Install Flexget RSS Manager

FlexGet is a multipurpose automation tool for all of your media

Support for torrents, nzbs, podcasts, comics, TV, movies, RSS, HTML, CSV, and more.

Installing Flexget#

Installing Flexget can easily be done through the creation of a python virtual environment. A virtual environment is the perfect option on a shared server as you can easily pull in any and all of the required dependencies you need to run your application without affecting other users on the server.

So to begin, we will create a new python virtual environment:

cd ~
python3 -m venv flexget

Enter the venv and install required packages

cd flexget/bin
./pip install flexget
## Optional if using deluge
./pip install deluge-client

Configuring Flexget#

Now that Flexget is installed to the virtual environment, we need to configure Flexget to tell it what we want to process and how we want to process it.

Add your config.yml to ~/.config/flexget/config.yml

mkdir -p ~/.config/flexget
nano ~/.config/flexget/config.yml

The config will be up to you. For complex configurations, you can reference the plugin guide on the Flexget website. Here is a sample config to get you started:

tasks:
  mycustomtask:
    rss: http://myrssfeed.com/feed.xml
    accept_all: yes
    deluge:
      port: 11111 (deluge daemon port)
      host: 127.0.0.1

note

There are also plugins for rTorrent, sabnzbd, nzbget, Sonarr, Radarr, etc -- I would highly recommend checking out the plugin guide for more info! The above is just the very basics of what can be accomplished via Flexget.

The above task will scrape the feed.xml at myrssfeed.com and send every new match it finds to Deluge. Filtering can be more complex than simply accept_all; however, for any complex tasks you should consult the documentation for specifics on how best to do them. accept_all works best in the case of pre-filtered feeds, i.e. those that you have setup with custom rules on your tracker in question.

Setting up systemd services#

With the configuration has been set, it's time to set a systemd --user timer and the coupled service file to go along with it. The timer will make sure that Flexget keeps running and will also ensure that it gets started in the event that the server reboots.

If the directory ~/.config/systemd/user doesn't already exist, it's time to create that now:

mkdir -p ~/.config/systemd/user

Now we must create our timer file:

nano ~/.config/systemd/user/flexget.timer

Inside that file, paste the following:

[Unit]
Description=Run flexget every 5 minutes

[Timer]
OnCalendar=*:0/5

[Install]
WantedBy=timers.target

You are welcome to change OnCalendar to something sorter (e.g. OnCalendar=minutely for once a minute); however be aware that not all trackers like being hammered for RSS feeds every minute and may result in an IP ban. Please consult your tracker documentation before setting the limit this low. More advanced OnCalendar notation can be found here in the systemd documentation.

Save and exit this file. Now we must write the associated service file:

nano ~/.config/systemd/user/flexget.service

Inside this file, place the following

[Unit]
Description=Run flexget

[Service]
ExecStart=%h/flexget/bin/flexget execute --no-cache

Make sure the user daemon has been reloaded:

systemctl --user daemon-reload

Now start and enable your timer (no need to enable the service!)

systemctl enable --now --user flexget.timer

If you run systemctl --user status flexget.timer you should see some information about the last time the timer was execute and the countdown until the next execution.

If you run systemctl --user status flexget.service you should see any potential debug information you might need about any issues.

You can also run ~/flexget/bin/flexget execute directly from the command line to trigger the same command have its outputs dumped to your current terminal session for debugging.

Finally, don't forget to checkout the Flexget Plugins for an indepth overview on how to configure all the various options available to you.