raschid

So all Ubuntu version work with NM (bionic, disco, eoan), both debian version currently do not (regardless of kernel version) but work with with /etc/network/interfaces. So @yoq I would suggest you stick with an ubuntu flavour for the time being

Yeah. XR819-Wifi stopped working with stretch and buster, but still works with bionic. Here's my syslog with a buster build: ov 30 16:45:56 localhost NetworkManager[1116]: <info> [1575132356.6690] wifi-nl80211: (wlan0): using nl80211 for WiFi device control Nov 30 16:45:58 localhost NetworkManager[1116]: <info> [1575132358.1449] device (wlan0): driver supports Access Point (AP) mode Nov 30 16:45:58 localhost NetworkManager[1116]: <info> [1575132358.1555] manager: (wlan0): new 802.11 WiFi device (/org/freedesktop/NetworkManager/Devices/4) Nov 30 16:45:58 localhost NetworkManager[1116]: <info> [1575132358.1656] device (wlan0): state change: unmanaged -> unavailable (reason 'managed', sys-iface-state: 'external') Nov 30 16:45:58 localhost kernel: [ 34.419112] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready Nov 30 16:45:58 localhost kernel: [ 34.423469] IPv6: ADDRCONF(NETDEV_UP): wlan0: link is not ready Nov 30 16:45:58 localhost NetworkManager[1116]: <info> [1575132358.1790] device (wlan0): set-hw-addr: set MAC address to AA:88:8F:8F:8F:AF (scanning) Nov 30 16:45:58 localhost NetworkManager[1116]: <info> [1575132358.2924] device (wlan0): supplicant interface state: init -> starting Nov 30 16:45:58 localhost wpa_supplicant[1115]: Could not set interface wlan0 flags (UP): Invalid argument Nov 30 16:45:58 localhost wpa_supplicant[1115]: nl80211: Could not set interface 'wlan0' UP Nov 30 16:45:58 localhost wpa_supplicant[1115]: nl80211: deinit ifname=wlan0 disabled_11b_rates=0 Nov 30 16:45:58 localhost wpa_supplicant[1115]: Could not set interface wlan0 flags (UP): Invalid argument Nov 30 16:45:58 localhost wpa_supplicant[1115]: WEXT: Could not set interface 'wlan0' UP Nov 30 16:45:58 localhost wpa_supplicant[1115]: wlan0: Failed to initialize driver interface Nov 30 16:45:58 localhost NetworkManager[1116]: <error> [1575132358.3304] sup-iface[0x1175c18,wlan0]: error adding interface: wpa_supplicant couldn't grab this interface. Nov 30 16:45:58 localhost NetworkManager[1116]: <info> [1575132358.3309] device (wlan0): supplicant interface state: starting -> down Nov 30 17:21:39 localhost NetworkManager[1116]: <warn> [1575134499.3085] device (wlan0): re-acquiring supplicant interface (#1).

Thanks actually I am not trying the get sound to work right now - I would like to understand why I does not function as it is supposed to, though ...

Interesting. When did you build that OPiZero image?

"current" on an OPi zero also seems to be missing analog audio at this time - so this is not specific to this board.

Just tried "current" - same audio issue: root@sunvell:~# cat /proc/asound/cards 0 [allwinnerhdmi ]: allwinner_hdmi - allwinner,hdmi allwinner,hdmi so H3_Audio_Codec is also missing in 5.3.x

@guidol I have a similar setup - PC GBit interface and R69 connected to the same switch => no up/down events in dmesg. IR indeed works. Audio indeed does not. The <codec> node for H3 devices which is activated by the sun8i-h3-analog-codec overlay is defined in sunxi-h3-h5.dtsi and thus part of our dts here. No idea why the overlay fails. By the way, building via armbian build environment works fine now: ./compile.sh BOARD=sunvell-r69 BRANCH=dev

@guidol, thank you for the kind hint - I updated the link. I added a cpu voltage regulator node to the device tree - oddly enough to build a viable kernel there seems to have to be one ... although this box reportedly does not have anything remotely resembling a PMIC (still working on that ...). Regarding the overlays: the IR overlay seems to be missing while the audio overlay appears to have been compiled for another kernel version ('bad magic'). No idea how armbian generates overlays, sorry. But IR is "OKed" in the DTS and should work anyway.

Would you give this a try, @guidol? https://drive.google.com/open?id=1aIUBR61RgxyKS6xi19kzsvWbCXJ0MeED

Mine (red PCB, v1.2) is running 5.3 - but with a modified device tree if I remember correctly. I'll check ..

Mine is the red 2017 version (rev.1.3) - the boards seem to be very different. Yours has 8G NAND, mine has 2 Samsung DDR3L SDRAM modules. Lot's of other differences from the pics. I built the OS using Armbians build environment. There is a "community supported" profile for the Sunvell R69 which in its current (mainline) version does not run on my device. I did not test the (FEX-based) legacy version which is available as well. The kernel image provided above was built with a somewhat modified kernel patch (I had to add a cpu voltage regulated to the device tree to make it work). Cheers!

This image is currently running on exactly this box (android TV-box R69). I used Etcher to burn the image to a Class 10 SAN-Disk card. Insert card and boot. There seem to be different versions of this box around - maybe yours is different from mine ... Cheers!

Hey @congtm - Try this armbian-build (Kernel 4.19.80): https://drive.google.com/open?id=1yoe5gkaeufAJQK2Z7hljda46VuHPdnvZ Cheers!

@sgjava the current device tree (dev) contains ... &usb_otg { dr_mode = "peripheral"; status = "okay"; }; can you check what the FriendlyElec version says here?

Thermals should work with dev for H2+/H3: see https://github.com/armbian/build/pull/992

Keep us posted. Good luck, @gounthar

You are right ... and out of luck for now: https://github.com/armbian/build/issues/1003#issuecomment-394987272 you could try a recent nightly build: https://dl.armbian.com/orangepizero/ One BananaPi running a recent kernel (Armbian 5.45 / kernel 4.17.0 rc7) has been up and running for 12 days now ...

@gounthar you might try downgrading the kernel: sudo apt install linux-image-next-sunxi=5.31 linux-dtb-next-sunxi=5.31 sudo apt install linux-u-boot-bananapipro-next=5.31

Have a look at this: https://forum.armbian.com/topic/6386-lime2-mainline-kernel-with-debian-9-stretch-becomes-unresponsive-forced-reboot-required/?page=2&tab=comments#comment-55225

Host is Xenial. As recommended here https://docs.armbian.com/Developer-Guide_Build-Preparation/

I have the same issue (with jessie and strech, xenial untried) - encountered the following error during build: Reading state information... hostapd is already the newest version (2:2.6-4~armbian5.38+1). sunxi-tools is already the newest version (1.4.2-2~armbian5.38+1). sunxi-tools set to manually installed. 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. [ o.k. ] Calling image customization script [ customize-image.sh ] [ o.k. ] Preparing image file for rootfs [ orangepilite stretch ] [ o.k. ] Current rootfs size [ 801 MiB ] [ o.k. ] Creating blank image for rootfs [ 1096 MiB ] 1.07GiB [ 37MiB/s] [====================================================>] 100% [ o.k. ] Creating partitions [ root: ext4 ] [ .... ] Creating rootfs [ ext4 ] Invalid filesystem option set: ^64bit,^metadata_csum tune2fs: Bad magic number in super-block while trying to open /dev/loop0p1 Couldn't find valid filesystem superblock. mount: wrong fs type, bad option, bad superblock on /dev/loop0p1, missing codepage or helper program, or other error In some cases useful info is found in syslog - try dmesg | tail or so. The image of course does not boot.

Thank you - that did it. And you are right: it did not solve my wifi issue

Wifi does not work with dev branch (4.17) on H2+/H3 systems: [ 10.807592] cfg80211: Loading compiled-in X.509 certificates for regulatory database [ 10.857239] cfg80211: Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7' [ 10.916465] platform regulatory.0: Direct firmware load for regulatory.db failed with error -2 [ 10.916479] platform regulatory.0: Falling back to user helper [ 11.432051] cfg80211: failed to load regulatory.db any hints?

Hmm, looks like the machine is running out of RAM. Maybe we have some memory leak ... Maybe you could check for decreasing available memory over some time?

Thx for the info @vlad59 - I would not recommend upgrading to more current versions at the moment.