I am running several docker containers and decided to reorganize them (split the configurations into folders gathering dependent services instead of having one big heap). Everything was going fine until I moved my Home Assistant configuration.
I did not change anything in the previous configuration, except enforcing a network common to all containers visible "externally" (outside of the private docker network). It works fine with the 20+ other containers.
hass:
devices:
- /dev/serial/by-id/usb-1a86_USB_Serial-if00-port0:/dev/serial/by-id/usb-1a86_USB_Serial-if00-port0
image: homeassistant/home-assistant:2022.7
network_mode: host
restart: unless-stopped
volumes:
- /etc/docker/domotique/data/hass:/config
- /etc/localtime:/etc/localtime:ro
depends_on:
- mqtt
networks:
- srv
(...)
networks:
srv:
external: true
name: srv
When starting this container I get an error:
root@srv /e/d/domotique# docker-compose up hass
[+] Running 1/0
⠿ Container domotique-mqtt-1 Running 0.0s
Attaching to domotique-hass-1
Error response from daemon: failed to add interface veth6eb29e9 to sandbox: error setting interface "veth6eb29e9" IP to 172.18.0.33/16: cannot program address 172.18.0.33/16 in sandbox interface because it conflicts with existing route {Ifindex: 7 Dst: 172.18.0.0/16 Src: 172.18.0.1 Gw: <nil> Flags: [] Table: 254}
It disappears when I comment out network_mode: host. The running container has then the IP 172.18.0.33/16. Removing existing Home Assistant containers to force a recreation does not help.
What does this message mean in this context?