Core Minutes 13 March 2012

Core Minutes 3/13/2012

Present: Joanne Bogart, Jim Chiang, Warren Focke, Heather Kelly, Michael Kuss, David Landriu, Tom Stephens

ScienceTools: (Jim) See Science Tools Development Notes for recent developments.
MySQL migration (Heather) and Joanne managed to overcome one of the more fearful aspects of the migration: testing CMT RM. RM web pages and various build types were exercised using a MySQL 5.5 dbs. (Joanne) The only issue encountered had to do with reserved words. MySQL 5.5 has an additional 40. We use one of the new ones for a table name and another for a column name. This is ok as long as they're properly quoted in sql queries. One of the Perl scripts making up RM had to be modified.

(Heather) We're ready for the move with the possible exception of the OpsLog database: waiting for confirmation from Tony J. on that one. Plans and status are summarized in Confluence.
Core talk:
- rhel5 migration and testing (Heather) would still like to make a systest comparison between CMT and SCons builds of the new P8 tag (as was the case before the meeting hiatus).
  
  GR HEAD builds successfully on rhel4 and rhel5 with SCons.
  
  M-E has been building her stuff against the L1 SCons build (which has new gaudi). There is a problem with that build, however: converters are not getting called for real data, apart from the one for the event header. Heather has been looking into it with the debugger (and has meanwhile built a rhel5 debug version of the new gaudi).
- SCons RM error handling (Tom S.) will change RM to do a better job of capturing errors and saving relevant information as soon as there is a consensus on what's needed. In addition to saving error codes and associated output, other options include sending email notifications and aborting on failure. Email notifications will be sent if there is a checkout failure. [after some discussion] Proposed handling of failure of LATEST tagging:
  - Continue tagging remainder of packages except for SConsFiles, normally done last. That way we will avoid creating doomed builds but will still get as much of the tagging done as possible.
  - Email error status and output to relmanlist for all failed packages.
  - Don't attempt automated clean-up or retagging for now, but keep it in mind as a possible enhancement.
- Macs (Tom S.) snowleopard is probably approaching end-of-life since mountainlion is on the horizon, making a second OS version newer than snowleopard. Will SLAC end support? Even on lion Eric W. has been struggling to get ST to go. In particular, he's had trouble with readline.
- rhel4-64bit (Tom S.) Builds have been turned off. We only have a single redhat4 machine, glastlnx14, and it's 32-bit.
- GR vc90 (Joanne) is waiting on a new, truly debug build of CLHEP. It seems we will need a complete set of debug externals and ultimately, if there is a need for optimized builds, a complete set of optimized externals. (Heather) will see about concocting one. Only non-debug optimized binaries are distributed, which is a mite disturbing. Why doesn't anyone else care about this?
- Archiving (Michael) It turns out the CHS builds were archived under the wrong names and hence don't sort in a natural way. He will rename but, just in case there are problems, he has already temporarily restored the builds to u52. If all goes well he won't need them.

minutes index

J. Bogart, Last Modified: