Experimental Physics and
Industrial Control System

Florian Feldbauer via Tech-talk <tech-talk at aps.anl.gov> · Mon, 15 Jun 2020 09:32:50 +0200

Hey Xiao,

we are using Docker containers for our IOCs as well. We start the

container with the "-dit" options.

The additional "-it" creates an interactive TTY which keeps the iocsh

alive within the container.

Also with this option you can "enter" the container and work with the

iocsh by using the command:

`docker attach <container id>`
Leave the iocsh with `CTRL+p CTRL+q`

HTH Florian

Am 15.06.2020 um 09:23 schrieb xiao zhang via Tech-talk:

Hi all,

The only way of starting an IOC I have learned is by running the

`st.cmd` script, which by default opens an iocsh shell on success. I

am seeing 2 problems when running IOCs in containers:

1. The iocsh shell seems redundant as nobody will actually use it when

run in containers. A way to start IOC which does not open up iocsh is

a more light-weight solution.

2. When running an IOC container with `-d` option only, the container

stops immediately. A dirty workaround is using `-it` option to attach

a pseudo-tty to the container. The caveat is that when the IOC is

broken the container keeps running. Although a liveliness probe can be

used to check the health condition of the IOC, a container exiting

when the IOC is broken is certainly a more light-weight and reliable

solution.

TL;DR, is there a proper way to start a persistent iocsh-less IOC

instance?

--
Dr. Florian Feldbauer

Ruhr-Universität Bochum
Experimentalphysik I AG
Universitätsstr. 150
Fach-Nr. 125
D-44801 Bochum

Office: NB 2/134
Phone:  (+49)234 / 32-23563
Fax:    (+49)234 / 32-14170
https://paluma.ruhr-uni-bochum.de

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System