Custom Docker bridge configuration¶

Custom Docker bridge configuration involves creating and managing a user-defined network bridge to facilitate container communication across different hosts or specific network segments, often replacing the default docker0 bridge.^[600-developer-docker-docker-bridge.md]

This approach allows containers to reside on the same network segment as the host or specific physical interfaces, rather than being isolated behind NAT.^[600-developer-docker-docker-bridge.md]

Overview¶

In a standard setup, Docker creates a default bridge named docker0. However, for more complex networking requirements—such as connecting containers across multiple physical hosts or assigning IPs from the local subnet—administrators can create a custom bridge device (e.g., br1) and configure the Docker daemon to use it.^[600-developer-docker-docker-bridge.md]

Because all network interfaces connected via a bridge must exist in the same network segment, it is necessary to explicitly restrict the IP address range allocated to containers on each host to prevent conflicts.^[600-developer-docker-docker-bridge.md]

Configuration Steps¶

1. Create and configure the virtual bridge¶

First, install the necessary utilities and create the bridge interface.^[600-developer-docker-docker-bridge.md]

# Install utilities
yum install bridge-utils

# Create the bridge (e.g., br1)
brctl addbr br1

# Assign an IP address to the bridge in the same subnet as your network
# Example for Host1 (192.168.2.1)
ifconfig br1 192.168.2.1 netmask 255.255.255.0

2. Bridge the physical network interface¶

Connect the physical network interface (e.g., eth0) to the new virtual bridge.^[600-developer-docker-docker-bridge.md]

brctl addif br1 eth0

3. Configure the Docker daemon¶

Modify the Docker daemon configuration (typically /etc/sysconfig/docker on older systems) to use the new bridge and define a specific IP range for containers using the --fixed-cidr flag.^[600-developer-docker-docker-bridge.md]

Example configuration for Host1:

DOCKER_OPTS="-b=br1 --fixed-cidr='192.168.2.64/26' "

Example configuration for Host2:

DOCKER_OPTS="-b=br1 --fixed-cidr='192.168.2.128/26' "

-b=br1: Specifies the bridge device to be used.
--fixed-cidr: Limits the IP allocation range for containers on that specific host to avoid overlaps.

4. Restart and verify¶

After saving the configuration, restart the Docker service and launch a container to verify connectivity.^[600-developer-docker-docker-bridge.md]

systemctl restart docker
docker run -it <image_name>
ping <target_container_ip>

Bridge management commands¶

The brctl utility is used to manage bridge devices.^[600-developer-docker-docker-bridge.md]

Command	Description
`brctl addbr <name>`	Creates a new bridge device (e.g., `brctl addbr br0`).
`brctl addif <bridge> <device>`	Adds a physical interface to the bridge (e.g., `brctl addif br0 eth0`).
`brctl delif <bridge> <device>`	Removes a physical interface from the bridge.
`brctl delbr <name>`	Deletes a bridge device.
`brctl show`	Displays the current bridge configuration and status.

Static network configuration¶

To make the bridge configuration persistent across reboots, network scripts (e.g., in /etc/sysconfig/network-scripts/) should be created for both the bridge (ifcfg-br0) and the physical interface.^[600-developer-docker-docker-bridge.md]

Key configuration parameters for the physical interface: * BRIDGE=br0: This directive is required to attach the physical device to the bridge.

[[Docker Networking]]
[[Network namespaces]]

Sources¶

^[600-developer-docker-docker-bridge.md]