Linux load average and other silly metrics Talk by Franck Pachot. Link to the details in the Reference section. The talk starts with the saying the load is often misunderstood. He shows a classic example of io wait with high load, low cpu. Lots of processes in D state, this can lead to high load. You can lower the io wait percentage by running ‘yes’. Well, it simply means that the cpu is no longer only waiting on io.

Opening up communication silos with Matrix 2.0 and the EU Digital Markets Act Two parts, matrix 2.0 and the EU act. First few graphs shows that Matrix is really popular. The graph showing government deployments is interesting to see. Not all governments fund the project. “Public money for public code” translates to adding new features, since this is visible in the final product. 2023 was rough year. This forced the team to focus on certain projects, matrix 2.

Where Did All the Fun Go Talk by Bogomil Shopov. Link to the details in the Reference section. Fun opening with an AI assistant replacing the speaker. Well, taking over the presentation. Directing it towards a Jean Claude van Damme presentation. Part of the fun comes from our heart. One example, is chair rowing. Not a lot of people remember this in the room (me included). Singing songs about what we do.

Soft Reboot Talk by Luca Boccassi. This talk is about a new systemd feature. Soft reboot. Targeted towards performance, service interruption. Service downtime should be minimized. This should pair nicely with kernel ’live patching'. For the kernel ’nothing at all is happening’. However, for userspace it is a full shutdown/reboot. With a File Descriptor Store applications can buffer connections to keep them alive while the system goes for a soft-reboot.

Using chroots in a single Linux Container as an alternative to docker-compose Talk by Aiden McClelland (details linked in the Reference section below). The talk is centered around StartOS. An operating system designed to enable non-technical users to self-host open source software. The OS hosts this via containers. With the base philosophy ‘one container = one service/app’. This makes managing IP addresses easier, and managing resources easier. The OS uses LXC to manage the containers.