Active threads

Go ahead and put that into a PR if it makes sense. That is the way to slowly improve the situation in armbian. These kinds of patches have to come from the community to be shared with everyone.

Thank you for sharing your work and insights. Please continue and don't give up!

Awesome, thank you so much for this. And it's already merged. Very nice indeed.

sounds like a genuine bug and not an error on your end. @Efe Çetin Can you confirm the issue?

It's super easy to compile your own image: "git clone https://github.com/armbian/build.git;cd build;./compile.sh" https://docs.armbian.com/Developer-Guide_Building-with-Docker/ https://docs.armbian.com/Developer-Guide_Build-Preparation/ https://www.youtube.com/watch?v=kQcEFsXEJEE FWIW, I just tried to download the two images from https://www.armbian.com/espressobin/ and did not have any issue. What problem exactly are you facing?

Wild guess: armbian-firmware or linux-firmware package?

Apologies for the very long delay; life happened. I grabbed the Ubuntu Noble 25.8.1 image, ran through initial setup, and re-ran the `apt` commands in my initial post. This got me Linux 6.12.56-current-rockchip, since time has advanced. Alas, here too `modprobe zfs` crashes in much the same way: [ 430.862366] spl: loading out-of-tree module taints kernel. [ 430.910755] zfs: module license 'CDDL' taints kernel. [ 430.910773] Disabling lock debugging due to kernel taint [ 430.910832] zfs: module license taints kernel. [ 430.911221] 8<--- cut here --- [ 430.914682] Unable to handle kernel paging request at virtual address 5e56ce80 when read [ 430.923821] [5e56ce80] *pgd=00000000 [ 430.927842] Internal error: Oops: 5 [#1] SMP ARM [ 430.933004] Modules linked in: spl(O) squashfs algif_hash algif_skcipher af_alg sunrpc snd_soc_hdmi_codec snd_soc_simple_card snd_soc_simple_card_utils snd_usb_audio snd_soc_rockchip_i2s snd_hwdep hantro_vpu snd_usbmidi_lib snd_soc_core rockchip_vdec(C) snd_ump snd_rawmidi rockchip_iep v4l2_vp9 snd_pcm_dmaengine rockchip_rga snd_seq_device v4l2_jpeg panfrost r8723bs(C) snd_pcm v4l2_h264 videobuf2_dma_contig rk_crypto dw_hdmi_i2s_audio gpu_sched dw_hdmi_cec v4l2_mem2mem snd_timer crypto_engine dw_wdt rk3288_gpiomem snd soundcore rockchip_thermal cpufreq_dt zram zsmalloc binfmt_misc sch_fq_codel fuse gpio_keys syscon_reboot_mode reboot_mode realtek [ 430.997122] CPU: 0 UID: 0 PID: 3102 Comm: modprobe Tainted: P C O 6.12.56-current-rockchip #1 [ 431.008012] Tainted: [P]=PROPRIETARY_MODULE, [C]=CRAP, [O]=OOT_MODULE [ 431.015204] Hardware name: Rockchip (Device Tree) [ 431.020456] PC is at resolve_symbol+0x134/0x36c [ 431.025522] LR is at cmp_name+0x14/0x18 [ 431.029805] pc : [<b01ac008>] lr : [<b01ab0ec>] psr: 30070013 [ 431.036796] sp : f27c1d78 ip : f27c1cb8 fp : f27c1dcc [ 431.042622] r10: f27c1ed8 r9 : af317102 r8 : af3152b8 [ 431.048447] r7 : af2859c0 r6 : af285c48 r5 : f27c1e50 r4 : af592300 [ 431.055737] r3 : 00000000 r2 : 5e56ce70 r1 : af317102 r0 : 00000001 [ 431.063028] Flags: nzCV IRQs on FIQs on Mode SVC_32 ISA ARM Segment none [ 431.071000] Control: 10c5387d Table: 06f2806a DAC: 00000051 [ 431.077407] Register r0 information: non-paged memory [ 431.083050] Register r1 information: 3-page vmalloc region starting at 0xaf315000 allocated at load_module+0x760/0x1b8c [ 431.095099] Register r2 information: non-paged memory [ 431.100740] Register r3 information: NULL pointer [ 431.105992] Register r4 information: 201-page vmalloc region starting at 0xaf57a000 allocated at load_module+0x760/0x1b8c [ 431.118242] Register r5 information: 2-page vmalloc region starting at 0xf27c0000 allocated at kernel_clone+0x9c/0x348 [ 431.130194] Register r6 information: 5-page vmalloc region starting at 0xaf284000 allocated at load_module+0x760/0x1b8c [ 431.142250] Register r7 information: 5-page vmalloc region starting at 0xaf284000 allocated at load_module+0x760/0x1b8c [ 431.154306] Register r8 information: 3-page vmalloc region starting at 0xaf315000 allocated at load_module+0x760/0x1b8c [ 431.166361] Register r9 information: 3-page vmalloc region starting at 0xaf315000 allocated at load_module+0x760/0x1b8c [ 431.178416] Register r10 information: 2-page vmalloc region starting at 0xf27c0000 allocated at kernel_clone+0x9c/0x348 [ 431.190470] Register r11 information: 2-page vmalloc region starting at 0xf27c0000 allocated at kernel_clone+0x9c/0x348 [ 431.202523] Register r12 information: 2-page vmalloc region starting at 0xf27c0000 allocated at kernel_clone+0x9c/0x348 [ 431.214578] Process modprobe (pid: 3102, stack limit = 0xa4dfd6e7) [ 431.221483] Stack: (0xf27c1d78 to 0xf27c2000) [ 431.226347] 1d60: b01ab490 b5f1dbc0 [ 431.235485] 1d80: 00000018 f2db7171 00000100 af2859c0 00000000 af3152b8 00000000 ac1f1f5c [ 431.244622] 1da0: b529a040 af592300 00005a8b 0006aa80 af59230c af7428b4 f2db7171 f27c1ed8 [ 431.253760] 1dc0: f27c1ebc f27c1dd0 b01ad948 b01abee0 f2e5c524 00000000 00000003 00000000 [ 431.262896] 1de0: 004ea3f8 00000000 00000001 b132d7f4 b132d80c b132d79c 00000000 00000000 [ 431.272033] 1e00: 00000028 b132d790 00000000 af592300 af5923fc 00000001 f2e5c4ac b1bd8b84 [ 431.281171] 1e20: 00000000 b1c8a214 b5eb76c0 f2e5c524 00000000 00455524 f27c1e64 f27c1e48 [ 431.290308] 1e40: 7fffffff 00000000 00455524 b5eb76c0 00000000 00455524 f27c1ebc f27c1e68 [ 431.299444] 1e60: b037a83c b06c390c 00000002 b02a4d7c f2a07000 00455524 00000001 00000000 [ 431.308581] 1e80: 00455524 00000000 00004004 ac1f1f5c f27c1ed4 00000000 b5eb76c0 004ea3f8 [ 431.317718] 1ea0: 00000000 b1c8a624 b663c708 b5eb76c0 f27c1f44 f27c1ec0 b01ae768 b01acb4c [ 431.326855] 1ec0: f27c1ed4 7fffffff 00000000 00000002 b0332b30 f2a07000 f2d154c1 f2c759c0 [ 431.335991] 1ee0: f2a07000 00455524 f2e5be6c f2e5bcec f2d8970c 00082824 000a81a4 000a639c [ 431.345128] 1f00: 000c6142 00000000 000a638c 00000028 00000029 00000016 00000000 0000001e [ 431.354266] 1f20: 00000000 ac1f1f5c 00000004 00000056 b1c8a214 00000001 f27c1fa4 f27c1f48 [ 431.363402] 1f40: b01aeb2c b01ae6d8 004ea3f8 b5eb76c0 00000000 b663c708 00000000 b1c8a37c [ 431.372539] 1f60: 00000000 00000000 f27c1f68 f27c1f68 fffffffc ac1f1f5c 00000006 004ec5bc [ 431.381676] 1f80: 0210c4ec c8b8ca00 0000017b b01002c8 b529a040 0000017b 00000000 f27c1fa8 [ 431.390812] 1fa0: b0100060 b01ae9a8 004ec5bc 0210c4ec 00000004 004ea3f8 00000000 004eba84 [ 431.399950] 1fc0: 004ec5bc 0210c4ec c8b8ca00 0000017b 0210c550 a6f40cd8 0210b190 ae8cb2b4 [ 431.409086] 1fe0: ae8cb170 ae8cb160 004e2707 a6b97ca2 40070030 00000004 00000000 00000000 [ 431.418222] Call trace: [ 431.421046] resolve_symbol from load_module+0xe08/0x1b8c [ 431.427084] r10:f27c1ed8 r9:f2db7171 r8:af7428b4 r7:af59230c r6:0006aa80 r5:00005a8b [ 431.435831] r4:af592300 [ 431.438654] load_module from init_module_from_file+0x9c/0xdc [ 431.445080] r10:b5eb76c0 r9:b663c708 r8:b1c8a624 r7:00000000 r6:004ea3f8 r5:b5eb76c0 [ 431.453827] r4:00000000 [ 431.456649] init_module_from_file from sys_finit_module+0x190/0x2cc [ 431.463754] r6:00000001 r5:b1c8a214 r4:00000056 [ 431.468907] sys_finit_module from ret_fast_syscall+0x0/0x54 [ 431.475232] Exception stack(0xf27c1fa8 to 0xf27c1ff0) [ 431.480864] 1fa0: 004ec5bc 0210c4ec 00000004 004ea3f8 00000000 004eba84 [ 431.490002] 1fc0: 004ec5bc 0210c4ec c8b8ca00 0000017b 0210c550 a6f40cd8 0210b190 ae8cb2b4 [ 431.499138] 1fe0: ae8cb170 ae8cb160 004e2707 a6b97ca2 [ 431.504772] r10:0000017b r9:b529a040 r8:b01002c8 r7:0000017b r6:c8b8ca00 r5:0210c4ec [ 431.513519] r4:004ec5bc [ 431.516343] Code: ea000036 e5922000 e1560002 0a000033 (e5923010) [ 431.523419] ---[ end trace 0000000000000000 ]---

You need to set this to y or m in your kernel config. https://github.com/NickAlilovic/build/blob/v20250306/config/kernel/linux-sunxi64-edge.config#L5745 https://www.kernelconfig.io/config_usb_acm

I guess that would answer the last part of my question, thanks. Any comments on the work in that branch?

Okay I gave up on it and done a fresh install.. Took me 5 hours to get it back up and running, to reconfig x11vnc and samba stuff. Starting a freshly baked image proved to be a nightmare - for whoever is interested I share my story; For some reason it did not detect wireless dongle for keyboard. So I could see it over hdmi but no input was possible. Did not have a wired keyboard at hand. at first run it did not became accessible over ssh. but an ip address was assigned to it by dhcp router; I could see a new ip when I'd start it up. but it would not respond to ssh requests. But serial connection saved the day. Luckily I did have a serial adapter. Created username/password and then it was accessible over ssh and I could ditch the crappy tv screen I was using Selected wrong locale on setup, was unable to get rid of it without commenting wrong locale lines in ~/.bashrc Then I spent some few hours to get vnc up and running with lightdm and crapload of other dependencies, packages. Thanks AI for assistance. When asking quality questions and sharing messages/outputs it was super useful (I dont keep anything confidential there) Manually created /home/domas/startxfve_headless.sh script, made it executable, the whole orderal Not to mention xfce was not installed by default on this debian/minimal image. It was fun one too. https://docs.armbian.com/User-Guide_Armbian-Software/Desktops/ - armbian-config --cmd XFCE01 just ran the script and nothing happened without any error Had to do it manually but I think it worked the first or second time. And the most fun one since it included multimeter and soldering iron: then it did not detect my disk over sata no matter what I do. Apparently measured 0v at hdd power connector. Not sure how I killed it. After some fiddling around I connected jst connector's for hard drive 5V power to one of gpio pins that supply 5v. HDD spins up - YAAAY. It was short-lived. Disk Still undetected. Probably usb to sata controller is not powered on since 5v regulator fried. (thanks AI for ideas) So then I jumped a wire from said gpio 5v pin to the back of what used to be 5V connector for hard drive so it both feeds the disk and backfeeds the boards 5v regulator with the needed usb to sata controller. I don't use gpio. ITS ALIVE. No, I did not believe it would work. But at this point I had nothing to lose. Sketchy? Yes. Have I ever pulled worse? Sure. Proud of myself? You bet.

Thank you for sharing it Can you include these specifications, for quick reference for future readers? Board, Debian version, Kernel, Graphics (console, X11, wayland)

Bluetooth also works

I finally successfully updated Nextcloud to version 31.0.10 on my Odroid N2+ 😅 Despite all official information, an update is possible on a system running nextcloud installed via the script ! It is not necessary to buy a new hardware just to run nextcloud on your Armbian SBC in order to get the update. I did the following: Updated nextcloudpi to version 1.55.4 which was never released but just tagged on github Nextcloudpi tagged page on github via the ncp-config menu system (Updates/nc-update) Updated nextcloud to version 31.0.10, again, via the menu system (Updates/nc-update-nextcloud version 31.0.10) I don't really know why this information is not readily available on the nextcloudpi github page nor anywhere else I looked. Have fun Norbert

@c0rnelius I would like to report back with the news of full success: I have updated cmdline.txt to start with: console=tty1 console=ttyAMA0,115200 earlycon=pl011,0x107d001000,115200n8 And also added the following to config.txt: usb_max_current_enable=1 enable_uart=1 dtparam=uart0 dtparam=uart0_console dtparam=pciex1 dtparam=pciex1_gen=3 dtoverlay=disable-wifi-pi5 dtoverlay=disable-bt-pi5 All this with the HAT+ (in the picture above) still connected. The early boot console also works, and I also get a login prompt, which is all I wanted. After this I tried replacing ttyAMA0 to serial0 in cmdline.txt, to see if there is any difference, but they both seemed to work in the end.

@AndroidNewbie, what kind of help do you expect to receive? @Ducdanh Nguyen, are you sure it was my modified update.zip (hint: open it with any archiver and see for file "wipedata.flag" - is it there?)? Are old settings and apps stays in place after flashing? Are update.zip located in root of sd card? If all answers is yes, then try to update using usb flash drive instead of sd card.

I had to use 6.1.75 to get the analog out working. It is the only version where the analog out works - it shows in other versions but there is no playback

/dev/serial is also gone on 25.8.2

There's a new version of the kwiboo ffmpeg+v4l2request function https://code.ffmpeg.org/Kwiboo/FFmpeg/src/branch/v4l2request-v3 And it was shown as a work in progress in the ffmpeg community https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20847 Then, it was noticed in the mpv community https://github.com/mpv-player/mpv/pull/14690#issuecomment-3500141427 They previously said that v4l2request first needs to be integrated in ffmpeg officially, then mpv would start working on v4l2request This is needed, to have mpv+v4l2request compiling successfully and working under Debian Trixie (my experience)

Installation of autheila package ("apt install authelia") does does end with error due to /var/log/journal beeing a symlink to /var/log.hdd/journal. Once the symlink is removed and the directory is created instead, the installation can be completed successfully. The symlink can be recreated afterwards and authelia wil steill run normally - until the next package update.

Looking for help getting mine running I have a very basic understanding of this device and am learning Linux on my own if someone could message me so we can work through this I would love to get this running

You have to change the fancontrol.service file and set the option PrivateDevices to no. Do systemctl edit fancontrol and insert [Service] PrivateDevices=no

Hello, I will soon be selling a working Helios64 from Kobol (with 2.5G fix done from factory). I also have 3x 6TB hard drive that could come with it if needed. Based in France, I'm able to send within EU. Please DM me if interested.

Same/similar issue. Armbian Bookworm v25.8.1 on BTT Pi. wlan0/wlan1 devices are visible, but after first reboot, do not connect anymore. Even after stopping any networking-related services/processes, "iw wlan0 scan" just returns "device or resource busy". Changing the netplan renderer to network-manager as recommended did not work for me either. I also experienced the sunxi64-mmc kernel error messages when trying to reload the rt8189fs module. After switching kernel from 6.12.47-current to 6.16.8-edge, WLAN worked right away. But in exchange the wired eth0 device is now gone completely... you can't have it all (Maybe related to this?)

Yes the WTL-40110175 3.8v 10.000mAh battery has 3x red (PB+) and 3x black (PB- GND) and 1x yellow (Overvoltage?) or Temperature) Here in Turkey I do only find a 3.7v 10.000mAh Cell, but without the yellow cable (only red and black): https://www.n11.com/urun/1260110-37v-10000mah-li-polymer-pil-devreli15a-87767963 The battery PDF gives me very less clues Pinebook-WTL40110175-3_8V-10000mAh-14inch.pdf

In the end, I couldn’t recover the device and ended up purchasing a new one. It was also a V5.1 board, but fortunately, it booted from the SD card without any extra steps and installed successfully. It seems we should keep in mind that even boards with the same version may not always work the same way.