[Article proposal] Automate podman generate systemd using Quadlet

There is the relatively young project ‘Quadlet’ which abstracts away the “systemd”-part of operating podman + systemd container setups and aims to reduce the config file setup to “what a container should be doing” while adding all the rest a systemd unit file requires based on the latest podman systemd generator spec automatically.

We could write an article like on the GNOME blog for FM: Quadlet, an easier way to run system containers – Alexander Larsson

Not sure however what “special sauce” we could add to make it distinct from the GNOME article except using Fedora Linux terminology and CLI. That could need a little more thinking - apart from that it’s nice to promote it on the FM nevertheless.

The project currently does not have an RPM package and has to be build from source.

Outline:

  1. Introduction - what is Quadlet, where does it come frome
  2. How to set it up
  3. How to use it
  4. What to expect in the future? (Integration in podman, RPM packaging?)
  5. Maybe: How to contribute - become RPM package maintainer

+1 from me. The post you are linking to is generally neutral to me. So adding some detials on how this looks in the Fedora Linux cli would be good I think. Your outline looks good too.

Can you think of some way to use the tool that the more typical Fedora Magazine reader might be interested in? I doubt that there are many Fedora Magazine readers who would be all that interested in Redis.

Hello @w4tsn , I’ve created Pagure issue (card) #46 to track your article.

Please follow all of the directions found at
https://docs.fedoraproject.org/en-US/fedora-magazine/writing-an-article/
except for the KanBan part (Pagure will replace that). We are using Pagure to track and comment on the article as it progresses and to answer questions from the authors, from time to time. It is a handy way of keeping track of the article progress for both the authors and the editors. Also a good place to ask for help about specifics of what you are writing if needed.

This is a change from the previous process since Taiga is being decommissioned.

The editors will move the card to “in progress” once you start writing the article. Please let us know by commenting on the issue (card) in Pagure.

After you have completed the article, and feel it is ready for review, please add a comment to the card so the editors know to move it to the Review column.

The editorial board meets weekly (Thursday’s at 1500 UTC) at #fedora-meeting channel on libera.chat (irc), and will review the article in the “review” status at that time and decide if it is ready for publishing or not. If publishing is decided, then an Editor and Image creator are assigned (often the same person) and the article gets scheduled. If the editors feel the article needs more work, it will be moved back to the in progress column, and the author will be informed of the reason(s), usually with helpful suggestions on what needs some more work.

I could take the Traefik reverse-proxy example from [Article Proposal] Migrate containers from Raspbian to Fedora IoT using [...].

Other possible examples are:

  • factorio - for the gamers on Fedora
  • telegraf - very easy to setup and maintain system metrics extractor
  • nginx in file-server configuration

More complex examples like matrix-synapse, nextcloud, mastodon or PiHole I’d like to save for their own articles. One of those probably for [Article proposal] Orchestrate multi-container setups using Podman + systemd

@glb do you have any specific example in mind that would be of great interest to the Fedora community?

The nginx file server idea sounds like it might be something that more readers might be able to find uses for. I think I’d go with that one.

I don’t have any better ideas. I just thought the Redis database seemed a little too obscure.