Experimental Physics and
Industrial Control System

D Peter Siddons <[email protected]> · Thu, 27 Jun 2013 11:44:48 -0400

Hi Jeff,
  Well, I was naughty and didn't mention that this is a Linux PC :( Sorry.

Looking at the same PV in a MEDM window I can see that the (single)

caput executes immediately, so there are no lingering puts around.

But the puzzle is that other installations don't show this behavior.

Has anyone else seen this effect?

Pete.

On 06/27/2013 11:03 AM, Hill, Jeff wrote:

Hi Pete,

I suspect that this delay might be related to CA's orderly shutdown of the winsock (windows socket) library or alternatively related to the CA's orderly shutdown of the TCP socket (the socket can be configured to block until the connection close handshake completes which can be important for example if there is a last CA put message lingering that hasn't been delivered to its destination). Such delays might also be dependent on the exact version of windows. At some point I should probably try to find out what version of windows and watch this in the debugger. FWIW, I don't personally recall seeing such delays (I am using windows 7), but I have been working mostly in Linux, virtualized Linux guest on windows 7, lately.

Jeff

-----Original Message-----
From: [email protected] [mailto:tech-talk-
[email protected]] On Behalf Of D Peter Siddons
Sent: Thursday, June 27, 2013 8:43 AM
To: EPICS
Subject: caget apparent timeout

Hi all,
    I have a problem which seems to crop up occasionally, depending on
the setup, but quite often. I have a PC running MEDM screens connected
to a remote IOC by a crossover ethernet cable. The MEDM apps all run
fine. Occasionally, I need to check or change a PV, for which I use the
caget and caput command-line utilities. The operation (put or get) is
executed immediately, but then the PC waits for several seconds before
the cursor is returned. Does anyone know what is causing this? It feels
like a network timeout, but all I have is a PC and an IOC with a
crossover cable between. Anyone know the fix?
Pete.

--
D. Peter Siddons
Detector Development Group Leader
Bldg. 535B
Photon Sciences Directorate,
Brookhaven National Laboratory
Upton, NY 11973

email: [email protected]
Phone: (631) 344-2738

--
D. Peter Siddons
Detector Development Group Leader
Bldg. 535B
Photon Sciences Directorate,
Brookhaven National Laboratory
Upton, NY 11973

email: [email protected]
Phone: (631) 344-2738

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System