All Activity

At some point we will bump edge to 6.17. These bumps always take time since our patchset on top always needs adjustments. There are 3rd party ppa's for newer mesa. I suggest to use kisak since those are briefly tested. Otherwise go for oibaf which are automated untested, therefore latest mesa packages. In terms of Debian, no clue if there is a similar apt repo for such.

If you used edge, revert to 6.15

So I'm back to working on this board. I don't understand how I understood your instructions about erasing the eMMC successfully the first time, and now, the second time I need to do it because of needing to install Debian, I am unable to follow them. I imported the config, flashed the loader, parameters and u-boot. The board does not restart. I'm not even sure why wipe_part is mentioned if it isn't actually used. What am I missing here? I thought I followed the instructions but apparently not.

I'm a bit embarassed to say they are confusing to me. I am going to ask Grok and see if it can help me, thank you

Last try of 6.12 1 or 2 weeks ago, was still with display failure. 6.1 works fine, now, with both HDMI out and I believe SATA working at each boot. I believe than since your last change SATA works far better with 6.1 vendor kernel. I'm mainly using the 6.16rc7 armbian kernel on archlinux now where I compile time to time last mesa git, to take benefits of more GL features, Vulkan 1.4 support + zink, however some softwares like obs studio, blender, wings3D, GTK4-demos fails with zink, due to problems with broken DRI. X11 have some slowdown too, but Wayland is perfectly smooth with mainline kernel + mesa drivers. noticed that 6.16.0 was compiled with GCC11, where 6.16rc7 was compiled with GCC13.3, there is a lot of optimizations in perfs between the two GCC releases, and Debian13 is now shiped with GCC 14.2. kernel 6.17rc1 bring AV1 accelerated video on RK3588. and rc2 is now released, I don't know if there are projects to upgrade mainline kernel and/or Debian version? Debian is sadely shipped with Mesa 25.1.0. Vulkan 1.4 support was reached with Mesa 25.2. It is however maybe more stable, I didn't tried 25.1.0 stable?

https://forum.banana-pi.org/t/bpi-r4-eeprom/17836/13 suggests we should figure out how to give the info to uboot? or it already does generate it for us in one case? I can say I find eth0's local-mac-address in the running device tree, but the other two mac@ entries lack it. root@ratatosk:/boot# dtc /proc/device-tree 2>/dev/null |grep -PA 10 'mac@[0-9] ' mac@2 { phy-mode = "usxgmii"; compatible = "mediatek,eth-mac"; status = "okay"; pcs-handle = <0x3a 0x3b>; managed = "in-band-status"; phys = <0x3c>; reg = <0x02>; openwrt,netdev-name = "sfp-wan"; sfp = <0x3d>; }; -- mac@0 { phy-mode = "internal"; local-mac-address = [22 16 68 60 40 c2]; mac-address = [22 16 68 60 40 c2]; compatible = "mediatek,eth-mac"; status = "okay"; reg = <0x00>; phandle = <0x2c>; fixed-link { full-duplex; -- mac@1 { phy-mode = "usxgmii"; compatible = "mediatek,eth-mac"; status = "okay"; pcs-handle = <0x36 0x37>; managed = "in-band-status"; phys = <0x38>; reg = <0x01>; openwrt,netdev-name = "sfp-lan"; sfp = <0x39>; };

I don't think it's an SD card issue. Performance wise it's about twice the requirement. And if it was a bad block, I doubt the issue would be the same across two different images. As for logs... where? I tried running "sudo dmesg -T --color=always --level=err,warn | more" and nothing about AdGuard came up. I checked the AdGuardHome.yaml file and confirmed bind_host was set to 0.0.0.0 (I also tried setting it to 192.168.0.6... no improvement). The command "sudo netstat -tulnp | grep 3000" said command not found, and "journalctl -u AdGuardHome" came back with no entries.

new Ubuntu/netplan bug: https://bugs.launchpad.net/netplan/+bug/2120999

https://github.com/canonical/netplan/pull/278 <-- related netplan PR

maybe not a strictly a systemd bug... but a netplan stupidity. root@ratatosk:~# ethtool --show-permaddr eth1 Permanent address: not set https://github.com/systemd/systemd/pull/14448/files <-- PR that created this systemd match feature

Latest armbian update 25.11 trunk 45 also breaks boot...

Thx for your reply. Yes i tried both together. And just one. No difference. # Enable DRM VC4 V3D driver dtoverlay=vc4-kms-v3d # Enable the Touch Display 2 DSI panel dtoverlay=vc4-kms-dsi-ili9881-7inch

making a copy of the file from /run/, changing the YAML to use MAC, then diff root@ratatosk:~# netplan generate root@ratatosk:~# diff -purN /run/systemd/network/10-netplan-eth1.network /root/netplan-eth1-namematch.network --- /run/systemd/network/10-netplan-eth1.network 2025-08-19 20:10:54.900000038 +0000 +++ /root/netplan-eth1-namematch.network 2025-08-19 20:10:36.980000037 +0000 @@ -1,5 +1,5 @@ [Match] -PermanentMACAddress=6a:c9:72:53:13:d5 +Name=eth1 [Network] EmitLLDP=true feels like a systemD bug ☹️

@Piotr sTry this image https://github.com/NickAlilovic/build/releases/download/20250306/Armbian-unofficial_25.05.0-trunk_X96q-lpddr3_bookworm_edge_6.12.11_xfce_desktop.img.xz to fix Ethernet you need to edit this file https://github.com/NickAlilovic/build/blob/v20250306/patch/u-boot/u-boot-h616/159-add-x96-q-lpddr3-v1.3-defconfig.patch. Change 16 to 0. Then compile a new image. +&mdio1 { + rmii_phy: ethernet-phy@0 { + compatible = "ethernet-phy-ieee802.3-c22"; + reg = <0>; + }; +}; more information can be found here

Ok, so using the match name *does* work, still makes no sense. root@ratatosk:~# head /etc/netplan/20-ethLAN.yaml network: version: 2 renderer: networkd ethernets: eth1: match: #macaddress: '6a:c9:72:53:13:d5' name: eth1 #set-name: ethLAN addresses: [ "192.168.88.253/23" ] root@ratatosk:~# networkctl list IDX LINK TYPE OPERATIONAL SETUP 1 lo loopback carrier unmanaged 2 eth0 ether degraded unmanaged 3 eth1 ether routable configured 4 eth2 ether off unmanaged 5 wan dsa off unmanaged 6 lan1 dsa routable configured 7 lan2 dsa off unmanaged 8 lan3 dsa off unmanaged 8 links listed.

The eth0/lan1 MAC does change... the eth1/eth2 MACs do not change. I've checked over multiple reboot cycles. 3: eth1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000 link/ether 6a:c9:72:53:13:d5 brd ff:ff:ff:ff:ff:ff 4: eth2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000 link/ether 32:b0:1b:6f:4e:e9 brd ff:ff:ff:ff:ff:ff root@ratatosk:~# uptime 20:05:19 up 5 min, 1 user, load average: 0.13, 0.39, 0.21 just rebooted, the MACs are stable 🐴 I could try reconfiguring the netplan config to use the name [like with lan1] instead of MAC... but that will of course mean I can't rename.

Weird. Check logs and verify matching - mac - if not changing random - is usually safest way. I haven't played under network manger lately ... perhaps there lies the problem or network devices gets up with a delay. I will try to recreate during week.

rules in /etc/udev/rules.d did *nothing*. tried with 00 prefix and 99 prefix. udevadm test-builtin net_setup_link didn't mention the filepath either. But I could make it do stuff with /etc/systemd/network/foo.link, which is what's current[ly not deleted yet from previous attempts]. Which makes it even weirder that it still won't rename at boot. Note that the problem is a lot less the renaming and more that networkd refuses to manage the interfaces, even if I give up on renaming,

OS is stable, errors you are experiencing are likely related to faulty HW component as @eselarm already exposed. And we would most likely notice such anomaly before it gets to you. There are thousands of users and this is the only report of this kind ... Boards are going to the market at high rate, while users support remains more or less the same in past 10 years https://docs.armbian.com/User-Guide_Board-Support-Rules/ Community builds and their upgrades are untested as size of Armbian can only cope with a small selection of boards we call as "standard support", where we have volunteers behind. Community builds are on Debian / Arch images level, but packed with more features. https://github.com/armbian/community?tab=readme-ov-file#purpose-of-this-repository For those - Armbian provides build and hosting infrastructure, place to meet - this forum - while everything else is anyone's problem.

Did you try renaming by using udev rule? Example: sudo nano /etc/udev/rules.d/76-rename-wifi-by-mac.rules SUBSYSTEM=="net", ACTION=="add", ATTR{address}=="b8:2d:28:5a:55:92", NAME="wlxb82d285a5592" and then proceed with whatever. Both systemd-networkd and network manager should work.

BPI-R4, but that is probably the least of my problems. setting up netplan, like the docs say I have to. using networkd, or trying... renaming the eth1 interface to ethLAN, keeping lan1 as-is. networkctl always says that eth1 or ethLAN is unmanaged. Note that I'm using the ttyS0 serial console. root@ratatosk:~# networkctl list IDX LINK TYPE OPERATIONAL SETUP 1 lo loopback carrier unmanaged 2 eth0 ether degraded unmanaged 3 eth1 ether off unmanaged 4 eth2 ether off unmanaged 5 wan dsa off unmanaged 6 lan1 dsa routable configured 7 lan2 dsa off unmanaged 8 lan3 dsa off unmanaged root@ratatosk:~# netplan apply [ 141.017704] mtk_soc_eth 15100000.ethernet ethLAN: renamed from eth1 [ 141.068835] mtk_soc_eth 15100000.ethernet ethEXT1: renamed from eth2 root@ratatosk:~# networkctl list IDX LINK TYPE OPERATIONAL SETUP 1 lo loopback carrier unmanaged 2 eth0 ether degraded unmanaged 3 ethLAN ether off unmanaged 4 ethEXT1 ether off unmanaged 5 wan dsa off unmanaged 6 lan1 dsa routable configured 7 lan2 dsa off unmanaged 8 lan3 dsa off unmanaged Here's my config. Note that it does in fact rename the interfaces... albeit not at boot 😑 root@ratatosk:~# cat /etc/netplan/20-ethLAN.yaml network: version: 2 renderer: networkd ethernets: ethLAN: match: macaddress: '6a:c9:72:53:13:d5' #set-name: ethLAN #tried with and without this addresses: [ "192.168.88.253/23" ] emit-lldp: true dhcp4: false dhcp6: false #ignore-carrier: true nameservers: addresses: - 192.168.82.55 search: - tabris.net routes: #- to: 192.168.88.0/23 # metric: 10 # from: 192.168.88.253 # scope: link # type: local - to: "default" via: "192.168.88.254" root@ratatosk:~# cat /etc/netplan/11-lan1.yaml network: version: 2 renderer: networkd ethernets: lan1: match: name: "lan1" addresses: - "192.168.88.250/23" nameservers: addresses: - 192.168.82.55 search: - tabris.net # routes: # - to: "default" # via: "192.168.88.254" note that lan1 works! ethLAN, whether named eth1 or ethLAN... does not. I tried using renderer: NetworkManager It then tries to restart networkd & stop NetworkManager. but it can eventually work... but I have to sacrifice a rubber chicken [stupid manual steps like manually setting the interfaces up with ip link] each time, and they're starting to pile up. it will *not* work at boot no matter what I do.

Aliexpress item: https://pl.aliexpress.com/item/1005006249858881.html Extracted dtb and dts files attached. It boots Warpme/miniarch image for X96 with LPDDR3 RAM https://github.com/warpme/miniarch/releases/download/f3b14fbae071/MiniArch-20240715-6.16.1-board-h313.x96_q_lpddr3-SD-Image.img.xz However internal Ethernet port is silent. It can work through USB->ETH adapter though. Any Idea on how to make Ethernet working? sun50.dtb sun50.dts

Hi. Did you add / enable required overlays ? https://chatgpt.com/share/68a49507-d420-8005-a687-04d8daf0e93a It looks like kernel support is enabled.

@buzz89

I just got an H618 board. It is able to boot using the Vontar image, but the onboard Ethernet is not working. USB Ethernet works fine. The Ethernet chip is RTL8211F, I’m hoping to get some guidance on this issue. Here’s the relevant dmesg output related to Ethernet: dmesg | grep eth [ 0.000000] psci: probing for conduit method from DT. [ 2.422638] dwmac-sun8i 5030000.ethernet: IRQ eth_wake_irq not found [ 2.422669] dwmac-sun8i 5030000.ethernet: IRQ eth_lpi not found [ 2.422678] dwmac-sun8i 5030000.ethernet: IRQ sfty not found [ 2.422989] dwmac-sun8i 5030000.ethernet: supply phy-io not found, using dummy regulator [ 2.423245] dwmac-sun8i 5030000.ethernet: PTP uses main clock [ 2.423316] dwmac-sun8i 5030000.ethernet: Current syscon value is not the default 53fe1 (expect 0) [ 2.426176] dwmac-sun8i 5030000.ethernet: No HW DMA feature register supported [ 2.426204] dwmac-sun8i 5030000.ethernet: RX Checksum Offload Engine supported [ 2.426211] dwmac-sun8i 5030000.ethernet: COE Type 2 [ 2.426220] dwmac-sun8i 5030000.ethernet: TX Checksum insertion supported [ 2.426226] dwmac-sun8i 5030000.ethernet: Normal descriptors [ 2.426232] dwmac-sun8i 5030000.ethernet: Chain mode enabled [ 2.469198] dwmac-sun8i 5030000.ethernet end0: renamed from eth0 [ 3.060680] asix 2-1:1.0 eth0: register 'asix' at usb-5200000.usb-1, ASIX AX88772B USB 2.0 Ethernet, 00:0e:c6:c5:f5:e5 [ 3.068667] asix 2-1:1.0 enx000ec6c5f5e5: renamed from eth0 [ 11.377369] dwmac-sun8i 5030000.ethernet end0: Register MEM_TYPE_PAGE_POOL RxQ-0 [ 11.388786] dwmac-sun8i 5030000.ethernet end0: PHY [stmmac-0:00] driver [Generic PHY] (irq=POLL) [ 11.388838] dwmac-sun8i 5030000.ethernet end0: No Safety Features support found [ 11.388849] dwmac-sun8i 5030000.ethernet end0: No MAC Management Counters available [ 11.388859] dwmac-sun8i 5030000.ethernet end0: PTP not supported by HW [ 11.393925] dwmac-sun8i 5030000.ethernet end0: configuring for phy/rmii link mode