Last week I got an opportunity to be part of Fedora Infrastructure Hackathon from 9th to 13th April in Fredericksburg, Virginia, USA. It was a very nice experience to work face to face with the folks I usually interact online. Additionally, as a new team member, it was good to get to know everyone in person.
It was a productive week for all of us – lot of things were discussed, planned and done. You can find these details on hackathon wiki page. Being there helped me a lot with improving my understanding around Atomic, our Fedora infrastructure setup and projects with their current state developed and maintained by Infra team. During the hackathon week, we discussed about topics such as rawhide gating via bodhi, Fedora infra team maintained apps migration to openshift, etc. More details can be read from Kevin’s and Pingou’s post.
We also discussed and planned where to put multi-arch content of the Atomic Two Week release on Fedora Website. Right now, if you go to Fedora website for Atomic downloads, there is only x86_64 related information and download links. After discussion, we decided on extending existing Atomic download page to have multi-arch content. Atomic is a fast release, and keeping all arches information at one place will help us maintain and update website easily with latest Two Week release content. Kudos to Ryan who volunteered to make UI changes in the website!
I also talked to smooge to get a better understanding of our Fedora infrastructure setup. For example – where to look for logs for a particular infra machine, querying packages installed in a builder. This is definitely going to help me with debugging and fixing some of the Atomic related issues.
There were further discussions with Patrick and Randy on having multi-arch images in Fedora registry which currently has x86_64 specific images. We discussed about things that need to be done to enable support for pushing multi-arch images. This will include fixing prep-docker-brew-branch script to support multi-arch.
Other than that I had lots of conversation and knowledge transfer with Dusty.
- Dusty shared with me the details of how we do Atomic Two Week release during Fedora GA release. This release is done for latest stable Fedora. For example: current Fedora Atomic Two Week release is based on F27. Other than that we also have nightly Atomic composes for pre-released Fedora (F28 at the moment) and rawhide. Things get tricky when a new Fedora GA release comes into picture. There are additional steps and scripts involved during this phase to make sure Atomic Two Week release with new Fedora GA happen successfully on time. We will soon have F28 GA release, it will be interesting to watch all steps happening and helping when needed.
- We also discussed about having a Fedora Atomic Host release dashboard. Before doing Fedora Atomic Two Week release, we usually go through various checks to ensure that everything looks good on all supported arches. This is a time consuming and manual process. Having a dashboard where we collect results status from various places for every Two Week and other pre-release nightly composes like autocloud test, atomic-host-tests, AMIs upload status and so on will be very useful. This is going to help us to always be ready for release with Atomic Two Week.
- We also looked into how vagrant images get updated currently after Atomic Two Week release happens. Right now we do it from the vagarantup web UI. We discussed and the plan is to do it through vagrant-cloud api.
I must say it was a very productive week for me. Being part of this hackathon made stuff happen quicker over IRC. Many thanks to Paul Frields who organized everything during this trip. Also thanks to Red Hat and Fedora Council for sponsoring this trip. It was very nice meeting you all and definitely want to meet you all again!
4 thoughts on “Things We did at Fedora Infrastructure Hackathon 2018”
[…] Original Post […]
[…] "Last week I got an opportunity to be part of Fedora Infrastructure Hackathon from 9th to 13th April in Fredericksburg, Virginia, USA." https://sinny.io/2018/04/18/things-we-did-at-fedora-infrastructure-hackathon-2018/ […]
[…] on jobs to get people enabled and working (generally refers to employment). I helped get Dusty and Sinny into a group which could log into numerous atomic staging systems to see what logs and builds were […]
[…] I did during this was working on tasks to get people enabled and working. I helped get Dusty and Sinny into a group which could log into various atomic staging systems to see what logs and builds were […]