Re: unable to see epics variables

[Pavel Maslov <pavel.masloff@gmail.com>]

Maybe iptables -F will help?

On Thu, Jan 3, 2013 at 10:31 PM, Mark Rivers <rivers@cars.uchicago.edu> wrote:

caget uses ports 5064 and 5065 by default.  It uses UDP to locate which IOC has the PV, and then TCP to read the value.

The fact that you can ping makes it sound like there is a firewall somewhere blocking ports 5064 or 5065.  Those both need to be open for both UDP and TCP.

 

Mark

 

 

From: tech-talk-bounces@aps.anl.gov [mailto:tech-talk-bounces@aps.anl.gov] On Behalf Of James F Ross
Sent: Thursday, December 27, 2012 12:51 PM
To: EPICS Tech Talk
Subject: unable to see epics variables

 

Hello everyone,
I am not sure if this is an epics problem or not, but if anyone has any insight I would appreciate the help.  The error is the following:

$> caget fgt:status
CAC: Unable to connect because "No route to host"
CA.Client.Exception...............................................
    Warning: "Virtual circuit disconnect"
    Context: "softioc3.starp.bnl.gov:52008"
    Source File: ../cac.cpp line 1214
    Current Time: Thu Dec 27 2012 13:44:42.760798000
..................................................................
cagChannel connect timed out: 'fgt:status' not found.

I have been unable to find anything blocking a route to the host, and I can ping the host from the machine I am trying the caget on. Historically this has been caused by firewall issues, but even with all security (iptables and selinux) disabled I still get this problem.  Is there anything epics related that can cause this issue?  Something that isn't running but should be in order to allow the command caget to connect? 

It would be helpful to know exactly how caget looks over a network.  Is there a particular port it uses?  Does it look via rsh or NFS or some other connection type that may not be working for some reason?

Three machines spontaneously started having this issue which points to it being a network problem and thus nothing tech talk can help me with, but the fact that I can ping each host from the three machines having problems indicates that there is in fact a network connection and may be epics related.

Thanks for the help,
James Ross

--
Best regards,


Pavel Maslov, MS
Controls Engineer at Pulsed power Lab
Efremov Institute for Electro-Physical Apparatus
St. Petersburg, Russia

Mobile: +7 (951) 672 22 19
Landline: +7 (812) 461 01 01