Then I can only recommend jdownloader to collect the links and download the files and filebot for organizing and consistent naming. The sharehosters cannot really be automated, because they have captchas everywhere and there exists no standardized index (like for torrents or Usenet). Instead, everything is forum based (like Serienfans or movie fans). And for just organizing manually downloaded files, I found the *arrs not really feasible.
The *arrs unfortunately don’t really work that well with oneclick hosters. I had a setup of jdownloader combined with filebot and a sharehoster premium account once in a while, and then using Plex to access the media. Since then I switched to *arrs in combination with Usenet and am glad I could leave the oneclick hosters behind. Especially when you are frequently paying for them, it’s absolutely worth the money (Usenet providers and indexers are mostly paid and I would recommend against free ones). If you are interested, this guide is an excellent entrypoint.
Looks like GitLabs CI Pipeline Needs/DAG visualization.
You could also search for the automation.turn_off service in your yaml files and see, if any match references your specific automations entity_id.
There should be entries in the history/logbook for the automation entity itself (automation.whatever). From there you should at least be able to determine, what changed the enabled state of your automation(s). If it is a service, you might have to search for specific calls of this service (e.g. in your other automations). It might also be an option to do a global text search for the entity IDs of your automations and/or any automation services. You could do this via the vscode addon if it is installed.
I myself have never encountered this error and got a few disabled automations, which haven’t changed their state over months and multiple updates/restarts.
Do you think the gaming part will be difficult because of the VM and passthrough, or was that simply a comment on the hardware choice? I don’t really care about anti cheat or anything and thought GPU passthrough and some assigned cores won’t be a problem. But I also don’t have any experience with this kind of setup.
Are you German? Then this guide would probably be the best way to start. If not, I cannot really recommend a good guide. But you’ll essentially need a Usenet provider (like eweka.nl) to access the Usenet and an indexer (like SceneNZBs) to find stuff. These two can be combined with e.g. SABnzbd as a downloader and the *arr software suite to automate the discovery and downloading of media.