Thank god, a static IP (for unraid) fixed the problem, know my docker-containers get loaded after a reboot with br0... 

Edited April 26, 20206 yr by derMeister

 

On 1/23/2018 at 12:59 PM, bonienl said:

 

Do the following to let docker rebuild the networks

rm /var/lib/docker/network/files/local-kv.db
/etc/rc.d/rc.docker restart

 

I try to run this command.  I can remove the local-kv.db but I cannot restart rc.docker.  When I do an ls-a to that location, I do not have rc.docker.  I am running into the same issue where I cannot select br0 for network device. 

 

I had the same issue.  

 

here's how I fixed it:

 

Restart the Unraid, 

dont initialize the array just yet....

go to SETTINGS, Network Settings,  change the desired MTU: to 1500( you have to put in 1500, what shows in the box is just a place-holder and not the actual number 1500)

Click on APPLY. 

you will lose network connectivity, so go to the server and restart it there. 

On 11/6/2020 at 4:14 AM, snizzard said:

I had the same issue.  

 

here's how I fixed it:

 

Restart the Unraid, 

dont initialize the array just yet....

go to SETTINGS, Network Settings,  change the desired MTU: to 1500( you have to put in 1500, what shows in the box is just a place-holder and not the actual number 1500)

Click on APPLY. 

you will lose network connectivity, so go to the server and restart it there. 

My VMs lost my br0 in update 6.9.0 to 6.9.1 and setting the MTU & restart did the trick.

unraid Cannot get interface MTU on 'br0': No such device

 

After tearing my hair and one sleepless night i can say that if this happens to anybody else and nothing mentioned in the thread works. Turn off bonding in network settings. For some reason it was enabled and set bridging option into a fritz (br0 was nowhere to be found) when i swapped motherboard. So i shut down my VM, Docker, disabled bonding and got an option for what eth0 to bridge to. 

On 1/24/2018 at 4:59 AM, bonienl said:

 

Do the following to let docker rebuild the networks

rm /var/lib/docker/network/files/local-kv.db
/etc/rc.d/rc.docker restart

 

Thanks mate. An old post, but helped me out with docker network issue.

On 1/24/2018 at 12:59 AM, bonienl said:

 

Do the following to let docker rebuild the networks

rm /var/lib/docker/network/files/local-kv.db
/etc/rc.d/rc.docker restart

 

Thanks for this solution. It works.
After upgrading Unraid to the latest version, the first boot after upgrading was ok, but I need to reboot it because my UPS battery needed replacement. Unraid docker definition menu can no longer see my br0. After performing the above two commands, the br0 is back and those dockers are now working with just restarting them again.

Not sure if this will help some - but after running 

rm /var/lib/docker/network/files/local-kv.db
/etc/rc.d/rc.docker restart

I was missing my custom docker network, no problem, I re-added it, but none of the containers automatically reconnected. I noticed I could go in and manually adjust the template and reselect the new custom network and the container would start fine...but this would take way too long to go through each one.

I ChatGPT made a script that will get list of all containers. Will check network of each container and if using the defined network, it will disconnect it and reconnect it the new one(of same name), skipping any containers that use some other network (br0, bridge, host...)

#!/bin/bash

# Network name
network_name="proxynet"

# List all containers
containers=$(docker ps -a --format '{{.Names}}')

for container in $containers; do
    # Retrieve the container's network information
    network_info=$(docker inspect --format='{{range $key, $value := .NetworkSettings.Networks}}{{if eq $key "'$network_name'"}}true{{end}}{{end}}' $container)

    # Check if the container is connected to the specified network
    if [ "$network_info" = "true" ]; then
        echo "Updating network for $container"

        # Disconnect from the old network (ignore errors if not connected)
        docker network disconnect --force $network_name $container 2>/dev/null
        
        # Connect to the new network
        docker network connect $network_name $container
        
        # Restart the container to apply changes
        docker restart $container
    else
        echo "$container is not connected to $network_name, skipping..."
    fi
done

 

Edited December 12, 20232 yr by khaotiq

Hello from 2024.

I rebooted my server today and I was seeing the

 Error response from daemon: network br0 not found Error: failed to start containers: <container name>

error like OP was.

 

Somehow I didn't have a static IP set in my Unraid OS settings. So going Settings > Network Settings and changing my IPv4 to a static address worked. To change the setting I had to turn off the Docker service, so when I turned on the Docker service again all of my containers came back up immediately. No need to run any commands for me, apparently.