1. Running on a system as a Python script [we will refer to this as a "local" install]
2. Running as a Docker container
GENERALLY SPEAKING, running as a Docker container is simpler, as you won't have to be concerned about installing Python, or support libraries, or any possible system conflicts generated by those actions.
For this reason, it's generally recommended that you install via Docker rather than directly on the host.
If you have some specific reason to avoid Docker, or you prefer running it as a Python script for some particular reason, then this general recommendation is not aimed at you. It's aimed at someone who doesn't have an existing compelling reason to choose one over the other.
Kometa communicates with all services [Plex, Radarr, Trakt, etc] via their network APIs, so Kometa does *not* have to be installed on the same machine as Plex. Kometa does not require [nor would it use] access to the file system behind your Plex libraries.
Perhaps your Plex server is remote and you want to run Kometa on a machine in your home. That's fine. The relative locations of Kometa and Plex have no effect on the installation [except perhaps the URL you would use in the config].
**this doesn't cover the Kometa setup specifics found in the guides above with regard to creating the config file and collection file, so you may want to go through the [Docker Walkthrough](docker.md) first on your computer to gain that understanding.**
These are high-level steps which assume the user has knowledge of python and pip, and the general ability to troubleshoot issues. For a detailed step-by-step walkthrough, refer to the [Local Walkthrough](local.md) guide.
- The `-it` flag allows you to interact with the script when needed (such as for Trakt or MyAnimeList authentication).
- The `-v <PATH_TO_CONFIG>:/config:rw` flag mounts the location you choose as a persistent volume to store your files.
* Change `<PATH_TO_CONFIG>` to a folder where your config.yml and other files are.
* The docker image defaults to running the configuration file named `config.yml` which resides in your persistent volume.
* If your directory has spaces (such as "My Documents"), place quotation marks around your directory pathing as shown here: `-v "<PATH_TO_CONFIG>:/config:rw"`
These docs are assuming you have a basic understanding of Docker concepts. One place to get familiar with Docker would be the [official tutorial](https://www.docker.com/101-tutorial/).
A `Dockerfile` is included within the GitHub repository for those who require it, although this is only suggested for those with knowledge of dockerfiles. The official Kometa build is available on the [Dockerhub Website](https://hub.docker.com/r/kometateam/kometa).
Kometa's behavior can be modified in a variety of ways using either runtime flags or environment variables. These flags and vars are detailed [here](../environmental.md).
This is optional, and is not necessary to run Kometa. Many if not most users will have no reason to do this and can use something more like the basic docker-compose just above.
This example docker-compose would create a container that runs immediately upon start (rather than waiting until 5AM), uses a particular config file, processes only overlays on only one library, and exits when done. Those four changes are made by the four `environment:` entries, which are discussed in detail after the example:
As with the one above, this is an example docker-compose which will have to be edited to suit your environment before use.