I’m trying to set up a Lemmy instance using Docker on Ubuntu. This is my first time interacting with the open source community, so apologies if this is a stupid question. I’m not familiar with any common knowledge that might be obvious for people who frequently install software from github.

Anyway, in the official Lemmy installation guide, it says to download the following two files from the Lemmy github:

https://raw.githubusercontent.com/LemmyNet/lemmy/main/docker/docker-compose.yml https://raw.githubusercontent.com/LemmyNet/lemmy/main/docker/lemmy.hjson

However when you look at lemmy.hjson, it has a comment that says “This is a minimal lemmy config for the dev / main branch. Do not use for a release / stable version.”

I’m a little confused because I would expect that the setup guide would want me to use the latest release version, not the dev version. Why is this not the case? If I’m not supposed to use the release version, then who is the release version intended for?

Anyways I tried looking for the latest release version branch, which appears to be 0.17, and I got confused again when I saw folders for “prod” and “federation” in the Docker folder. What’s the difference?

Anyways can someone please help me understand the branch/folder structure? If I just want to run a regular Lemmy server with the latest stable code, which files should I use? Thanks

  • @BurningnnTreeOP
    link
    English
    11 year ago

    Thanks for the info. Sounds like Docker isn’t the way to go unless I want to spend a lot of time troubleshooting issues. Have you tried the Ansible method? Is it any easier or more reliable than the Docker method?

    • ubergeek77
      link
      fedilink
      English
      21 year ago

      Oh no, Docker is 100% the way to go for reliability. The Ansible script is a mess. It only works on Debian based systems (which I didn’t have previously), I couldn’t get email working, and I had a weird issue where I couldn’t sign up for my own instance.

      Granted, I didn’t even try to get email working on Docker because I don’t need it, but the Ansible build is much less maintainable for me. On Docker compose, I change 2 version numbers in the file to update and that’s it. For Ansible, I’m not sure how it would work, I guess I have to wait for the Ansible author to update the script first, so I would be reliant on them.