So... An update happened, then the download clients and indexers disappeared

[4C3T]

Describe the bug
I was unable to contact you, @nls44 since they found this off topic, so I'm trying it here instead.
It did an automated update, then it suddenly couldn't find the download clients nor indexers.

Platform Information (please complete the following information):

• OS: Windows Server 19
• Radarr Branch: At first, unchanged Master. Then tried nightly no difference
  Log file here:
  radarr.txt

Trace Logs
radarr.debug.txt

[nls44]

Sorry, thought the auto update was disabled in the settings. Can you please disable that and reinstall my version? It should pick up your settings. I'll have to do some tests on Windows too since I use Linux myself.

I've added symlink support for Windows too, there's rar2fs binaries for Windows and probably something to pool drives too, so everything should in theory work.

[4C3T]

Ok. Figured as much.

I don't use RAR-archives for my movies, and I have no take on how these symlinks (Symbolic Links I get, but not what it's for and why etc) work.
These UNC pathways are weird. They're suppose to work, but they just don't anymore.
As I wrote in the other "forum", I managed to get it to work maybe 1-1½ year ago, and ran it for a short time until I moved the Radarr (etc) to another puter. It just wasn't optimal.

I'm a bit reluctant to swap the existing version on my server for yours as I'm allowing for updates since there are various fixes daily (nightly updates).
I don't know if you've spoken to these guys who managing Radarr about your project, but it would be nice if you could collaborate, or whatever would be necessary for these updates to work, and that'd be nice. Plus, you'd probably be relieved of some trouble shooting as well.
Just a though.
It's quite extensive this project. Haven't bothered to dive into it any, and besides I don't have the skills either. I do however do the odd tweak here & there from time to time, but I'm no programmer.
I do understand it takes a crap-load of time and effort to create something like this, and my hat off to you for even attempting anything like this.

Second question, again, sort of:
How do I go about to make your version of Radarr to post new arrivals in a relase-bot?
Yesterday I found myself logged into the AirdDC++ web-GUI, checking out add-on scripts, but I felt a bit lost after a while, since I don't know how these works (and these kind of software are for the most part poorly documented and any info that is available, is under the assumption that you already know everything else, so not pedagogically explained - One can tell it's written by a programmer!), so I'm quite lost here.

I was hoping there would be something that could pass on a few parameters like <!cmd>
Guess this gets kind of murky, as the categories per say doesn't quite exists, they'd have to be substituted for Quality Profile.
Here I'm thinking there would either have to be a script within your Radarr that does this, of settle for renaming the profiles to correspond with the existing ones in the hub-script. Pain in the butt to say the least.

I wish there were an easier way to do this so that other hub-users could add stuff with ease and have this sent to the release script as easy as "1, 2, 3".

[nls44]

I think in your case it's best to wait for the plugin system of Radarr to be implemented, then I can just add a plugin for AirDCPP as an indexer + download client and you would be able to use the regular Radarr and get the nightly updates. It's better since you don't require symlink support.

If you want to post new arrivals, best would be to create a small custom script that you would call in Radarr (under Connect -> Custom Script) that uses the airdcpp-web api.

[4C3T]

I think in your case it's best to wait for the plugin system of Radarr to be implemented, then I can just add a plugin for AirDCPP as an indexer + download client and you would be able to use the regular Radarr and get the nightly updates. It's better since you don't require symlink support.

If you want to post new arrivals, best would be to create a small custom script that you would call in Radarr (under Connect -> Custom Script) that uses the airdcpp-web api.

Ok. That might be easier than having to forgo the updates. Have a bead on when that's going to happen?

That script you speak of - Any ideas how I go about that?

[nls44]

One of the developers of Radarr has told me it will be added pretty soon for Lidarr; it should be included in Radarr soon afterwards I think.

For the script you can check out the API docs here: https://airdcpp.docs.apiary.io/ . Should be pretty easy with cURL to create a small script. Couple different calls you have to make (get hub overview, get session ID for the hub you want to send the message to and finally send a message to the hub).

[maksis]

Couple different calls you have to make (get hub overview, get session ID for the hub you want to send the message to and finally send a message to the hub).

It's actually possible with a single call: https://airdcpp.docs.apiary.io/#reference/hubs/hubs/send-chat-message

[4C3T]

Ok. That got a bit complicated...
So, where do I start?:

Got Radarr running on a Windows 7
Have AirDC running on Windows Server 2019 (Not as a server), with network shares to the above
Running Luadch hub on a Linux Mint x64 box (This have curl capabilities, although I have no clue about what it really does or is)

That's the outline of my rigs.

So, I begin with what, making a script - Where? Settings/Connect->Add/Custom Script in Radarr?
@maksis That was a bit fuzzy, but I take it I use the code on the right hand side, like this:
{
"text": "Hello there",
"third_person": false,
"hub_urls": [
"adcs://myhub.com:6423",
"mynmdchub.com"
]
}

But, I take it that's only the sending of the actual text. I still have to login and probably a few other things too, right?

The Radarr custom script allows me to open (not tested yet since I don't have a script) for validation/adding, which I presume this above comes into play, in whatever form it would take?

[maksis]

So, where do I start?

I'd recommend starting from the beginning of the API documentation where the introduction is. If you aren't able to figure out how to make the API call, you could also join the dev hub at adcs://web-dev.airdcpp.net:1511 (or chat with someone who's familiar with using APIs). You can also start with the example curl command in the documentation triggering a refresh to see that your script is working before moving on to the chat message sending.

[4C3T]

Hi again. Just thought I'd ask how it's going with that plugin for Radarr, any updates?
I haven't looked for it anything. I didn't use it any since that botched update, and just been running with the regular Radarr.

[kicker83]

Hi all,

I've been waiting a lot for this feature in Radarr, is there any update about this plugin?

Thanks!

[4C3T]

It would seem that something's happened to nls44. He's not been active since June, and I sent a msg to the staff at GitHub, but they're unable to send any kind of notice of concern to him. It would be nice if more could urge them to send some kind of mail or something else to him to see if he's alive or something.
It would be a shame if this project went to the grave as it would be most useful...

Edit, some time later he did reply, it just took a while

[Nischi85]

I really looked forward to this project. Is there any alternative?

[nls44]

I'm actually still using both this and the version I created for Sonarr. Both still running the latest release builds I published on Github. Just make sure to disable automatic updates and everything should be fine if you set it up according to the readme.

Both Radarr and Sonarr (the official branches) have received quite a few updates in the meantime but I haven't felt the need yet to update since everything (Plex updates, email notifications, fetching metadata, ...) is still working fine.

If you're using a raspberry pi, you can use this Docker image for Radarr btw: nls44/radarr-custom.

[4C3T]

A couple of months ago I gave it another try and let it wreak havoc on my other machine, but I wasn't getting the downloads that I hoped as I want non-RAR, but that's pretty much what it went for. I'm sure there's a way around it, but I didn't find any setting that made it stay away from them.

I'm a member of a hub that I don't share such archives (admin), and the client didn't omit that hub so I had to monitor it all the time and cancel tons of downloads.

I ended up uninstalling the client since it was unusable for my purposes.

There are a few things that differ from your version and the most current, and I don't remember what I was thinking of last I used it. One was, if I remember correctly, that I was unable to import the database from my newer version, which was a bummer.
I had to run import again, and that sucks when you've got some 15'000 movie folders to churn through (not to mention the other +10k movies in the Wanted-list, as well as other lists of collections and actors).

Which begs the question, since you mention Raspi - Without spending "big bucks" for a 8 to 16 GB version - How does a budget-Raspberry (Zero/Pi 1-4, 512MB to 1 GB RAM?) fare with a large db and imports from download clients?
My Dell PowerEdge 1950 mk II with 32 GB of RAM, Windows 10 Enterprise had an hell of a tough time processing large download queues (>30) and at times went on it's knees, if it were to have any that is. Did have qBittorrent on the same machine (which has gotten way slower lately and consumes resources like crazy!) but apart from the app being busy it ran pretty well by itself. Radarr was the culprit.

I have no intention of even trying to use any Raspberry for Radarr due to the size and all. Also, I'm running a dual multi-core CPU computer (Dell PowerEdge had 2 x 4 Cores, this has 2 x 16 so there's a lot of difference!).

I was just curious in general.

[nls44]

It should work fine for non-RAR too, but indeed no way to select between rar or non-RAR. Maybe in the next version I can add a way to exclude/include hubs from the airdcpp searches. Import most likely doesn't work because of the change in table structure between the different versions.

I'm not sure about the lower-end versions tbh, I'm running it on a 8 GB raspberry pi 4 myself but I'm pretty sure it ran fine on a 4 GB version too. I don't have a huge database though (I don't keep my current collection in there, only the things I'm interested in) but I also don't have to extract anything so there's little load usually.