Ray,
If it isn't too late, I've made a small but significant change to
SetupHarnessPipeline.py again for r.561. It's not enough to affect the
correctness of our runs, but it keeps us from writing the same log
messages twice to the central log file, which can be a bit of a hassle.
In other news, I did a totally dry run of the new stageleader (stages
now contain only zero-operation memflop events and no files) and found
that our overhead is about 30 seconds per image.
-Jon
_______________________________________________
LSST-data mailing list
[email protected]
http://www.lsstmail.org/mailman/listinfo/lsst-data
