Debian 13 template v13.1-2 and systemd-networkd-wait-online timeouts

[vmguy]

I have systemd-networkd-wait-online disabled but I still having those timeout messages.

Oct 02 06:27:30 deb systemd[1]: Starting apt-daily-upgrade.service - Daily apt upgrade and clean activities...
Oct 02 06:28:00 deb systemd-networkd-wait-online[1563]: Timeout occurred while waiting for network connectivity.
Oct 02 06:28:00 deb apt-helper[1561]: E: Sub-process /lib/systemd/systemd-networkd-wait-online returned an error code (1)
Oct 02 06:28:01 deb systemd[1]: apt-daily-upgrade.service: Deactivated successfully.
Oct 02 06:28:01 deb systemd[1]: Finished apt-daily-upgrade.service - Daily apt upgrade and clean activities.

systemctl status systemd-networkd-wait-online
○ systemd-networkd-wait-online.service - Wait for Network to be Configured
     Loaded: loaded (/usr/lib/systemd/system/systemd-networkd-wait-online.service; disabled; preset: disabled)
     Active: inactive (dead)
       Docs: man:systemd-networkd-wait-online.service(8)

 

[leesteken]

Post in thread 'Network Wait Error'

Mar 11, 2024

Hi,
AFAICT, this service is not enabled by default and Proxmox VE does not use systemd-networkd.service for configuring the network. It uses /etc/network/interfaces and ifupdown2. Did you install anything that requires systemd-network? Otherwise, you should be able to just disable these services. The system logs/journal might also contain further information.

• fiona

 

[reinob]

I think the problem is that the apt-daily-upgrade.service does this:

ExecStartPre=-/usr/lib/apt/apt-helper wait-online
ExecStart=/usr/lib/apt/apt.systemd.daily install

The apt-helper wait-online probably calls systemd-networkd-wait-online, which returns an error.
No idea (at least right now) how to fix that.

 

[blinkinglightsclub]

Ran into the same thing on Debian 13 containers and tracked it down. Picking up from the apt-helper finding above:

The container has two network stacks. ifupdown actually brings up eth0 (from /etc/network/interfaces), but systemd-networkd is also running and manages nothing. networkctl list shows every link as unmanaged. apt's timers run apt-helper wait-online, and when networkd is active that hands off to systemd-networkd-wait-online. That waits for networkd to report a managed link as online, which never happens, so it times out every run.

That's also why disabling the service alone doesn't stick: systemd-networkd.socket can start networkd again. Masking all three fixes it, but only if networkctl list shows every link as unmanaged:

systemctl mask --now systemd-networkd.service systemd-networkd.socket systemd-networkd-wait-online.service

Afterwards eth0's address and default route are unchanged, and /usr/lib/apt/apt-helper wait-online returns immediately with exit 0. To undo: systemctl unmask the same three units, then systemctl start systemd-networkd.

I wrote it up in more detail, including a guarded script for doing it across many containers (it rolls back if a container's address or route changes):
https://blinkinglightsclub.hashnode.dev/proxmox-lxc-apt-wait-online-timeout