cscoppa

Just do something like this, I don't even have that L2 Discovery option turned on, and I can manage my devices just fine. (and Adopt)

I 100% agree with you. Years ago I was self-hosted because I started with the Access Points. When I got the Dream Machine I was disappointed in the speed compared to what I had before. I also like the MANY options I have for backups.

Just wanted to say thanks for creating this! My UDM (base) died 2 days ago (the classic no lights, full fan death) and needed to get something up quick. This worked out great and is running really fast on my hardware, much better than the UDM. I have the UXG-Lite in transit and will be adding that shortly. For now I'm up on my old Asus RT-AC86U for the router function, and managing my 2x NanoHD's and 1 U6 Pro with this container.

THANK YOU Squid and everyone else that contributed to fixing this. It had been driving me crazy for quite some time!!! Zigbee2MQTT was the first one that started showing not available awhile back, then 2 more popped up in my list.

PS, I color coded my previous screenshot. The red lines are the "real" location of the GPU and HDMI audio devices respectively. The blue are the virtual ones that are presented to the VM. "vfio" is what gets created when you bind the devices in the Unraid UI.

I'm not using a ROM file, I don't believe it's needed with my card. (too old to need one I believe) For 2 I let the UI create the XML. (in basic mode) ... then I switched to advanced and modified the XML based on some guides / videos I've seen. Not sure if this is required, as I believe it's mainly to get HDMI audio working. From what I remember, Window sees the GPU, but it sets it at a Code 43 or something, so I had to force the driver in to get it working. I downloaded the driver from Nvidia, ran the installer, and then when I knew it had extracted the files to a temp folder, I switched to Device Manager and installed the driver from there, then I just killed the Nvidia installer. HA=N means Nvidia in Blue Iris ... the "HA" is Hardware Acceleration. If it wasn't working it shows as a "-" instead.

It wasn't too hard, I just had to "force" the driver in by starting the installer, switching to Device Manager, and pointing it at the temp driver install folder to get the driver to go in. TOOLS --> SYSTEM DEVICES:

Same here, when I first built the box. I ended up flipping it around and using the iGPU for Plex and passing through my old GTX 730 to a Win11 VM for Blue Iris to use. I don't transcode in Plex very often (prefer direct play) so would be interested in getting iGPU passthrough to work.

Thank you so much for posting this. I wanted to reconfigure things so that I could turn OFF "Host access to custom networks" as that generates a new MAC address on every boot, and it really is a hacky implementation. I could NOT get local LAN redirects working until I saw your post, and had the idea to move the Unraid UI ports so that NPM runs on the standard 80/443 like your screenshot. Now it's working great so thanks!!!!

I believe I took that screenshot right after starting playback, so it is probably still building the buffer.

APC units are probably the safest bet, but I would imagine others might be supported. More info here: http://apcupsd.org/manual/manual.html

Pretty happy with the power usage on my 12700. (non-K on purpose) This is with a Windows 11 VM running with 8 cameras in Blue Iris going, about 20 Dockers, and all drives (4) spun up at the moment:

I can confirm the mac still changes on every boot with 6.11. (for shim-br0, which is what the router sees as the main)

Working here too on the i7-12700. NOTE: You have to be on the Plexpass version. ( plexinc/pms-docker:plexpass )

What I did is install Watchdog on another client, then copied both its runtime and config files to the Docker. I use this to pickup new recordings that my HDHomeRun records. Once the files are in place you toggle it on via the web UI.