From lindy@ligo.mit.edu Mon Mar 10 01:40:50 2008
Date: Mon, 10 Mar 2008 01:36:26 -0400
From: Lindy Blackburn <lindy@ligo.mit.edu>
To: Keith Riles <kriles@umich.edu>
Cc: Laura Cadonati <cadonati@physics.umass.edu>, Shourov Chatterji <shourov@ligo.caltech.edu>
Subject: new version of H1-H2_lockloss, etc flags

Hi Keith, Laura, Shourov,

I'm not sure of the procedure to update DQ flags, but I made a new "v2" set
of flags for H1:H2-lockloss, etc:

http://lancelot.mit.edu/~lindy/s5/scatter/flags

This addresses a bug in my previous lists where day boundaries were
identified as lock loss or lock acquisition if a Not_Locked flag intersected
a day boundary.

Since this could only happen if one instrument was out of lock, I don't
think it can impact triple-coincident livetime easily (there would need to
be an brief (minutes) lock loss just overlapping a day boundary). But I've
asked Shourov to see if the new flags also show the 2-3% impact in triple
coincident livetime that the old ones had.

Thanks,
Lindy