts would be much appreciated!
Core Face-to-face Meeting (Heather) Still aiming for the week of July 7th.
GR tags (Heather)
Reprocessing (Tom) The reprocessing done by Richard using a HEAD release was successful — only 1 stream out of about 3000 failed — but maybe it needs to be done again anyway with an official tag. (Richard) Liz found some runs with duplicate events and there may be others, so some runs that claim success are still suspect. It would be well to redo, preferably using a tag from the v15 branch. (Heather) believes everything on the reprocessing branch has by now been incorporated into v15r<something>.
Skimmer and CEL [Thanks to David for the following]
The proof of concept is done that the Skimmer can generate a CEL as output, instead of real ROOT files. This only applies to the data types that can currently go into a CEL, i.e. not pointing history and jobinfo. The final tests and commit are on the rails.
For what concerns how to upgrade the CELs when the data files are moved, the current suggestion is to write a program which transforms all the file paths within the related CEL. An example of such a program can be found in CVS: rootUtil/apps/celRelocate.cxx. One just needs to rewrite the function relocateInputFilePath (which currently simply removes the directory part of the path).
New RM, SCons (Navid) The new RM is running in parallel with the old. He has been working on a gui for the tag collector; it's going a little slowly since gui interfaces are new territory for Navid. (Heather) The old tag collector is not functioning. (Navid) The tag collector script, existing only locally on glastlnx02, was erased when that machine was upgraded to RHEL4. He will sub as human tag collector until the new one is done. (Jim) uses the tag collector regularly. The upgrade to RHEL4 was a while ago, yet he used the tag collector shortly before going to Florida. [Navid and Jim will huddle over that one.] (Navid) The new tag collector uses the new tagging convention, so the new RM will do the opposite thing from the old one: that is, generate old-style tags from new ones.
previous
|
minutes index
|
next
 |
J. Bogart, Last Modified:
01-Jun-2010 15:45:09 -0700
