Experimental Physics and
Industrial Control System

From: Lucas Russo <lerwys at gmail.com>
Date: Thursday, October 28, 2021 at 5:59 AM
To: "Paduan Donadio, Marcio" <marcio at slac.stanford.edu>
Cc: EPICS tech-talk <tech-talk at aps.anl.gov>
Subject: Re: IOC in Docker not responding to caget from another host

Hi Marcio,

Hey Marcio,

 

one think that comes to my mind: If you run the IOCs in containers with network mode host there might be multiple caRepeaters now listening on the same IP address and ports as each IOC might start its own caRepeater within its container.

 

If you start multiple IOC on the same host without containers there will be only one instance of the caRepeater process.

 

I'm not sure if this might cause your problem.

 

Best regards,
Florian

 

On 10/28/21 4:54 AM, Paduan Donadio, Marcio via Tech-talk wrote:

This is the scenario:

 

In HostA I have one container running an IOC and another container with caget/caput installed. Both containers have these env variables defined:

 

EPICS_CA_ADDR_LIST=127.255.255.255

EPICS_CA_REPEATER_PORT=5065

EPICS_CA_SERVER_PORT=5064

 

I can caget in one container and receive data from the IOC. When both are in the same host it works well. With tcpdump I can see the traffic between the two containers, using localhost to communicate.

 

The containers are running with network host mode.

 

Now I want to run caget in HostB that is in a different network. In HostB I've defined:

 

EPICS_CA_ADDR_LIST=134.79.219.255

EPICS_CA_REPEATER_PORT=5065
EPICS_CA_SERVER_PORT=5064

 

With tcpdump in HostA, I can see that 5 caget requests arrive from HostB to HostA, but the IOC doesn't seem to be aware of the packages. I understand that the IOC responds to broadcast requests coming from any of the interfaces by default. Is that right?

 

I've tried Ralph Lange / Rodrigo Bongers script (https://wiki-ext.aps.anl.gov/epics/index.php?title=How_to_Make_Channel_Access_Reach_Multiple_Soft_IOCs_on_a_Linux_Host) but nothing has changed.

Any clues? If you need more data, I'm happy to add them to this thread.

 

Thank you,

 

Márcio Paduan Donadio | Control Systems Engineer Advanced Control Systems Department SLAC National Accelerator Laboratory | Menlo Park, CA p: 650.926.5007 | w: slac.stanford.edu

-- 

Ruhr-Universität Bochum

AG der Experimentalphysik I

Dr. Florian Feldbauer

NB 2/131 / Fach 125

Universitätsstr. 150

D-44801 Bochum

Office: NB 2/134

Phone:  (+49)234 / 32-23563

Fax:    (+49)234 / 32-14170

https://paluma.ruhr-uni-bochum.de