uek2wooF

Holy crap guys after messing with this for 2 days and posting here, I just found a rom I dumped a year ago. Tried with that one and it worked!!!! That one is 129k /mnt/user/isos/vbios/dump.rom: BIOS (ia32) ROM Ext. IBM comp. Video (123*512)

Oh I also dumped my video bios but it is only 59k, seems small. Tried passing that with every combination (iirc). /mnt/user/isos/vbios/gtx780vbios.rom: BIOS (ia32) ROM Ext. IBM comp. Video (117*512)

Hi I had passthrough working a year ago in windows (and linux and osx) so I know it is possible with my GTX 780. This is an amd ryzen 9 3900x with no onboard video, just the one GTX 780 card. I upgraded to 6.9.2 and tried again to do passthrough. I used the new way (tools -> system devices) to check the boxes for the video and sound: IOMMU group 33: [10de:1004] 12:00.0 VGA compatible controller: NVIDIA Corporation GK110 [GeForce GTX 780] (rev a1) [10de:0e1a] 12:00.1 Audio device: NVIDIA Corporation GK110 High Definition Audio Controller (rev a1) Those are the only 2 things in that iommu group. I have tried legacy with seabios, and I tried uefi with ovmf. I tried both machine types with each. I tried the old (echo 0 > /sys/class/vtconsole/vtcon0/bind) 2>/dev/null (echo 0 > /sys/class/vtconsole/vtcon1/bind) 2>/dev/null (echo efi-framebuffer.0 > /sys/bus/platform/drivers/efi-framebuffer/unbind) 2>/dev/null I tried that with and without the system devices boxes checked. I tried modifying the vm XML and putting both the video and sound on the same virtual pci slot with multifunction. I tried HDMI, I tried DisplayPort. I can't remember what else I tried but I feel like I have tried everything. No matter what I still get Code 43 in my windows 10 vm for the video driver and a black screen on the monitor. It says Code 43 before I install the latest nvidia drivers and after. Goal is a steam game stream server.

Just to add, I get code 43 in the win 10 vm, and just a black screen. Fresh install of win 10. I am using the new way of passthrough, tools -> system devices, check the boxes for the gpu and sound. I also pass though a usb card and that works fine. ryzen 9 3900x. tried both uefi and legacy unraid boot. I also tried the bios but it is only 59K in size, seems small. also set multifunction so both gpu and sound appear on the same pci slot in kvm.

yeah those lines don't work anymore as a userscript or just pasting them into an ssh terminal as root since upgrading to 6.9.2 for me either (used to work on 6.8.2). echo 0 > /sys/class/vtconsole/vtcon0/bind echo 0 > /sys/class/vtconsole/vtcon1/bind echo efi-framebuffer.0 > /sys/bus/platform/drivers/efi-framebuffer/unbind root@Tower:~# echo 0 > /sys/class/vtconsole/vtcon0/bind root@Tower:~# cat /sys/class/vtconsole/vtcon0/bind 1 root@Tower:~# echo 0 > /sys/class/vtconsole/vtcon1/bind -bash: /sys/class/vtconsole/vtcon1/bind: No such file or directory root@Tower:~# echo efi-framebuffer.0 > /sys/bus/platform/drivers/efi-framebuffer/unbind -bash: echo: write error: No such device root@Tower:~# ls -l /sys/bus/platform/drivers/efi-framebuffer/ total 0 --w------- 1 root root 4096 Apr 18 21:40 bind --w------- 1 root root 4096 Apr 18 21:41 uevent --w------- 1 root root 4096 Apr 18 21:39 unbind

Oh and to the OP, what I do is just run fedora in a kvm and then install all the qemu-sytem-* emulations I want in there. The nested kvm performance seems ok, I am not doing any heavy lifting in mine (ppc and s390x).

+1 for qemu-system-ppc, -s390x, -sparc, and well anything else I might need someday.

Are you talking about the terminal window you get when you click under the "uptime" in the top right? Did you try increasing the size with openapple-+ (control-+ on other os). Or just ssh in using whatever your favorite terminal is.

Oops never mind. I forgot I made a qrcode generator docker container. I pasted the config in there, modified it, and generated a new qrcode for my phone to use.

Ah ok. Well it is easy to do on a laptop but sort of a pain on a phone. It would be nice to customize it from the unraid gui. Thanks.

I want something between "remote access to lan" and "remote tunneled access". Basically I just want to add another network route for the client through wg. I believe this should go in "peer allowed IPs" but unraid is putting that route on itself to the tunnel. I think this is a bug. If I download the config to the client and then edit it and add the net to "peer allowed IPs" everything works as expected.

Is there something in nerd tools to burn a dvd? Or maybe a docker to do it? I am not finding anything.

Some of these guys are just so helpful. If you just want to host static content locally, copy one of the example confs and put this in your server {} section and set the server_name appropriately. Make sure that path is available by editing the container, go to the bottom, click + Add another Path, Port, Variable, Label or Device and map a path there. I think what I did was made a share called wwwroot which makes a dir /mnr/user/wwwroot and then I mapped that to /wwwroot in the container. Then you could host each domain below that like this. location / { root /wwwroot/marv1; } It sounds like maybe you did this other part already, but if not: make sure to support extra domains by editing the container, going to advanced view and adding them to Extra Parameters like -e EXTRA_DOMAINS=marvsamurai.com,samuraimarv.com

Ever get this working? I need them too, (no passthrough needed).

Seems like you can only have one TLD and then multiple subdomains. Is there a way to get all my domains in this? Like: www.example.tld plex.example.tld www.example2.tld

Is there any way to get the VM without restarting all my other VMs (and restarting the whole array)?

Why are you using s6-svc to reload nginx after a cert renew instead of nginx -s reload anyway? If adding a conf file is heavy customization then how else do you suggest using it?

trying to debug your broken container isnt what broke it

Restarting the container makes everything run fine again, for a week or so. The docker run command isn't really exposed for community apps in unraid that I can tell. I think it is somehow dynamically created. I don't really tinker with unraid much lately so I can't think of anything that has changed. I did add another nginx conf file for another host and this problem started a couple of weeks after that. But if there was a problem with the conf why would it work fine for a couple of weeks first? It looks like 3 things run out of cron, logrotate, libmaxmindb, and /app/le-renew.sh. That last one is for renewing certs and it restarts nginx. I bet that one is the problem. I guess I will disable it and just set a calendar reminder to manually renew the certs.

After days or weeks of running fine my letsencrypt docker logs start filling up with this and all my nginx stuff goes down: nginx: [emerg] bind() to 0.0.0.0:80 failed (98: Address in use) nginx: [emerg] bind() to [::]:80 failed (98: Address in use) nginx: [emerg] bind() to 0.0.0.0:443 failed (98: Address in use) nginx: [emerg] bind() to [::]:443 failed (98: Address in use I tried nginx -s reload and it says it can't find the pidfile. I tried running logrotate again (with the command from the crontabs) and that doesn't fix it. Any ideas what is causing this?

So the question now is how to do updates to the app. I did a new image build with docker build, did the tag and push so it is in the repo, but I don't know what to do so it can be upgraded with community applications. Any ideas?

I have been trying to figure out how to add my own containers to Unraid and I couldn't find all the docs in one place. These instructions include creating a docker image, using your own docker registry instead of dockerhub, and incorporating your image into the private apps section of Community Apps where it can be installed and maintained like any other docker container. I hope this helps someone. The example used here is a QR code generator I made. It takes a string and when you submit the form it just shows the qrcode image in the browser. I use the letsencrypt docker to proxy to it. Set up your docker image. I just use a Dockerfile, I think docker-compose is the more modern way. Here is the Dockerfile: FROM node:12.9.1 WORKDIR /usr/src/app COPY package*.json ./ RUN npm install COPY server.js . EXPOSE 3016 CMD [ "node", "server.js" ] Here is package.json: { "name": "qrcode", "version": "1.0.0", "description": "", "main": "server.js", "scripts": { "test": "echo \"Error: no test specified\" && exit 1", "start": "node server.js" }, "author": "", "license": "ISC", "dependencies": { "express": "^4.17.1", "qrcode": "^1.4.4" } } Here is server.js: const express = require('express') const bodyParser = require('body-parser') const app = express() app.use(bodyParser.urlencoded({ extended: false })) const QRCode = require('qrcode') app.post('/qr/qrpost', (req, res) => { QRCode.toDataURL(req.body.qrstring, (err, image) => { const arr = image.split(/,/); const buf = Buffer.from(arr[1], 'base64'); res.set({ 'Content-Type': 'image/png'}); res.send(buf); }); }); app.listen(3016); Create a docker image for this: docker build -t yourname/qrcode . If you wanted to run it and test it you would do: docker run --net=yourprivatenet -h qrcode --name qrcode -d yourname/qrcode At this point you could just be done. The container will work fine, and you can see it in the Docker tab in Unraid. You just can't do much with it from there. I wanted to have my own containers controlled through the Unraid interface just like containers installed from Community Apps. If you continue make sure you get the container id from docker ps and then: docker stop $id && docker rm $id I wanted to maintain my own docker registry instead of using dockerhub. If your image is on dockerhub then you can skip this part: Save this xml into a file called registry.xml. (The icon is bogus, sorry. I am too lazy to put the right one.) <?xml version="1.0" encoding="utf-8"?> <Container> <Beta>False</Beta> <Category>HomeAutomation:</Category> <Date>2020-04-01</Date> <Changes> [b]1.APR.2020:[/b]Added[br] </Changes> <Support>https://docs.docker.com/registry/</Support> <Name>Docker Registry Docker</Name> <Description> The Registry is a stateless, highly scalable server side application that stores and lets you distribute Docker images.[br][br] This is the from the Docker team - I just created the template.[br][br] </Description> <Registry>https://docs.docker.com/registry/</Registry> <GitHub>https://github.com/</GitHub> <Repository>registry:2</Repository> <BindTime>true</BindTime> <Privileged>false</Privileged> <Networking> <Mode>bridge</Mode> <Publish> <Port> <HostPort>5000</HostPort> <ContainerPort>5000</ContainerPort> <Protocol>tcp</Protocol> </Port> </Publish> </Networking> <Data> <Volume> <HostDir>/mnt/cache/appdata/DockerRegistry</HostDir> <ContainerDir>/var/lib/registry</ContainerDir> <Mode>rw</Mode> </Volume> </Data> <Banner>http://i.imgur.com/zXpacAF.png</Banner> <Icon>http://i.imgur.com/zXpacAF.png</Icon> </Container> Create a directory on the usb stick that community apps will use (I'm actually not sure if this persists through reboots and Unraid upgrades): mkdir -p /boot/config/plugins/community.applications/private/myrepo Then copy that xml there: cp registry.xml /boot/config/plugins/community.applications/private/myrepo Now when you go to the Apps tab in Unraid you will see Private Apps on the left side. Install the registry you set up there with defaults. You can change the port from 5000 if you are already using it for something else. Just remember to change 5000 in later instructions to that port. Tag your qrcode (or whatever) image and push it to your registry: docker image tag yourname/qrcode localhost:5000/qrcode docker push localhost:5000/qrcode (I think it is probably safe to docker rmi $imageid to remove the image you created earlier since it is now in your own registry. I didn't test that yet.) In the rest of the instructions that is how you refer to the image. So if you had it uploaded to dockerhub it would be yourname/qrcode but if you have it in your own registry it would be localhost:5000/qrcode. Now you create an xml file for your app, qrcode.xml or whatever. Edit it for your app: (Again, the icon is bogus. Use a proper icon.) <?xml version="1.0" encoding="utf-8"?> <Container> <Beta>False</Beta> <Category>HomeAutomation:</Category> <Date>2020-04-01</Date> <Changes> [b]1.APR.2020:[/b]Added[br] </Changes> <Support>http://forums.unraid.net</Support> <Name>QRcode</Name> <Description> QRcode will take a string and return the QRcode image[br][br] QRcode will run on port 3016.[br][br] </Description> <Registry>http://unraid:5000/qrcode/</Registry> <GitHub>https://github.com/</GitHub> <Repository>localhost:5000/qrcode</Repository> <BindTime>true</BindTime> <Privileged>false</Privileged> <Networking> <Mode>bridge</Mode> <Publish> <Port> <HostPort>3016</HostPort> <ContainerPort>3016</ContainerPort> <Protocol>tcp</Protocol> </Port> </Publish> </Networking> <WebUI>http://[IP]:[PORT:3016]</WebUI> <Banner>http://i.imgur.com/zXpacAF.png</Banner> <Icon>http://i.imgur.com/zXpacAF.png</Icon> </Container> Copy that xml to where Community Apps can find it: cp qrcode.xml /boot/config/plugins/community.applications/private/myrepo Reload that tab. Install your app (set the network to your private net). Done. Whew. So much hassle, haha. Good luck! And I'm already back. I forgot, if you actually want to try out the qrcode app you will need this qrcode.html: <html><head></head><body> Enter string to qrencode:<br> <form action='/qr/qrpost' method='POST'> <input type='text' name='qrstring' id='qrstring'/> </form> </body></html> Copy that to your letsencrypt nginx document root: cp qrcode.html /mnt/user/appdata/letsencrypt/www And the qrcode.subfolder.conf location ^~ /qr/qrpost { include /config/nginx/proxy.conf; resolver 127.0.0.11 valid=30s; proxy_pass http://QRcode:3016; } Copy that to your letsencrypt nginx proxy-confs: cp qrcode.subfolder.conf /mnt/user/appdata/letsencrypt/nginx/proxy-confs And restart your letsencrypt docker. Go to yoursite.com/qrcode.html, type in some words and hit enter to submit. Voila! I apologize for any mistakes. Let me know so I can fix them.

What is the viewDockerLogSize script telling me exactly? The zoneminder log gets big pretty fast. It doesn't seem to be the same info you get from docker logs <container>. Can I move that log outside the container somehow? Thanks!

Where are the nerdtools vim settings coming from? There is no /etc/vimrc or ~/.vimrc. Typing :scriptnames in vim shows a /usr/share/vim/vimrc but even removing that doesn't get rid of some of the settings, like breaking normal mouse select for example.

For VNC you have to put the password in the config, it won't prompt for it like rdp and ssh do. For ssh my unraid box didn't have a dns search entry in /etc/resolv.conf so neither did the container (I just made a userscript to add a search domains at array start). Also figured out you need to bind mount any rdp fileshare dirs into the container. Not sure how you do that with the unraid docker interface so I'll just use the other containers for now I guess.