Lacy

Sorry for the delay, here are my diagnostics: library-diagnostics-20221009-1327.zip

Friendly user Davo1624 on the Unraid discord was able to help me troubleshoot, it turned out to be the 4th point in the SWAG `502` docs: https://docs.linuxserver.io/general/swag#502. I had run swag before I setup the custom docker network. - I had to point the filebrowser proxy-conf at port 80, since that is what filebrowser uses by default - Running `docker exec swag cat /config/nginx/resolver.conf` returned an ip on my subnet like `192.168.1.254`, which was incorrect. - I removed the `resolver.conf` file, `rm /mnt/user/appdata/swag/nginx/resolver.conf` - Restarted swag - Now, `docker exec swag cat /config/nginx/resolver.conf` returned a local ip like `127.0.0.11` and everything worked!

I have followed spaceinvaderone and ibracorp videos exactly and have swag up and running using duckdns and cloudflare. I can access my swag instance at swag.domain.com but when I try to access filebrowser.domain.com the browser spins for a minute before I see a cloudflare timeout error. Steps I’ve taken: forward ports 80 and 443 to 180 and 1443 (used by swag) setup dyndns with duckdns create CNAMEs in cloudflare for various containers Enable “preserve user networks” in docker settings and restarted docker service created a custom docker network and placed swag and desired containers on the network renamed containers to match the "proxy-conf" settings, as well as matched the ports ("filebrowser", "8080") tried both spaceinvaderone and Ibracorp methods to route dns to containers using docker socket: changed proxy-conf files from “.conf.sample” to “.conf”, and changed the upstream port to match the container’s webui port AND added the auto-proxy mod to swag and “swag” label with to each container with value “enable” restarted containers and swag I could really use some help troubleshooting.

I have the main GPU mapped to VFIO. I have tried with the second GPU mapped to VFIO and without, the result is the same. My cache drive is currently full and the diagnostic download is failing, I will post after I clear it out.

Just want to note, this feature has been introduced in the latest update, 6.11

I have two NVIDIA GTX 1080 GPUs, one in a regular slot, one in a 4x riser. I almost have the server completely setup and working, with the VM using the main GPU. When I plug in my second GPU (on the riser), the VM shows a glitchy, static screen and will not boot. I've tried having the second GPU bound to IOMMU and not bound, both show the same thing. How can I get my second GPU working with Unraid (not passed through to the VM), while the VM uses the main GPU?

This worked for me as well

I have a USB switch that I switch between 3 computers, including one Unraid machine. I want USB devices to work with my Windows VM _when_ they are connected, but not have to be connected. Currently the VM will not boot without them connected. If I boot with them then I am able to use the switch without issues. Is this possible, to have a USB device associated with a VM, but not have to be present in order to boot the machine?