This is a docker image:
https://hub.docker.com/r/docbill/fedora-plex/
This is a github repository:
https://github.com/docbill/docker-plex
Get the difference? You cat do "docker pull" on a git repo file, but you can on a docker image.
Overall, I would say
is overly complicated. You don't need to create a separate volume for every media type. You can always pass those as subfolders, mounts etc.
The important things is you want your directory structure to remain consistent. That way you can always remount it into another container any time.
The next thing is BOTH my docker image, and plexinc/pms-docker suffer from the same fundamental flaw... That is you don't want a docker image to be built just from copying a binary into the image. That is just bad design, and will eventually break when the URL changes or such. I'm actually very disappointed I had resort to using curl to pull the latest image... What you want, is for the docker file to contain the necessary instructions to BUILD the necessary binary. Or take one from an official Linux distribution that is built from source code.
Here is an example of a docker container that builds an application from source:
https://github.com/docbill/docker-drive
And here is one that installs a version of from an official Linux package:
https://github.com/docbill/fedora-eclipse
I would say the last is most preferred. As in means the application has been added to linux distribution, which is the most preferred state.
I am sadly disappointed the best PLEX could do for an "official" package is make a copy of the MANY user may plex Dockerfiles already existing. Absolutely nothing was done to resolve the fundamental flaw in all the existing plex images available on hub.docker.com.
And trust me this is NOT a plug for my docker image. There are hundreds of images for plex available. The most common is:
limetech/plex
With all those different images available, why in the world did I choice to build my own?
1. I was able to mirror the structure on my existing disk closely enough I did not need to rebuild the config and tag all the movies again.
2. I always base my images on Fedora. It saves me disk space and memory to use the same linux distribution for all my images, and I prefer to work with Fedora.
3. By build my own image, I can control when I upgrade the version of plex. Whenever I see the message there is a newer version available I kick off the a new build and pull the new image.
That said, I would say if you want to build your own image. My image is a fairly simple example on how to do that. If you want to just use an existing image, the linuxserver/plex seems to be the best go-to. Until there is actually an official plex image that does something better than this curl non-sense, there really isn't a huge advantage over using one image over another image. The all work pretty much the same.
