The goal of this project is to get an overview of the state-of-the-art technology on training and deploying machine learning projects with kubernetes and apply that to a SUSE CaaSP cluster. With that in mind, we will train and deploy a model for summarizing github issues:

Update: Ruqola ('zypper in ruqola') is now in Tumbleweed and Leap 15.2! For Hack Week 19 (Feb 2020) main goal is to test, polish packaging, potentially do fixes and select a snapshot to submit to openSUSE Tumbleweed. Upstream made one release but it was not optimal, development pace has sped up a lot lately and the client has become more stable and usable.

I want to learn more about robotics. The goal is to have a small robot at the end of the week. Since I have no clue of what I can do with robotics / what is possible I leave it open what the

SUSE Manager uses an ivy repository on the R&D network to get the packaged jar to build against. To remove the manual maintenance of that repository and allow external contributors to hack the Java code of Uyuni / SUSE Manager we need to provide an Ivy plugin that gets the jars directly from OBS.

I have previously implemented a major portion of the C++ standard library for HelenOS [0][1][2] as part of my master thesis. In this project, I will be adding

I plan to work on libgphoto2 Lumix WiFi support. Status after Hackweek:

The Parrot seems to have many fans meanwhile, in particular since we're running Rocket.chat at SUSE. Some may find these emoticons distracting, some find it just cool and I would certainly love to see something like the Partyparrot but as a "SUSE parrot", probably someone manages to make the SUSE Geeko move in a similar fashion, as party_geeko. Finally I'd love to see this uploaded to the cultofthepartyparrot.com under "Party Guests". Because clearly the Geeko insists! :-) Bonus: do the same with TUX for total ssǝuızɐɹɔ

The idea is to look at some technologies: - C++17 parallel execution

Use mouse to select a word on the screen, popup a small icon, click icon, then show a window to display word online translation result.

motivation

Many people ask how to start with openQA, ask "can you run these two lines in bash in an openQA" test and some shun the effort to try out openQA because it is "too big".

We are running the geekos app at geekos.prv.suse.net since some time now in the SCC CaaSP cluster. The goal we have is to bring this to production in the CaaSP cluster maintained by the engineering infrastructure team.

Description

On aarch64, ppc64le and x86_64, there is a common os-autoinst backend called qemu, in which qemu is launched from an installed SLE.

LanguageTool is a style and grammar checker for 25+ languages https://languagetool.org Packaging it will require packaging the following maven packages and their dependencies:

Given that free time is always a scarce resource, I'd like to use the hack week to work on v2.0 for OTPClient Github project: https://github.com/paolostivanin/OTPClient/projects/4

During the previous Hackweek I did a workshop in a Montessori school. This time I would like the same but with focusing on Robotics. Who is this for?

I would like to convert a very cheap foam toy plane into a radio control plane Materials

In order to see the SPD (detailed memory information) data, the user currently has to manually load the needed kernel driver. Which driver to load depends on the memory type. Depending on the driver user, the devices may even have to be instantiated manually and this is a non-trivial multi-step task. Plus you need to be root to do it. I would like to attempt to automatize all this at least in the most common and simple cases like Intel x86 desktop. The idea would be to figure out the memory type and the I2C address of the SPD EEPROMs based on DMI data. If the DMI data is of good quality then it should be possible to automatically figure out which driver to use and to instantiate the devices at boot time.

Produce a ~3 min demoscene using modern javascript libraries combining graphics with original music.

The youngest architecture addition to the mainline Linux kernel was C-Sky (arch/csky/). I have a GX6605S board booting a downstream 4.9 kernel. It uses a proprietary GxLoader bootloader (similarities with U-Boot exist but no sources...) with uImage and gx6605s.dtb files in a FAT partition on USB stick.

I maintain the mailprocessing utility, a Turing complete maildir and IMAP capable filter with rules written in Python, both upstream and in openSUSE. It is currently suffering from some performance problems I'll devote my Hackweek (or part of it) to improving. Most notably the following areas will need some attention: * Ability to deal with large inboxes. This one is already done and I have been using it myself for a while but not released, yet. If you urgently need it feel free to grab it from the master branch. I'll give it another once-over before release, but it should be fine.

Few weeks ago whilst analyzing the possibility of using Terraform[1] via terraform-libvirt [2] plugin, I came across Terratest [3], which is an automated testing framework. My initial thoughts have been written here [4] but I'd still like to check out Terratest in more detail.

Learn Rust by following The Rust Programming Language book.

So far the style (color, shape, ...) of the devicegraph output of libstorage-ng is hard-coded. The user of the library should somehow be able to customize the style.

Following the work started in the last hackweek, Improve OBS service scripts, I will try to migrate current service script for workers to systemd unit, and at the same time, try to get rid of the sysv code.

Create a sound board that has the following requirements: * Works in Browser

The idea is to have a friendly way to access live cluster's nodes, compare outputs of simple commands and find out differences in tables data / structure. Current state: implemented and tested scripts which split table into ranges and then for each range check rows count and rows checksums.

When all of the SUSE Manager squads switched from SCRUM to a Kanban we lost estimations and therefore also the ability to do predictions. But there are other ways to get insights that are even more reliable, since they are based on anecdotal data. The lead and cycle times of issues are the two most important here. I'd like to extract those two for all of the issue from the spacewalk project and explore how they can be visualized in meaning- and helpful ways.

Create a new generic CUPS wrapper backend /usr/lib/cups/backend/monitor that runs in parallel with the actual backend to monitor it.

Following trademark and licensing issues with Grafana, explore the possibility of debranding Grafana and use that in SUSE Manager (and maybe others) Products are available from GitLab: https://gitlab.suse.de/susana

At SUSE Manager and Uyuni we use right now a set of bash scripts called sumaform-test-runner to run terraform and cucumber, send notifications and store cucumber results. However such scripts are currently hard to maintain and extend, and bash is clearly showing it's limitations when it comes to parsing cucumber results, working with JSON stuff, or using APIs. Besides it forces us to have credentials hardcoded at the main.tf files, which is a huge problem for making a public CI for Uyuni.

We've (mostly @cbruckmayer) put a significant amount of time into our performance measurement dashboards based on influx/grafana. I want to share this with the world because it's most awesome and easy to use. It's also nice if people don't go to some PaaS provider just for this...

Currently, the SUMA test-suite takes about 6 hours to complete, often fails in the first tests, which set up the environment for the rest of the tests, those tests are what we named "core features". To solve this problem we had planned to move from Jenkins Job to a Jenkins Pipeline, having stages to split the test suite into core features, initialize clients, secondary features. So,if one stage fails, the rest of the stages will not be executed.

There are attempts to solve some kernel deadlock with using lockless printk ringbuffer. The proposed implementation is pretty complex (6 stages, 6 write and 6 read barriers, two buffers, entries linked via list, ...) I have a idea how to make it easier with tracking the state and sequence number in one atomic variable. It might allow to remove the lists and many barriers. It is possible that it will just not work. Let's see.