Bridging ethernet on Fedora

Bridging ethernet on Fedora

Post by Nat » Sat, 17 May 2008 02:05:11



I installed Fedora 8 from the Live CD image on an HP DL320 server and
set it up in a datacenter to do mail relay.  Now I need to bridge the
two network cards in the server to act as a switch, while still having
it act in the mail relay role.  For some reason, I can't get this to
work properly.

/etc/sysconfig/network-scripts/ifcfg-eth0 has:
DEVICE=eth0
BOOTPROTO=static
ONBOOT=yes
TYPE=ethernet

/etc/sysconfig/network-scripts/ifcfg-eth1 has:
DEVICE=eth1
BOOTPROTO=static
ONBOOT=yes
TYPE=ethernet

I ran:
yum -install bridge* -y
to install the bridge utilities, then I ran:
brctl addbr br0
brctl addif br0 eth0
brctl addif br0 eth1
ip link set br0 up
ip addr add 10.1.1.52/24 brd + dev br0
route add default gw 10.1.1.1 dev br0

Now I when I check ifconfig, it shows that eth0 has gotten an IP
address in our DHCP range, as well as br0 having the ip address
10.1.1.52.  Also, the computer connected to eth1 cannot ping the
router on eth0.  It gets a message from 10.1.1.52 that the destination
is unreachable.

Any ideas?  Should I be going about this a different way?  Let me know
if I need to post additional information.  I'm still fairly new to
Linux, so you may also need to tell me where to look to find said
information.

 
 
 

Bridging ethernet on Fedora

Post by Philippe Weil » Sat, 17 May 2008 19:40:13



> I installed Fedora 8 from the Live CD image on an HP DL320 server and
> set it up in a datacenter to do mail relay.  Now I need to bridge the
> two network cards in the server to act as a switch, while still having
> it act in the mail relay role.  For some reason, I can't get this to
> work properly.

> /etc/sysconfig/network-scripts/ifcfg-eth0 has:
> DEVICE=eth0
> BOOTPROTO=static
> ONBOOT=yes
> TYPE=ethernet

> /etc/sysconfig/network-scripts/ifcfg-eth1 has:
> DEVICE=eth1
> BOOTPROTO=static
> ONBOOT=yes
> TYPE=ethernet

> I ran:
> yum -install bridge* -y
> to install the bridge utilities, then I ran:
> brctl addbr br0
> brctl addif br0 eth0
> brctl addif br0 eth1
> ip link set br0 up
> ip addr add 10.1.1.52/24 brd + dev br0
> route add default gw 10.1.1.1 dev br0

> Now I when I check ifconfig, it shows that eth0 has gotten an IP
> address in our DHCP range, as well as br0 having the ip address
> 10.1.1.52.  Also, the computer connected to eth1 cannot ping the
> router on eth0.  It gets a message from 10.1.1.52 that the destination
> is unreachable.

> Any ideas?  Should I be going about this a different way?  Let me know
> if I need to post additional information.  I'm still fairly new to
> Linux, so you may also need to tell me where to look to find said
> information.

On centos 5 ( like rhel 5)

I have for this


DEVICE=eth0
MASTER=bond0
SLAVE=yes
ONBOOT=yes
POST_UP_SCRIPT=ifenslave-eth0


DEVICE=eth1
MASTER=bond0
SLAVE=yes
ONBOOT=yes
POST_UP_SCRIPT=ifenslave-eth1


DEVICE=bond0
TYPE=Bonding
BOOTPROTO=none
ONBOOT=yes
IPADDR=
NETMASK=
BROADCAST=
GATEWAY=


eth0      Link encap:Ethernet  HWaddr AA:00:86:9D:B0:5C
           UP BROADCAST RUNNING SLAVE MULTICAST  MTU:1500  Metric:1
           RX packets:1188120484 errors:0 dropped:18763 overruns:0 frame:0
           TX packets:1086428707 errors:0 dropped:0 overruns:0 carrier:0
           collisions:0 txqueuelen:1000
           RX bytes:1301605677778 (1.1 TiB)  TX bytes:1007780187968 (938.5 GiB)
           Base address:0x3000 Memory:d8200000-d8220000


eth1      Link encap:Ethernet  HWaddr AA:00:86:9D:B0:5C
           UP BROADCAST RUNNING SLAVE MULTICAST  MTU:1500  Metric:1
           RX packets:622228703 errors:0 dropped:559 overruns:0 frame:0
           TX packets:1187691040 errors:0 dropped:0 overruns:0 carrier:0
           collisions:0 txqueuelen:1000
           RX bytes:63921012314 (59.5 GiB)  TX bytes:1780803629493 (1.6 TiB)
           Base address:0x3020 Memory:d8220000-d8240000