Correct way to configure Bridge network in centos

Question

I am trying to configure bridge network in the HP server. I dont have console access. I am connected using ssh to the server remotely. The configuration files have been placed in the proper location /etc/sysconfig/network-scripts

cat ifcfg-eth0 
DEVICE=eth0
ONBOOT=yes
TYPE=Ethernet
IPV6INIT=no
USERCTL=no
BRIDGE=br0

cat ifcfg-br0
DEVICE=br0
TYPE=Bridge
BOOTPROTO=static
IPADDR=IP of the server
NETMASK=255.255.255.0
ONBOOT=yes

When I restart the network service after configuring this /etc/init.d/network restart, I am getting disconnected from the terminal. What is the proper way to do this configuration?

The above configuration works perfectly on the virtualbox setup, but when I try to implement the same to a physical server, it fails to set up the bridge.

I am using Centos 6.8 OS.

Answer 1

I finally figured out on this.

During the bridge configuration,

• For bridge interface (br0) you need to mention the actual Hardware Address, disable NetworkManager control and enable the onboot.
  
• For ethernet interface (eth0), you need to mention a different Hardware Address (last digit can be changed), disable NetworkManager control and enable the onboot.
  

Example configuration

ifcfg-br0
=========
DEVICE=br0
TYPE=Bridge
GATEWAY=192.168.1.1
HWADRESS=00:1C:23:59:5A:92
IPADDR=192.168.1.109
#NETMASK=255.255.255.0
ONBOOT=yes
#BOOTPROTO=dhcp
NM_CONTROLLED=no
DELAY=0

ifcfg-eth0
==========
DEVICE=eth0
TYPE=Ethernet
HWADRESS=00:1C:23:59:5A:93
BOOTPROTO=dhcp
ONBOOT=yes
NM_CONTROLLED=no
BRIDGE=br0

you can create a shell script and run it in nohup mode.

$ echo  "service network restart" > restart_network.sh
$ chmod u+x restart_network.sh
$ nohup ./restart_network.sh &

I was able to configure the bridge network on multiple physical servers without the network getting disconnected. I did not have console or physical access to server, but used only remote connection via ssh.