Getting to know each other

Defining containerisation and hands-on Docker / Docker-compose

• Remi Voirin (CERN)

Container security discussion How to address container-related security issues in controls systems?

Sirius Diagnostics Container Deployment for EPICS IOC

• Lucas Russo

Sirius beam diagnostics group is responsible for specifying, designing and developing EPICS IOCs for most of the diagnostics in the Booster, Storage Ring and Transport Lines for Sirius Synchrotron Light Source. In order to ease maintenance, robustness, repeatability and dependency isolation a set of guidelines and recipes were developed for standardizing the IOC deployment. It is based on two main components: containerization, which isolates the IOC in a well-known environment, and a remote boot strategy for our diagnostics servers, which ensures all hosts boot in the same base operating system image. In this presentation, the containerization guidelines and usage, as well as the remote boot strategy along with its constituent parts will be discussed.

Storage and deployment in CERN controls: status and plans + group work on application storage and deployment

• Remi Voirin (CERN)

Presentation of current release and deployment techniques of software in CERN controls, and changes that we want to introduce. Group work around storage of container images and how they reshape the way we deploy software.

Containers in the ESS integrated control system

• Karl Vestin

Containers have rapidly evolved into the de-facto standard for deploying software infrastructure for the integrated control system at ESS. This short presentation gives an overview of the current status of containerization, shares some positive and negative experiences and indicates some areas for future exploration.

Introduction to container orchestration and hands-on minikube + group work on container orchestration

• Remi Voirin (CERN)

General presentation of container orchestration and practical hands-on session with minikube, a small distribution of the Kubernetes control plane. Using the six thinking hats methodology, we will explore possible use of container orchestration within controls systems.

Current status of service management at SuperKEKB

• Shinya Sasaki

We manage tens of server computers to provide various services for accelerator control at SuperKEKB. Most of the services are managed with init system such as systemd or SysVinit. We also use procServ to manage services. ProcServ is a command to run an application as a daemon into the background with telnet access to stdin and stdout. The services are manually migrated when we have a system trouble or a server replacement, however, preparing for execution environment imposes a burden on us. Therefore, we require effective service management system. I will report the current service management and expectation for container technology.

Quiz

• Remi Voirin (CERN)

General Q&A and wrap-up