porg

Was able to omv-release-upgrade my Odroid HC2 from my restored OMV6 to OMV7 simply by having the single line in /etc/apt/sources.list.d/armbian.list commented out. Now I will just need to figure out how to update Armbian itself.

In the OpenMediaVault forums I reported Seemingly cannot update and upgrade anymore due to repository having no release file (OMV6 on Armbian on Odroid HC1) So if my OMV 6 with Armbian Bullseye both have not updated for a while, will this your instruction let me run apt update and apt upgrade again and also omv-upgrade? Note: As omv-upgrade in OMV 6 did not work anymore, I thought to omv-release-upgrade to OMV 7 right away. But that upgrade script included an implicit omv-upgrade too. So failed too. Hence I had then run an omv-release-upgrade (with its omv-upgrade line outcommented) which resulted in a broken system: Armbian seemingly up to date, but all OMV stuff gone or stuck in v6. I restored my SD-card with OMV6/Armbian-Bullseye from my openmediavault-backup with fsarchiver successfully. Now I would like to do things correct. Hence my questions regarding your steps: Ad 1) a) From where shall I get that armbian.gpg file reliably? Is it cryptographically ok/safe enough as @BrewNinja did it? Note: Hobby NAS on a private LAN with no services exposed to the Internet, LAN only services) If ok, then please integrate into your instruction for TL;DR users. Thanks! b) /etc/apt/sources.list is empty Ad 3) That part is hard to understand for someone who deals with SBC / Armbian only occasionally. How do I ensure apt to not make any changes to /etc/initramfs-tools/initramfs.conf ? Please clarify that. Do you mean there are interactive questions which I shall answer accordingly? Or do I need to lock something? Or monitor something and revert it if changed? Ad "4")? Then comes another code block with "apt upgrade --no-new-pkgs ; apt full-upgrade ; apt dist-upgrade" — Is this step 4? Or preceding part 3?

Possibly this solves it? https://forum.armbian.com/topic/35402-upgrading-armbian-from-bullseye-to-bookworm

Thanks for the fix!

Environment: Armbian on an Odroid-HC2 (Homecloud 2) which uses the odroid-xu4 firmware. Reproduction: After initial installation on the welcome banner "RX today:" always shows a nice value. Install and run some docker containers. This changes /sys/class/net/ . AFAIK it adds so called "virtual network interfaces". From then on the Armbian welcome screen displays like this: Error: Interface "veth•••••••" not found in database. Note: • redacted by me. Analysis: /etc/default/armbian-motd … PRIMARY_INTERFACE="$(ls -1 /sys/class/net/ | grep -v lo | egrep "enp|eth" | head -1)" … The current regular expression falsely matches veth••••••• (created by Docker). $ ls -1 /sys/class/net/ br-•••••••••••• docker0 enx•••••••••••• lo veth••••••• I fixed it for myself by changing PRIMARY_INTERFACE to this: "$(ls -1 /sys/class/net/ | grep -v lo | egrep "^en" | head -1)" Fix Proposal: A developer with proper knowhow of the network interface naming conventions changes the default regular expression / pipe for PRIMARY_INTERFACE so that it safely only matches a real hardware interface for all typical systems, also working fine when using docker or other software which may create virtual network interfaces or other stub like entries there. Notes: This bug was already discussed in a forum thread in a "how to help yourself" manner. This here now is a formal bug report with the inquiry to please fix it. I ran through the bug submission wizard at https://www.armbian.com/bugs/ which eventually led me here.

I run Armbian on an Odroid-HC2 (Homecloud 2) which uses the odroid-xu4 firmware. On the welcome banner RX today: always showed a nice value. Since December 2022 when I last updated it suddenly showed Error: Interface "veth•••••••" not found in database. And I can confirm: I use 2 docker images on that device, so that may cause unusual devices at /sys/class/net/ which the motd script handles wrongly. Note: • redacted by me. /etc/default/armbian-motd […] PRIMARY_INTERFACE="$(ls -1 /sys/class/net/ | grep -v lo | egrep "enp|eth" | head -1)" […] $ ls -1 /sys/class/net/ br-•••••••••••• docker0 enx•••••••••••• lo veth••••••• Simple fix: Change PRIMARY_INTERFACE to the grep expression "First device that starts with "en", in case you have only one physical Ethernet connection. PRIMARY_INTERFACE="$(ls -1 /sys/class/net/ | grep -v lo | egrep "^en" | head -1)" After that "RX today" on the welcome banner again shows some meaningful statistics.