Agenda and Minutes, 26 September 2007
Agenda: Wed, 26 September 2007
Minutes
Attendance: Brian Scott Patrick Adam Filippo Ben Greg Stuart Charles Kent Ed John Z Kipp Steve Carsten Keith T. Announcements: Patrick to follow up on time of the call. Stuart announced that there is years worth of coincident data. The next calibrated data release will be coming after the measurements are completed post-run. DMT: John Z reported that the xml reading is not working correctly with the rpm. John has been having problems with tablepgm built and installed in the rpm. He has noticed a couple of bugs in metaio as a result. Ed suggested to check that the same configuration options are being used for the rpm build and the manual build. Ed asked if we could have people use valgrind to check for problems with memory allocation in search codes. Adam will take a look this under a debugger and strace. GEO++: Asked Steve and Carsten to ask if this is still active. Glue: No report. LAL/LALApps: No report. Stuart asked about Sukanta's problem with the frame writing code. It is not solved, but will be followed up. LDAS: No report. LDG Client/Server: Greg reported the VDT version 1.8.1 came out yesterday and is building ldg 4.5 out of this. Patrick asked if all the issues relating to scripts are in place from Warren. Greg will test this for Mac OSX. Warren reported that it is the perl modules that is the central thing. Warren commented that in the process of helping somebody renew their certificate he realized that the script does not use quotes around passphrases right now. Scott asked about myproxy. It is in. LDR: Scott reported that he has been working with Gerald to recver their LDR from hardware failures on their side. Scott will take what is running at LHO and add the bugfix for Filippo and then that it would be deployed at all the Lab sites. Scott reported that the Filippo bug is a parsing problem associated with an underused method. Scott will try to do this either tomorrow or Friday. Ben gave an update on the publication of S5 hoft data. About 25% of regeneration was not published at Caltech; it should be published very soon if he identified the fix correctly. LIGOTools: No report. MatApps: Nothing new this week but will get a tentative list for rpm for next weeks call. Metaio: No report. Onasys: No report. OSG-LIGO: Britta is starting to look into possibility to run E@H on OSG. We have scheduled a telecon for Friday @ 10am. Kent reported they are also talking to the Pegasus people to understand if this can be done on TeraGrid too. LSCSoft: Adam reported that the FC4 testing repo now has builds of the latest gsl 6.10; he will announce them by e-mail. Stuart asked if the problem with the gsl info package. In 64bit it was working just fine; Adam also reported that one can force it. He also reported that there is a repo for CentOS 5 with libmetaio and libframe are in the testing repo x86_64. Patrick explained that Adam will be working on gravity for the next few days and will switch to resolve the GSL issue properly right after that. AOB: Scott asked if CIT has moved to Condor 6.9.4 yet? Not at CIT. Warren asked to get a report on the NR data format on a future agenda. Patrick will follow up on this. General issue of doing better test/build facility to improve our ability to detect problems. Ed suggested trying to use valgrind to help detect memory issues.
$Id: 070926.html,v 1.2 2007/10/31 16:59:39 patrick Exp $