Active threads

If I understood a bit better what the script is doing, i.e. this part: Then maybe I could help with updating it to work with the Radxa products.

Raised up a PR with a patch. 🤞

Was the auditd disabled ? Without it is difficult to see where the sysrq comes from. Re. monitoring ttyS1: you could try use screen: sudo screen /dev/ttyS1 9600 Replace 9500 if needed with the baud rate configured in the UPS

@oscylator678 I used this PPA https://launchpad.net/~ernstp/+archive/ubuntu/mesaaco Mesa 23.0, PanVk has already Vulkan 1.4.

@Fredrik thank you for testing.

It seems like there is a mix-up between specs for the framework and for the OS. A lightweight desktop can run on 512MB, yes but probably not much fun, depending on applications. Won't recommend running a web browser. I'd probably get some existing cheap sbcs with similar specs and play with them to elaborate if the performance is sufficient.

Thank you very much. I will try it. Thank you very much. I will try it.

I can confirm that my changes have now been merged into the upstream linux_openfd repo, so I have deleted my fork. Unfortunately I am unable to edit my instructions, but hopefully it should be clear from the instructions that if my fork is unavailable, then the original repo should be used (alternatively, if a mod could edit the post to remove the reference to my fork, to make the instructions clearer, that would also be appreciated).

Small update. The hack below seems to work as well. $ sudo systemctl edit armbian-hardware-optimize.service Add the following content between comments [Unit] Before=basic.target systemd-networkd.service netplan-wpa-wlan1.service Please note that armbian-hardware-optimize was picked at random to force anything that touches wlan1 to wait until slightly later in the boot process. No tests were done to determine at what exact point in the boot process wlan is "safe" to interact with.

Thanks for the reply, Tall Man. In a way, that was the point. My experience is really only with my home network use and with nvme information displayed by linux tools which seem to insist on the xnypz suffix on the root word nvme. I continued reading up on the general topic overnight and it seems that in the dtb tree world, the focus is on structuring cohesion and delaying actuating instances, so I am guessing different (more general) descriptors are used. My query was whether the precise descriptor was correct as I could see other options possible, one of which had, I think, popped up elsewhere. Since I had seen other reports of difficulties in loading systems to devices, I thought it a remote possibility that a typo had crept in. I still havent found anything definitive in my reading re the formats of the abstractions in the dt stack. I am interested in knowing about this but I as my Armbian is up and running now, and I identified the actual issue for me was a sd card error, the pressure is off . So, I hope this ramble was not too tedious and I do appreciate that you took the time to answer, and again, thanh you cheers

I am not sure if I am using the git command correctly to get mpv-0.3.9 with PR14690.. Please help roberto@orangepizero3:~$ mkdir mpv-official roberto@orangepizero3:~$ cd mpv-official/ roberto@orangepizero3:~/mpv-official$ git clone https://github.com/mpv-player/mpv roberto@orangepizero3:~/mpv-official$ cd mpv roberto@orangepizero3:~/mpv-official/mpv$ git checkout -b release/0.39 roberto@orangepizero3:~/mpv-official/mpv$ gh pr checkout 14690 <- downloads and applies the v4l2request changes roberto@orangepizero3:~/mpv-official/mpv$ nano video/out/gpu/hwdec.c <-- I see the added changes in the hwdec.c file roberto@orangepizero3:~/mpv-official/mpv$ meson setup build <-- everything configures ok, but at the end, it shows that I got mpv-0.40 I asked in the mpv github and a contributor said they are waiting for ffmpeg first incorporate v4l2request, then they will add the feature in mpv Applying the pr14690 manually on top of mpv 0.3.9 is something that i could do, if there’s no choice

i've updated the image to include info on screen like IP / cpu usage / ram usage (in %) any software can write to /tmp/screen to change content you can download here

This MainsailOS pull request contains a cb1 trixie image that is working. It does forces NetworkManager instead of netplan. After flashing look for a wifi txt file on the BOOT partition to pass wifi credentials.

Hello everyone, I got a tv box: J15 pro with rk3328 inside I tried the trunk version but it failed to boot then I tried the achived 23.11.1 version, It can boot normally, but if i do a full upgrade , the box failed to boot again. I think maybe the box do not compatible with kernel 6.12 so i tried to hold the 3 packages: linux-dtb-current-rockchip64 linux-image-current-rockchip64 linux-u-boot-rk3318-box-current and then upgrade , It works what can i do to help to fix the compatible problem? ps lshw shows the box comes with a wifi chip: rtl8189es

Small guide for fixing Ethernet on the latest Armbian. 1. Make sure you have the correct device: cat /proc/device-tree/model → OrangePi 3 LTS 2. Decompile the dtb to dts: dtc -I dtb -O dts /boot/dtb/allwinner/sun50i-h6-orangepi-3-lts.dtb -o ./sun50i-h6-orangepi-3-lts.dts 3. Check the correct pins (yours may differ): grep -i reset-gpio sun50i-h6-orangepi-3-lts.dts | head -n 1 → reset-gpios = <0x1e 0x03 0x0e 0x01>; 4. Create the file sun50i-h6-ethernet.dts. Use the provided template and substitute your pin values. My file for example: /dts-v1/; /plugin/; / { compatible = "allwinner,sun50i-h6"; fragment@0 { target = <&emac>; __overlay__ { snps,reset-gpio = <0x1e 0x03 0x0e 0x01>; snps,reset-delays-us = <0 10000 1000000>; snps,reset-active-low; mdio { ethernet-phy@1 { reset-gpios; reset-assert-us; reset-deassert-us; }; }; }; }; }; 5. Add your overlay: armbian-add-overlay sun50i-h6-ethernet.dts 6. Reboot and enjoy working Ethernet: reboot

Coming from the future, I created a fork with a patched version of moonlight-qt that works with Armbian 25.8.1 using vendor kernel 6.1.115 and Gnome as well as the mali blob g610-g24p0-wayland-gbm and of course, rkmpp. The performance is great. In a nutshell, what I actually patched were the direct OpenGL calls since these were not directly exposed in OpenGL ES that the mali blob uses.

I ended up creating a patched fork of moonlight-qt that works with the mali blob and rkmpp with excellent performance.

Hi community, I am experiencing an issue reinstalling Armbian on my Helios64. I tried all methods including booting from an SD card, flashing the image onto the eMMC directly but every time I boot, I get kernel panic (flashing status light). I also can't connect to the USB via TTY to diagnose the issue. My macbook keeps detecting the /dev/tty.usbserial-DT03OEIM and the next second it is gone. I wanted to install the original Buster image but I cannot find it anymore in any of the archives. It's kind of disappointing because what's the purpose of an archive then if not hold on to an image for a few years. My current theories are limited and I can only think that the newer images might not have the correct drivers, but as I said, I can't connect via TTY to even observe the boot log. I tried pretty much everything at this point: all of the images available (devbian, ubuntu), all flavors of them (minimal, omv). Booting into U-boot, Setting P10, P11 jumpers etc, etc. Can anyone help?

I try installing Debian Trixie using the installer by 2 differents approachs but It didn’t succeeds. 1) using a bootable USB key with ARMhf (Generic version), found on Debian web site 2) Has described in DEBIAN ARMhf installation manual, using hd-media to include the asus c100p correct .dtb (tree) files during the creation of the external USB. Regardless the method, the USB key with intaller, is not recognized, when typing CTRL+U, in developper mode. The Chromebook didn’t see the key, but it work well with Zutchi group, but not here. I only received a serie of bibs, when doing CTRL+U. I also try building an ARMBIAN image, but, Asus C100P is not available in choices, and I look in EXPERT mode, not supported, but didn’t find a generic ARMhf file having features: RK3288 processor, 4Gb of RAM, 16Gb of emmc, 2 x USB, USBc, hdmi, wifi, so I didn’t complete the process. I am open to suggestions, but my next trial is using ‘dd’ and install the ‘Velvet trixie’ image found on GitHub directly on emmc ‘drive’, if not succeed, I will try Archlinuxarm with PostmarketOS. Thanks for your comments and suggestions.

@Tomi Skies try the x96q images https://github.com/NickAlilovic/build/releases/tag/20250306

Everything worked fine, even tried some ffmpeg for hardware decode

This is on minimal. So no xorg or Wayland. But I use xorg normally. I generally start with minimal and work up. I despise both Ubuntu and gnome. You won't find those on any computer I own. I compromise with xfce for the little time I might use it before I have everything set up. I've just reversed direction and installed debian xfce stable. Mostly to avoid the hassle of replacing net plan with network manager. A PIA. I can uninstall all the stuff I don't like that comes with the desktop version. Probably easier. With xfce there are two screens, but the one on the built in screen seems to be double the size it should be. I can only see half the dog. I can live with that for now. At least it uses the Fullscreen of the HDMI.

"*** Warning - bad CRC, using default environment" This is safe to ignore. It is strange why Ubuntu doesn't work, while Bookworm does. This could be some build infrastructure issue. https://docs.armbian.com/Developer-Guide_Build-Preparation/

Welcome to Armbian, I'm your emotional support, I'll give you all the support if you want to start developing a custom firmware for this device, we have all the time in the world, one day this will be ready and we will put your name on it, it will be cool, I believe you have the ability to start volunteering, get in touch

for Pi-Hole i'm using: "cbcrowe/pihole-unbound" https://hub.docker.com/r/cbcrowe/pihole-unbound For docker Containers I'm on Casa-OS curl -fsSL https://get.casaos.io | sudo bash pihole-unbound.yaml