Greyberry

Members
  • Posts

    105
  • Joined

  • Last visited

Posts posted by Greyberry

  1. I ended up disabling autostart of the default network, as suggested here.

     

    virsh net-autostart --disable default

     

    I want to mention this from the same mailing list, as unraid desperately lacks important network management features for vms, which would make handling those issues much easier:

    Quote

    Any network defined with <forward mode='nat'>, <forward mode='route'>, or no forward element at all, will result in libvirtd creating a new bridge device for that network. If the network also has an IP address defined, then a dnsmasq instance will be started, listening *only* on that IP address (i.e. only on the bridge) for DNS requests. If the <ip> element also has a <dhcp> section, then the dnsmasq instance will also listen on the dhcp port *only* on the bridge.

    For <forward mode='bridge'> and <forward mode='hostdev'>, no bridge device is created by libvirt, no IP address is configured on the host, and no dnsmasq is started.

     

  2. Hi everyone,

     

    Quote

    docker: Error response from daemon: driver failed programming external connectivity on endpoint lancache-dns (c4af6531e9ae67e9ea29ab158d435ee6fdccbc54f502488fdad71c1062aeae4a): Error starting userland proxy: listen udp4 0.0.0.0:53: bind: address already in use.

     

    Is there a way to modify /etc/libvirt/virt*.conf to disable dnsmasq? As i only use VMs connected to my network anyways and therefore do not need DHCP and DNS services internally.

     

  3. you have to use a custom network for your containers to reach each other via hostname

    docker network create --subnet 172.20.0.0/16 mysupercooldockernetwork

     

    now it works.

  4. Hello everyone,

     

    I want my containers to communicate between each other via hostnames.

     

    • Do I have to set the --hostname parameter on every container manually, or is there a default hostname?
    • Is this possible in the standard Bridge network unraid creates, or do i need to create a custom network?

     

    So far I tried using the container names as hostname which did not work out. But communication between containers using the internal 172.17.0.0/16 address obviously works.

     

    I need a hint in the right direction, how is this supposed to work?

  5. Feature Request: SVGs and Material Design Icons

    Hi, uploading and fetching Icons every time I open up the Docker page is a little bit too complicated for my taste. And I could not get base64 encoding to work with svgs... Can someone please do a short checklist for how someone can do that?

     

    It would be really nice to only write in the icon-name of a Material Design Icon and beeing done with icon handling, like Flame does implement it.

     

     

    SVG Hack

    Meanwhile I worked arround that limitation by a little hack, which is basically a html injection, lol.

    How to do it:

    • Select an icon at https://pictogrammers.com/library/mdi/
    • Click "Copy SVG" on the right upper corner
      • (you could basically use any svg you want)
    • add a width="36px" attribute to the svg element
    • add a height="32px" attribute to the svg element
    • add "></img> in front of the whole string (note: with aprostrophe)
    • add <img style="all: unset; visibility: hidden; on the end of the string

    It should then look like something like this:

    "></img> <svg width="36px" height="32px" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><title>television-classic</title><path d="M8.16,3L6.75,4.41L9.34,7H4C2.89,7 2,7.89 2,9V19C2,20.11 2.89,21 4,21H20C21.11,21 22,20.11 22,19V9C22,7.89 21.11,7 20,7H14.66L17.25,4.41L15.84,3L12,6.84L8.16,3M4,9H17V19H4V9M19.5,9A1,1 0 0,1 20.5,10A1,1 0 0,1 19.5,11A1,1 0 0,1 18.5,10A1,1 0 0,1 19.5,9M19.5,12A1,1 0 0,1 20.5,13A1,1 0 0,1 19.5,14A1,1 0 0,1 18.5,13A1,1 0 0,1 19.5,12Z" /></svg> <img style="all: unset; visibility: hidden;
    • post this into Icon field
    • enjoy your svg as icon

     

    Definitely not perfect, but gets the job done.

    Support for it would be better though.

     

    1364934923_Bildschirmfoto2024-02-07um00_44_56.png.fbb38753d136161eaadf5ae2ae0764d6.png

     

    418315538_Bildschirmfoto2024-02-07um00_45_55.png.c26d9872adc96408f98dbd4d9455c4b4.png

  6. Is it possible to make a vpn network without access to the server itself (webinterface, dockers, etc.)?

    only the clients should see each other + access to the standardgateway of unraid (internet).

     

    and: when a wireguard-client is routing its traffic through the tunnel to the internet, does Unraid do NAT or is the wireguard ip address shown on the standardgateway?

  7. Hi,

     

    I installed an application from dockerhub. I configured everything and it is running, however the webui of the application is giving me an error:

    Unbenannt.PNG.3aee6f8c1b918e03c1f5e585652212cb.PNG

     

    How can i make this compatible with the image/container? (note: /data is mounted into /mnt/cache/appdata/application)

     

    I experimented with variables PUID, PGID, UMASK but could not make it work.

  8. On 8/30/2023 at 11:23 AM, nuhll said:

     

    So, command in the go file it is?

    sudo mount -o remount,size=XXXG /dev/shm

     

     

    On 8/30/2023 at 11:23 AM, nuhll said:

    I have all my dockers on dev shm and made it nearly full ram ("only" have 64gb)

    Cool! Is there a guide how to do that?

  9. On 4/7/2023 at 8:25 AM, Avsynthe said:

    Hey all,

     

    So I'm trying to set up Recognize. I'd like to run it without Tensorflow WASM mode as I understand Video tagging is not available with that setting.

     

    Under Tensorflow WASM mode, it reads:

     

    Could not check whether your machine supports native TensorFlow operation.

     

    And with it disabled, under Node.js it reads:

     

    Could not load libtensorflow in Node.js. You can try to manually install libtensorflow or run in WASM mode.

     

    Under that is the note:

     

    If the shipped Node.js binary doesn't work on your system for some reason you can set the path to a custom node.js binary. Currently supported is Node v14.17 and newer v14 releases.

     

    I'd also like to run Tensorflow GPU mode, so I'd like to know if GPU passthrough works in the traditional sense with the --runtime=nvidia extra parameters and the NVIDIA_VISIBLE_DEVICES variable set.

     

    Thanks in advance!

     

    I would like to push this issue, as it is also affecting me.

    Quote

    Could not check whether your machine supports native TensorFlow operation. Make sure your OS has GNU lib C, your CPU supports AVX instructions and you are running on x86. If one of these things is not the case, you will need to run in WASM mode.

     

  10. I have a (for me) huge amount of ram installed on my homeserver and tested creating dummyfiles on /dev/shm.

    Unfortunatelly it seems that only half of the installed RAM is available. After a certain point i get "No space left on device." even though RAM is only half filled.

     

    Is there a way to change that so that my RAM is not installed but unused all the time?

  11. On 8/20/2023 at 3:39 PM, dlandon said:

    You want to be careful doing that.  Scripts will not run directly from the flash drive for security reasons.  You would need to copy it ti /tmp/ somewhere, change it to execute, and run it from there.

    Thats why I am asking if there is a method to use User-Scripts (the plugin) somehow to start a script, not from gui but via terminal.

  12. 8 hours ago, alturismo said:

    then it should look like this as sample when you directly connecting to your nextcloud server in LAN (with its direct IP)

     

    Yes when I'm connecting directly it looks like this, but i never use http in the first place on the client side. I wanted to use it on the "backend" so to speak.

     

    [Client] -------- https --------> [Reverse Proxy] -------- https -------> [NextCloud] // No Error Occurs

    [Client] -------- https --------> [Reverse Proxy] -------- http -------> [NextCloud]  // cardav Error Occurs

     

    8 hours ago, alturismo said:

    if you dont get this error (like me posted above) while connectd directly ... my step would be (as im using swag i cant really tell) if NPM is calling this issue, may ask there if anyone may could help ... or if it needs to be added there too ...

    Okay, i might look further into this in two weeks (I'm not at home the next days). Thank you for taking me seariously in the end, and thank you for your time.

     

    8 hours ago, alturismo said:

    and btw. your expected benefit (reduce overhead) wont be noticeable if there is any at all ... just as note.

    I remember reading an article with benchmarks some time ago, where there was showing quite a difference on low-end machines. I can't find it at the moment unfortunatelly.

    But maybe you're right and it is not worth the hassle on newer machines.

     

  13. Na, i got a user-script already in place, but want to call it from the terminal.

     

    This in itself would not be the problem, because one could run:

    /boot/config/plugins/user.scripts/scripts/[Name_Of_Script]/script

     

    But if you do that, it would run as a standard script in terminal.

     

    I wanted to know if there is a way to run it, so that the progress is also represented in the User-Script Plugin GUI, like running it from there.

     

    Unbenannt.PNG

  14. 5 hours ago, alturismo said:

    this has nothing todo with http or https, this is a known message and you will find plenty of infos if you just google it and add some extras in the proper file, im pretty sure its even mentioned several times in this thread actually.

    it has everything to do with http and https, because (A) i have the proper rules in place:

    location ^~ /.well-known {
            # The rules in this block are an adaptation of the rules
            # in `.htaccess` that concern `/.well-known`.
    
            location = /.well-known/carddav { return 301 /remote.php/dav/; }
            location = /.well-known/caldav  { return 301 /remote.php/dav/; }
    
            location /.well-known/acme-challenge    { try_files $uri $uri/ =404; }
            location /.well-known/pki-validation    { try_files $uri $uri/ =404; }
    
            # Let Nextcloud's API for `/.well-known` URIs handle all other
            # requests by passing them to the front-end controller.
            return 301 /index.php$request_uri;
        }

    and (B) the warnings do not occur, when i use https behind the reverse proxy.