Igor

When a current Le potato will fire apt update and reboot it will get stuck, right? Per board we can control: u-boot, bsp and DTB, kernel is identical to all meson64 family. Perhaps @Neil Armstrong can give us a hint with the error http://ix.io/2bUZ ?

The problem will be updating. We have to fix this kernel (hack DTB for Lepotato) somehow.

If this https://github.com/armbian/build/blob/master/lib/distributions.sh#L186 works for us, it should work for you.

I don't know.

Repository will be updated with matched headers in few hours from now. If you need this faster, you can build an image with headers and latest kernel https://github.com/armbian/build All kernels support DKMS but one from rockchip. Not sure if this is the problematic one.

We drop support for Odroid C1 due to not sufficient resources / help. Currently it doesn't compile so this job represent several days of R&D ...

I can check this fastest in 1-2 weeks.

... while Odroid C2 works fine.

Last try is to try clean git clone build script and start from scratch?

./compile.sh docker single [ o.k. ] This script will try to update Already up to date. Already on 'master' Your branch is up to date with 'origin/master'. [ o.k. ] Using config file [ /home/igorp/Dokumenti/build/userpatches/config-docker.conf ] [ .... ] Using existed a armbian Docker container [ o.k. ] Running the container [ o.k. ] Using config file [ /root/armbian/userpatches/config-single.conf ] [ o.k. ] Preparing [ host ] [ o.k. ] Build host OS release [ bionic ] [ o.k. ] Running in container [ docker ] It works for me - latest sources, using Debian Buster as a host. Yes - I would propose to delete the container / image.

@Alexey Manually update Docker file and docker config - copy from config/template ... and it will work.

It seems like a bug.

Agree. We don't want to see spam and wrongly positioned messages which is why we need someone to take care of this. Volunteers are helping (me at the end) covering this, sometimes dirty, job. Exactly. And this tiny change in tooling will help them to help us. Perhaps not at all. At virtual no cost on users liberty. Now, do we need to vote for this small unimportant function to be implemented? Perhaps we should only do that in first place?

It is possible that we need to enable this: https://github.com/armbian/build/blob/master/config/kernel/linux-sunxi-current.config#L2560 https://linux-hardware.org/index.php?id=usb:2357-0109

I made a test with my Cubox and network worked fine. The only difference is that I used a kernel from trunk, 5.4.20 ...

Basic hardware support for Odroid N2 was developed in private kernel 4.9.y and functions are slowly getting to the mainline kernel. Some functions has to be written from scratch, some come in with minor changes, some will never get there ... This situation is normal and expected. Having functionality of development and stability / modernity of modern kernels is rare and expensive luxury. Hardware support is identical in all cases. If it doesn't work under one, it doesn't work elsewhere. Different versions are for applications like OMV which works only with Debian. Some people needs this and that Ubuntu for whatever reasons regardless of hardware support.

https://dl.armbian.com/_download-stats/

Strange. I have no other ideas and no resources to dig in. Half or non functional WiFi on not supported board is not a release showstopper. Will need to wait for the future.

Can you try if this helps: https://github.com/armbian/build/pull/1780/files Creating a file /etc/NetworkManager/conf.d/10-override-wifi-random-mac-disable.conf with this content following by reboot or NM restart: [connection] wifi.mac-address-randomization=1 [device] wifi.scan-rand-mac-address=no

This forum provides support for Armbian (on Duo and others). Welcome to check it. Also we have different/better way of handling hardware and I/O: https://github.com/sgjava/userspaceio

Email validation is not good enough to fight spam. New users need to have one approved post .. but sometimes is hard to decide is it a bot, human or spammer. That's why we need something in between. Typical spammer replies to few posts, then edit those posts and add links. We want to deal with that.

There is a text "USB Type-C is not PD-compliant" with link to your topic since months. I did update wording and add it to the missing boards.

The plan is to put more limit to newbies, perhaps create another group between validating and users. No limiting normal users. This is how I see it, but other might see it different which is why we talk first and do action on some common grounds later. If most of us will think this is a stupid idea, we won't proceed.

Freedom to edit will only be limited to users we don't know. Nobody will even notice.

Bionic is somehow still preferred for desktops ... You mean also to remove Buster desktop from building? https://github.com/armbian/build/blob/master/config/targets.conf