Hi
MAAS version: 3.1.0
We have a large working cluster, however we deployed today for the first time in a month and network config failed to come up properly, looking in the logs compared to a machine with the same config booted one month before:
Original working:
Cloud-init v. 21.4-0ubuntu1~20.04.1
2022-03-20 12:19:13,313 - activators.py[DEBUG]: Attempting command ['netplan', 'apply'] for device all
2022-03-20 12:19:13,313 - subp.py[DEBUG]: Running command ['netplan', 'apply'] with allowed return codes [0] (shell=False, capture=True)
2022-03-20 12:19:13,923 - util.py[DEBUG]: Writing to /etc/netplan/50-maas.yaml - wb: [644] 1257 bytes
2022-03-20 12:19:13,923 - util.py[DEBUG]: Changing the ownership of /etc/netplan/50-maas.yaml to 0:0
And today:
Cloud-init v. 22.1-14-g2e17a0d6-0ubuntu1~20.04.3
2022-04-20 11:04:44,026 - activators.py[DEBUG]: Attempting command ['netplan', 'apply'] for device all
2022-04-20 11:04:44,026 - subp.py[DEBUG]: Running command ['netplan', 'apply'] with allowed return codes [0] (shell=False, capture=True)
2022-04-20 11:04:44,450 - activators.py[WARNING]: Running ['netplan', 'apply'] resulted in stderr output: Failed to connect system bus: No such file or directory
Falling back to a hard restart of systemd-networkd.service
2022-04-20 11:04:44,746 - util.py[DEBUG]: Writing to /etc/netplan/50-maas.yaml - wb: [644] 1257 bytes
2022-04-20 11:04:44,746 - util.py[DEBUG]: Changing the ownership of /etc/netplan/50-maas.yaml to 0:0
This is causing issues as it’s not bringing up the ips on the interfaces nor the routes. It looks like the latest image from maas.io is the problem.
I’m trying to drill down into what changed on the image to cause this.