coppit

Community Developer
  • Posts

    496
  • Joined

  • Last visited

Everything posted by coppit

  1. A couple of things I noticed... I got this working, but both my keyboard and my mouse seem to go to "sleep" after 3 seconds. I have to mash a bunch of keys or frantically move the mouse to wake the device up. After that the mouse and keyboard work fine, until I don't use them for 3 seconds. It looks like "powertop --auto-tune" was to blame. FYI, in case anyone else is running into this. I have two identical keyboards for two different VMs. Sadly, /dev/input/by-name only lists one of them. I'm using by-path, but I don't think that's any better than using the raw /dev/input/event. I wasn't sure how this would interact with USB Manager, where I had bound root hubs to the VMs. I unmapped those and instead mapped only the audio devices, leaving the keyboard and mouse to the XML config. Edit: Forget the comment below. I needed to use the one ending with "event-mouse" rather than just "mouse". I'm having trouble with one of my mice... I get this error: internal error: qemu unexpectedly closed the monitor: 2023-04-18T05:26:01.673092Z qemu-system-x86_64: /dev/input/by-id/usb-Gaming_Mouse_Gaming_Mouse-if01-mouse: is not an evdev device I tried with "/dev/input/by-id/usb-Gaming_Mouse_Gaming_Mouse-if01-event-mouse" as well and got the same error. Is there any hope for this mouse?
  2. @byb I think you want to remove the 2nd "<qemu:arg value="-object"/>" if you're not doing a mouse as well as the keyboard.
  3. I'm having trouble enabling amd_pstates with my 1950X Threadripper, on an ASRock X399 Taichi motherboard. I have the latest BIOS. I added "amd_pstate=passive" as a kernel boot option. When I run "cpufreq-info", I see that it's still using "acpi-cpufreq" as the driver. In dmesg I see "amd_pstate: the _CPC object is not present in SBIOS or ACPI disabled". Given that I see tons of ACPI messages, I'm guessing that ACPI is not disabled. P.S. I don't know if it's related, but I'm also seeing this: [Firmware Bug]: ACPI MWAIT C-state 0x0 not supported by HW (0x0) ACPI: \_PR_.C000: Found 2 idle states
  4. No. Using the Oculus Quest with wireless instead.
  5. Precisely. The version of Java inside the current container is 16. We need an updated container.
  6. In case it helps anyone... I ran sudo fs_usage -w | grep Volumes.Backups to see what was accessing my backups volume, besides my backup software. (I'm using Carbon Copy Cloner instead of Time Machine.) It turns out that Finder was updating every time the directory was being written, because I had a window open to the directory. After closing that window, speeds shot up. Hopefully that helps someone...
  7. Does anyone know how things change for USB Type C ports? I see the device under usb1, but after I pass that through, the VM doesn't recognize the device on that port. (An Oculus Quest.) I'm ordering a Type C -> USB A cable in the meantime to try a USB3 port.
  8. This container stopped working a bit ago due to an API change. In updating it, I realized that FileBot went to a paid model, and that there's another better supported container available. I'm removing this container. Please use the other one. Personally, I'm switching to Sonarr.
  9. How about if I added a bash script to the config dir, and you could put any command you want in there, including "yum install comskip"?
  10. I reverted the Alpine Linux changes late last night, so things should be back to how they were.
  11. Are you saying that where it prints: Configuration: USERNAME=foo PASSWORD=<hidden> DOMAINS=bar INTERVAL=baz DEBUG=zap It displays your password in the USERNAME line? A couple of thoughts: Make sure that your password is within single-quotes like: PASSWORD='your password goes in here' However, if your password contains a single quote, then you have to escape it. LMK if that's the case. Second, make sure that you aren't defining environment variables for the container that are overriding the config file.
  12. Moved to Alpine linux, reducing the image size from 232MB to 90MB Also: Make the change monitors more like services that restart if they crash. Added a feature to use polling, so that Windows shares can be monitored too.
  13. Does the client connect and complain about the version? Or does it not connect at all to the port? Are you remapping 10090 or 8090, or both?
  14. I moved the image to Alpine linux, which drops the size from 229MB to 64MB. I also added the ability to configure it using env vars instead of the config file. This is a pretty big change, so let me know if you guys see any issues.
  15. I switched the container from the phusion base image to Alpine linux. The size dropped from about 250MB to about 16MB. It also handles options set using environment variables rather than the config file better. This is a pretty big change, so let me know if you guys see any issues.
  16. New update is out. I moved from the phusion (Ubuntu-based) base image to Alpine. The image is 5x smaller. Also, you can now specify all the settings using only environment variables. In this case you don't need the config file. Env vars take precedence over the config file values.
  17. Hi everyone, I found a bug in the container that causes the recordings to be saved inside the container, instead of to the attached disk volume. I'll release a fix soon. You must run the following command before updating your container, or your videos will be lost! docker exec -it Xeoma bash -c 'cd /usr/local/Xeoma/XeomaArchive; for n in *;do mv "$n"/* "/archive/$n";done' I'll wait about a day before releasing the fix, in case anyone has auto-update configured for their docker containers.
  18. The author (or moderators of Community Applications) of the plugin template (https://raw.githubusercontent.com/coppit/unraid-mosh/master/mosh.plg) has specified that this plugin is incompatible with your version of unRaid (6.5.3). You should uninstall the plugin here: Maximum OS Version: 6.4.1 I didn't specify any such thing. Tested as working with 6.5.3. How do I get this fixed?
  19. User reports it works: https://github.com/coppit/unraid-mosh/issues/1 I installed it and it works as well. Are you the keeper of plugins? Can someone please point me to the latest plugin documentation? My plugin files aren't part of a txz like I see others have. When it gets installed by the system, the directory is missing "other" read and execute permissions, which means that emhttp can't read the icon. I also see that the latest plugin plg's don't have "iconfile". I'd like to renovate my plugin, but don't know how.
  20. Can someone please point me to the latest XML template for plugins, so that I can update my mosh plugin? It's currently not even showing up in CA, I guess because it's assumed to be incompatible with the latest version of UNRAID until I indicate otherwise.
  21. Whoa. My SageTV and UNRAID worlds are colliding! To be honest, I'm not very familiar with USB devices and docker. This post seems to suggest that you can pass the devices through. The container is based on Ubuntu, so you could try getting into it with "docker exec -it xeoma bash" and installing USB drivers there. Heck, maybe it will "just work" without any special drivers, being "universal" and all. If that works, I could see about including those drivers in the container. But to be honest, USB cameras are not that great. I would get HikVision cameras, which are affordable, have good features, and support power-over-ethernet. Then maybe get some powerline adapters to provide PoE and also networking over your power outlets. Something like this and this.
  22. Sorry. That was a goof on my part. I just pushed a fix. As soon as the container rebuilds on the hub, you'll have the fix.
  23. Thanks everyone for helping folks to enable the UI. I appreciate the assistance. I've updated the first post to this thread, so hopefully people won't have to keep asking.