Experimental Physics and
Industrial Control System

"Bertrand H.J. Biritz" <[email protected]> · Thu, 15 Jan 2009 13:56:45 -0800

Dear Tech-talk,

Happy New Year :)

Here is a curious problem: the IOC we have been using without a

problem in the past all of a sudden will no longer read back correct

values, but it can still write them.

A little background information to set the stage. The supply is a CAEN

SY1527LC mainframe with four A1534P/N boards. The IOC was written by

Ru Igarashi from the Canadian Light Source (http://www.aps.anl.gov/epics/modules/manufacturer.php#CAEN

) and we modified it a little for our purposes at BNL during the

summer of 2006. No further modifications were made to the source code

since then.

Three months ago we were using the IOC without any problems. Last week

when I launched the IOC I noticed I could set parameters and turn on

channels – verified this via the built-in telnet interface of the

mainframe. However, I was no longer able to read back the channel

parameters properly. There is no error message, it just reads 0 all

the time. The PV which read the momentary voltage output always reads

0 now, similarly for the channel status PV and so on.

I did noticed with one PV it would retrieve the correct value

eventually. The initial current limit is 100uA and the read back PV

says as much. Then I set it to 50uA but read back 100uA – even after

the scan period has long expired. Set it to 75uA and now I read back

50uA.

The trouble shooting I have done thus far is (all of which was

unsuccessful):

re-compiled the IOC
turned off the firewall

verified the CAEN provided C wrapper library to which the IOC calls is

the same as the backup version

I used the demo program CAEN provided with the C wrapper library to

verify I can read out the channel parameters successfully with it – so

I believe the mainframe is fine.

Right now I am at a loss as to what has changed. The only thing I can

think of is we updated the system software via yum to the latest

Scientific Linux 4.x.

However, the vlinac and another IOC are fine.

Any thoughts or suggestions would be greatly appreciated,
Bertrand

P.S. I hope this is a really trivial problem due to some simple

misconfiguration I am not thinking of :)

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System