[evlatests] Software time Feb 22

[Barry Clark]

A confusing day.

A network configuration problem (since fixed) prevented running Idcaf on
Igloo, so that the standard route into the data archive could not be invoked.
However, multicasts were routed to the VLA subnet, so that I could run 
the Antsol screen on Pokey.  The Antsol screen invoked from the operators
screen did not appear to work, although a separately invoked Ansol screen
did.  Under investigation.

Jerry Langevin was at antenna 14, using the Executor to set the antenna up
at various bands.  This was causing the Executor to organize a single dish
subarray and reconfigure the correlator.  (It shouldn't have, but the 
subarray software isn't quite right yet.)  Having the correlator reconfigure
every few minutes at random made a fairly difficult working environment for 
doing other things.  (Telephone service to antenna 14 doesn't seem to work.)
So I spent an hour getting set up for debugging subarraying problems.  (It
would have been quicker to walk over to antenna 14 and tell Jerry to lay off.)
I collected some data which may let me find and fix the problem.

Keeping an occasional eye on Idcaf while this was going on, I have the 
impression that there may be problems upstream - it seemed to occasionally
take several minutes before data started flowing after a reconfiguration.
But this needs to be looked at more carefully.

I found a bug that sometimes permits the Executor to send baseline lists to
the correlator controler even when told not to.  I suspect that that may 
lie at the root of last week's problems with reverting to Modcomp control.
This week that went more easily, although not without some bumps - I have
no idea which pieces of the incantations I muttered were really necessary.

In all, I think we are in good position to get done next week what I had 
hoped to accomplish yesterday.