Log entry time 12:30:40 on February 13, 2011

Entry number 346664

This entry is a followup to: 346657

For now, shift workers please try doing this:

If at the end of run, the usual end-run GUI pops up, everything is
fine (my present hypothesis, see below).  If not, wait a few seconds,
don't start another run yet .... type ~/scripts/end_all from a-onl
account on adaql2.  This should pop up the GUI and you'll get end-run
info.  Observe whether or not this procedure results in 100% success
rate of halog end-run entries.

--------------------------------------------------------------------

Added notes

1. If this works, it suggests CODA 2.6.1 is sometimes not calling
end_all.

2. If it doesn't work it suggests a problem with the script end_all.

If 1, I'll investigate more tomorrow, perhaps with the DAQ group looking.
If 2, I guess it's easy to fix (??).

Hypothesis: Sometimes (~80%) the end-run halog entry is not made because
the end-run scripts are not called at all.  This would be a new problem.

It may help to consider the difference between start-run and end-run,
since start-run script seems to always work.  E.g. maybe something
messed up in the COOL DB ?  I'm at home, so I dare not make too many
tests and I cannot pop up any GUIs.