laibsch

That is a wise thing to do and might be a good default for armbian to adopt.

OK, can you do the following for me, please, to be sure this is properly fixed? Kindly share the console output for me to have a look at. sudo apt reinstall wireless-regdb dpkg -S /lib/firmware/regulatory.db* Then reboot and see if your problem remains fixed. For now, I suggest you keep the wireless-regdb-2025.02.20.tar.xz file around on the NanoPi in case you need it again.

please share more information about the problem you ran into for the sake of the next guy coming to the forum to look for a solution to the same problem. Thank you.

some progress $ sudo rfcomm bind /dev/rfcomm0 41:BB:00:D0:D3:D8 && ls -l /dev/rfcomm* crw-rw---- 1 root dialout 216, 0 Jul 7 22:46 /dev/rfcomm0 but $ sudo hcitool info 41:BB:00:D0:D3:D8 Requesting information ... Can't create connection: Input/output error

please be sure to share whatever learnings you have with us

I'd say, boot from a different medium and then try the dd commands again (use pv for a nice progress bar) to see if you have the same errors again.

Thank you for reporting back. I am happy to hear you were able to find another way to solve the underlying issue without needing the v4l2loopback-dkms package anymore. I see you already have backports enabled but probably with low priority (as you should) by default. You can check "man apt" for information how to install that specific version (or use aptitude like I do). That being said, "sudo apt install v4l2loopback-dkms/bookworm-backports" should probably get you the version you need and hopefully it would install fine.

Thank you for the report and this indeed does look like a general bug. It appears as if this comes from the file 10-armbian-header and in particular line 199 or possibly line 214, but I cannot look into it further right now. It's probably got something to do with network not being configured at the point at which you encountered this. I do certainly encourage you to open a ticket on github with the first entry on the pulldown "bug report for the armbian build framework".

$ hciconfig hci0: Type: Primary Bus: UART BD Address: B0:F1:EC:FD:65:CE ACL MTU: 1021:8 SCO MTU: 64:1 UP RUNNING RX bytes:10796 acl:47 sco:0 events:418 errors:0 TX bytes:40220 acl:43 sco:0 commands:331 errors:0 hcitool does see something, but bluetoothctl still does not $ hcitool scan Scanning ... 41:BB:00:D0:D3:D8 Baseus Bowie WM02 $ rfcomm connect 1 41:BB:00:D0:D3:D8 Can't connect RFCOMM socket: Connection refused $ bluetoothctl pair 41:BB:00:D0:D3:D8 Device 41:BB:00:D0:D3:D8 not available There is no rfcomm device $ ls -l /dev/rfcomm* ls: cannot access '/dev/rfcomm*': No such file or directory $ sudo rmmod rfcomm;sudo modprobe rfcomm -vvv

I believe you likely have an outdated version of the package. Here is the latest changelog entry for Ubuntu Noble. If you are using an older version, consider upgrading. If you are on bookworm you might want to consider the backports pocket. Or you can think about simply applying above patch locally. Next time, please provide relevant information and logs in your first post. Don't make us guess.

please give the output of "apt policy v4l2loopback-dkms" FWIW, installation worked just fine right now on my BPi M2+ $ sudo apt reinstall v4l2loopback-dkms Reading package lists... Done Building dependency tree... Done Reading state information... Done 0 upgraded, 0 newly installed, 1 reinstalled, 0 to remove and 5 not upgraded. Need to get 30.6 kB of archives. After this operation, 0 B of additional disk space will be used. Get:1 http://ports.ubuntu.com noble/universe armhf v4l2loopback-dkms all 0.12.7-2ubuntu5 [30.6 kB] Fetched 30.6 kB in 1s (20.8 kB/s) (Reading database ... 143046 files and directories currently installed.) Preparing to unpack .../v4l2loopback-dkms_0.12.7-2ubuntu5_all.deb ... Deleting module v4l2loopback-0.12.7 completely from the DKMS tree. Unpacking v4l2loopback-dkms (0.12.7-2ubuntu5) over (0.12.7-2ubuntu5) ... Setting up v4l2loopback-dkms (0.12.7-2ubuntu5) ... Loading new v4l2loopback-0.12.7 DKMS files... Building for 6.12.30-current-sunxi Building initial module for 6.12.30-current-sunxi Done. v4l2loopback.ko: Running module version sanity check. - Original module - No original module exists within this kernel - Installation - Installing to /lib/modules/6.12.30-current-sunxi/updates/dkms/ depmod....

Are you trying to copy the rootfs of the system you are currently booted into? That won't work. You need to boot a separate installation, the filesystem you want to dd should not be mounted. You didn't really say what went wrong, by the way.

https://github.com/alexl83 and https://github.com/chraac are listed as the maintainers for the Opi Z3

My Banana Pi M2+ is running quite fine on noble but something is amiss with bluetooth. bluetoothctl never shows any devices even when scanning for them and there are plenty of devices within arm's length. I am hoping for your help in diagnosing and ultimately fixing the issue. Logs:

please show the output of "apt policy zfs-dkms" but essentially, this looks the same as your other rodeo/struggle with this board and dkms. As mentioned there, you are much better off with btrfs instead of zfs on your SBC.

The way I understand it is that you only need the full firmware package in a few exceptional circumstances. What is the wifi chipset on your SBC? Do you have the wireless-regdb package installed? "dpkg -l wireless-regdb"

Thank you for sharing. Would be good to have your findings corroborated by some of official documentation. Thanks again for sharing.

If they boot, I wouldn't worry about any difference.

sounds good. good luck and keep us posted!

Yes, your logs show hundreds of errors, so it will be hard to pinpoint which one it is exactly. Here's a few things I can suggest that might move you along. "sudo service restart lightdm" # see if X now comes up on your TV, assuming that lightm is what you are using "sudo nano /home/*/.xsession-errors" # check the errors listed in those log files, they are more specific Can you try to flash an (older) image to a microSD card, boot off it and see what the results are? @Efe Çetin is the maintainer of your board, maybe he has a comment? Your board is in community-maintained status for armbian. There are no guarantees.

Do you know how to compile your own images? Do you know how to use "git bisect"?

I am always happy to see people keeping old hardware running. But I think it would be best to make the necessary changes in Armbian itself. I am happy to help out where I can be useful.

The ones I linked. You mentioned that you have no experience to compile stuff. One option is to target another SBC or a virtual machine so that you can focus on learning the build framework (explicit instructions are already given in some of the links) instead of having to deal with a multitude of issues (setting up your compilation host, getting compilation to succeed, finding the artifact to flash, only to then run into the next issue that armbian currently does not support your board). With a virtual system, you can be sure that you should get a working image. You can then focus on the compilation and familiarize yourself with that before jumping into real hardware enablement.

I suggest you check the logs for clues what is going wrong. syslog, .Xsession-errors, etc.

Thank you for reporting about this issue. What is the latest image that still works? Is it still available for you to download? Maybe this is the change you want to revert and see if that brings things back in line? But supposedly, that commit was boot-tested by Niklas. Your board does not have an official maintainer in armbian, so I suggest you contact Niklas to find out what the situation is for him.