Docker won't start due to existing bridge

Support
1

I’ve just installed Rockstor on a repurposed machine, and it suffered a random reboot. I’ll be testing the RAM, etc., but now I can’t start Docker due to an old instance of the docker0 bridge.

dockerd[5941]: failed to start daemon: Error initializing network controller: error creating default "bridge" network: cannot create network 738d60b2edf39c960034579b6bb5396e66fe446558773be494ee6fd3dd39238b (docker0): conflicts with network 5238a1ddcf98fe0ca223df4ae429dd7d51f575dcade3ac3fde5361e4202700ed (docker0): networks have same bridge name

nmcli shows the existing docker0 bridge, even after a reboot, so I tried to remove it:

nmcli connection delete docker0
Connection 'docker0' (3472d784-8a34-48dc-87f6-9340ef12fce5) successfully deleted.
nmcli device delete docker0
Device 'docker0' successfully removed.

Docker still can’t start (same error). Various guides online suggest removing the contents of /var/lib/docker/network but /var/lib/docker is empty on my installation.

I’m running Rockstor 5.0.15 on Leap 15.6.

Anyone have any ideas?

2

Very brief idea: most of the Docker data is stored in your rockons root, so maybe you need to delete your rockons root and then try recreating it, and reconfiguring the rockons service using that freshly and newly created rockons root?

Sorry for the very brief answer.

1 Like
3

I’d not clocked that everything would be in /mnt2/rockons rather than /var/lib/docker! Found the networking files, but I’ve done more diagnosis since then and have reinstalled the machine.

2 Likes