Each service is a separate docker-compose.yml, but they are more-or-less the same as the example configs provided by each service. I did it this way as opposed to a single file to make it easier to add/remove services following this pattern.

I do have a higher quality version of the diagram, but had to downsize it a lot to get pictrs to accept it…

The factory must grow

The wiki is a great place to start. Also, most of the services have pretty good documentation.

The biggest tip would be to start with Docker. I had originally started running the services directly in the VM, but quickly ran into problems with state getting corrupted somewhere. After enough headaches I switched to Docker. I then had to spend a lot of time remapping all of the files to get it working again. Knowing where the state lives on your filesystem and that the service will always restart from a known point is great. It also makes upgrades or swapping components a breeze.

Everyone has to start somewhere. Just take it slow and do be afraid to make mistakes. Good luck and have fun! 😀

The systemd.timers are basically cronjobs for scripts I wrote to address a few of the pain points I’ve encountered with the setup. They’re either simple curl or wget and jq calls or use Python for more complex logic. The rest are services that are either a part of or adjacent to *arrs.

As for k8s, personally I feel that would add more complexity than it’s worth. I’m not looking for a second job. 😛

The problem I’ve found is that the services will query indexers and that not all of the trackers allow you to use multiple IPs. This is where I found it easier to make all outbound requests go through the VPN so I didn’t get in trouble. It’s also why I have the Firefox container set up inside the network with it exposed over the local network as a VNC session. So I can browse the sites while maintaining a single IP.

I do have qbittorrent set up with a kill switch on the VPN interface managed by Gluetun.

There’s some overlap with my torrrents.py and qbitmanage, but some of its other features sound nice. It also led me to Apprise which might be the notifications solution I’ve been looking for!

Some of the arr-scripts already handle syncing the settings. I had to turn them off because it kept overwriting mine, but Recyclarr might be more configurable.

Thanks!

I get what they’re saying and it may be ‘technically correct’, but the issue is more nuanced than that. In my experience, some trackers have strict requirements or restricted auth tokens (e.g. can’t browse & download from different IPs). Proxying may be the solution, but I’d have to look at how it decides what traffic gets routed where.

The server itself is running nothing but the hypervisor. I have a few VMs running on it that makes it easy provision isolated environments. Additionally, it’s made it easy to snapshot a VM before performing maintenance in case I need to roll back. The containers provide isolation from the environment itself in the event of a service gone awry.

Coming from cloud environments where everything is a VM, I’m not sure what issues you’re referring to. The performance penalty is almost non-existent while the benefits are plenty.

For a long time, that was the case. Then the greed nation attacked. Now they’ve reproduced the cable model on the web and more than half of which have terrible clients / infrastructure.

If I could pay for a single service that operated similar to this setup:

• Tell it what I’d like to watch while also displaying similar content for discovery.
• Tracking progress in every show (while not forgetting!).
• Not losing content I have been watching as it’s now in ‘another castle’.
• A single place to view all tracked shows rather than loading each service individually.

I probably would sign up for it as that’s what was so successful for Netflix until all of the studios thought they could do better. And now the consumer has to suffer the consequences.

Maybe if you’re new to all this and/or have no interest. But if you’ve been tinkering for more than a few years, it’s just a PC version of project car. It’s something you tinker with on the weekends, adding and refining as you go. I would never be able to negotiate multiple streaming services in a unified way to my satisfaction. So it’s not as if I really even have the option of paying for what I actually want from a service.

This is taking it above and beyond I’ll agree. I’m still in the old times where I’m manually finding my movie/show/etc and doing all the leg work by hand, only because I haven’t had time to learn all the modern stuff. But things like this are a great resource to get up to speed.

I honestly took a shot at that. I didn’t have everything but I had most of it. I also had every channel available on cable.

The streaming wars have honestly ruined it. If you’re just looking for something to watch you of course will be able to find something. But if you specifically want specific content you might as well flip a coin. Oh it’s on Netflix, no Netflix lost lost that license, Oh Max has that, Wait no Max went under, no wait they’re back but they don’t have it anymore. Oh that’s a Disney property Disney+ should have it, nope Disney pulled that offline for the time being.

Screw it I’ll make my own streaming service with hookers and blackjack.

You can simplify it way down to Kodi + RD and have your own streaming service. This looks more like a hobby though to get every little thing just right.