Experimental Physics and
Industrial Control System

"Kenneth Evans, Jr." <[email protected]> · Fri, 4 Mar 2005 18:16:49 -0600

Ralph,

>> Hope this helps (well, I know it doesn't really, but ...;-)

Actually it does help.  You seem to be able to reproduce it.  (I didn't know
you were having a problem.)

I just started MEDM's on two computers looking at the same Linux Gateway and
did the sliders (on the same PV) with both hands while running camonitor
(with the third hand).  The PV is in an IOC.  The numbers in camonitor
stayed in range.

However, the PV did get set to DRVH or DRVL sometimes (after releasing one
or both sliders), even when I was careful to not hit the ends of the
sliders.  And at other times the value was not what I expected after
releasing the sliders and getting the final value change.  The Gateway only
has one instance of this PV and trying to set it at the same time from two
MEDMs while getting independent value change callbacks may indeed cause
problems, especially considering the round trip time. I need to find two
computers next to each other on the subnet with the IOCs to try this w/o the
Gateway.  Also, this may be a different problem than Rolf has.  I don't see
any sign of large numbers.

I repeat that I have no problem when only changing the PV from one MEDM at a
time.

And there are other reasons than the Gateway to set DRVH and DRVL correctly,
especially if it destroys the equipment to set it out of range.

The code that handles puts should not have changed from Gateway 1.3.
However, CAS has changed.

	-Ken

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System