strike

Simply put: wireguard And I'm not sure, but I believe you have to rename it to wg0.conf And here is the rest of the documentation:

Yeah, that's correct. I like to keep as many containers as I can on the custom network with their own IP (and host access disabled) just for security reasons. But it adds some extra complexity, so it's not for everyone.

I've confirmed it works and sends notifications now. But a note for other users with the same macvlan setup as me. If your signal-cli container runs on a custom network with its own IP you either need to go to settings-Docker, disable docker and enable host access to custom networks (this will enable it for all containers). I don't want that, as I want as few containers as possible with access to the host. So if you just want to enable access just for the signal-cli container, you can add this script to the user script plugin and set it to run at array start. #!/bin/bash # 1. Create a new macvlan interface on the host named "host-shim" attached to br0 ip link add host-shim link br0 type macvlan mode bridge # 2. Assign an UNUSED local IP address outside you DHCP range to this new interface ip addr add 192.168.2.250/32 dev host-shim # 3. Bring the interface online ip link set host-shim up # 4. Tell the host to route traffic for your signal-cli IP specifically through this new shim ip route add 192.168.2.222/32 dev host-shim

Yeah, I know, thanks :)

Now it installs and connects as long as the signal-cli container is in bridge mode. It will probably work with a custom network if I either enable host access to custom networks or just manually add a route myself. Testing that later, but now I'm gonna test that it works in bridge first. Thanks for the support so far. I'll be back if I see some more issues.

Yeah, I did that.

I get this error when installing: plugin: bad file MD5: /boot/config/plugins/signal-notification/signal-notification.txz

Aha, I thought I could use any signal-cli image. Awesome, thanks :)

I'm running the latest of bbernhard/signal-cli-rest-api.

@rgreen83 I was able to reproduce the issue, and I've fixed it. And it was related to the empty folder issue. Whenever it encounters an empty folder, it skips it, but instead of moving on to the next item in that path, it skips to the next managed path, leaving the disk under the free space threshold. I'll push an update in the next few hours, including the fix and the largest first sorting option. I just have to test that there are no regressions first.

Yeah, I've been thinking of adding that for some time. Just haven't got around to it. Initially, this plugin was created to just keep drives from becoming full and keep it just above that line. The goal was never to move a large amount of data, but the plugin has later evolved, so it makes sense to add. While it was not designed for that purpose initially, it can be used to empty drives or rebalance the array and move the files to other drives. Always feel free to post feature requests or other ways the plugin can be improved on. It keeps me motivated to actually work on it. The reason why I haven't added it yet was that there was really no request for it, and there have been no more bug reports, so I've prioritized other things instead. But yeah, keep an eye out for the next update, and it will be added. :) That's interesting. I haven't really tested it on moving large amount of data, as I mentioned it was not initially designed for that. But are you sure that the folders that are skipped actually have files in them? A side effect of using the remove source files command the script uses is that it sometimes leaves empty folders behind. And in earlier versions, there was no cleanup done to mitigate that. So you might have folders that are already moved, but have empty folders in the source disk. If the script is unable to determine the size of the folder/files (like when it's empty) it assumes it's either locked/in use or empty and will skip it. I'll take a look at it, but please check the folders that are skipped if they actually have files in them. If not, delete them manually, and they will stop appearing in the log. I've seen the issue myself a few times, but it has always been empty folders left behind from earlier moves. But it also sounds like it could be a bug if you have confirmed that it leaves the disk under the free space threshold. Sounds like it's maybe related to the empty folder issue somehow, and it messes up the calculations. IIRC I think I saw the same thing the few times this has happened in my runs, but I use it on a threshold of 60 GB and only move the smallest folders first, so I didn't think much about it when it was a few GB off. On big moves, this will of course add up. But for now, until it's fixed, I think the workaround is to make sure to delete empty folders if they appear in the log, as this is probably the cause of the wrong calculations, somehow. I'm really happy you like it. This is the right place to ask for support. Or did you mean like a "buy me a beer" link? I'll think about it :)

Seems to be a database issue. Can you restore from backup?

Thanks for this, but I get a 404. I'm running the API on a custom br0 network, but I've tried running it on the normal bridge as well with the same result.

It also looks like you have a lot of files with conflicts. example: [Scan] \"Anthony Ryan/To Blackfyre Keep\": Library file with path \"/audiobooks/media/allbooks/Anthony Ryan/To Blackfyre Keep/To Blackfyre Keep.azw3\" not found, but found file with matching inode value \"27021597764458986\" at path \"/audiobooks/media/allbooks/Anthony Ryan/To Blackfyre Keep/To Blackfyre Keep (1).azw3 Lots of these. Have you recently copied duplicate files to your library and deleted the original? It finds the files with (1).azw3. And (1) is usually appended when there is a duplicate.

I'm thinking it's a metadata issue too, but not with your files. You're using Google Books as your metadata provider, and your log is full of errors because it can't find your books. So your books are probably all in your library, but when you go to the author, you can only find 3 books because it fails to match the rest. Switch to the audible.com provider in the settings. and run a quick match on your library again. I've tested with one of the books in your log with google and I couldn't find it either, but with Audible it was fine.

Post your daily log instead of your scan log. Which version are you running?

Or you have a filter on that's only showing those 3 books.

Most likely it's because of your folder structure/naming. Have you checked the docs? https://www.audiobookshelf.org/docs/#book-directory-structure

So, according to the log, only 2 books were not imported, and it clearly says why: {"timestamp":"2026-02-10T13:00:20.531Z","message":"Library item at path \"Sarah Noffke/Determine the Future\" has no audio files and no ebook file - ignoring","levelName":"WARN","level":3} {"timestamp":"2026-02-10T13:00:20.601Z","message":"Library item at path \"Robert Jordan/The Fires of Heaven - Kate Reading + Michael Kramer\" has no audio files and no ebook file - ignoring","levelName":"WARN","level":3} Check that there are actually audio files in there.

Run this instead: curl ifconfig.io

During renewal, validation requests (like HTTP-01) might hit servers in different countries, so you can't restrict validation to just one country. They use servers and CDN points globally for validation, not just one location. Use the geoblocking in swag instead, I've never had an issue with that.

Then I would check the firewall or anything that has to do with dns, if you use pihole for example. The problem isn't with the VPN server, but something blocks name resolution causing the container not to be able to connect.

When it stops working, if you restart the container does it work again on first try? Try changing your name servers in the template to 1.1.1.1,1.0.0.1 If that doesn't work you need to monitor your firewall logs to see if it's blocking.

Do you use pfsense with snort/suricata or any other form of IDS/IPS? I had this issue a few months ago and it was due to snort adding a new rule, blocking delugevpn.

Have you seen this one? https://forums.unraid.net/topic/194592-support-scanopy-automatically-discover-and-visually-document-network-topology-ex-netvisor/ It may solve some of your needs. I'm looking to try this one myself as it seems very useful.