All the system
admin would like to avoid server outage by having redundancy for root
filesystem using mirroring, Multiple FC links to SAN with help of
multi-pathing and many more.So here the question is how do you provide
redundancy in network level ? Having a multiple network card will not
give any redundancy.In redhat Linux you need to configure
bonding to accomplish the network level redundancy.Once you have
configured the bonding/teaming by using two NIC cards, kernel will automatically detect
the failure of any NIC and work smartly according to that without any
riot.Bonding can be used for load sharing as well between two physical links.
The below diagram will explain how the bonding is happening.
Operating system used: Redhat Linux 6.3
NIC Details:
NIC Details:
[root@server2
/]# ifconfig -a|grep eth
eth0 Link encap:Ethernet HWaddr 00:0C:29:C0:16:56
eth1 Link encap:Ethernet HWaddr 00:0C:29:C0:16:6A
eth2 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
eth3 Link encap:Ethernet HWaddr 00:0C:29:C0:16:74
|
Goal:
Configure bonding between eth2 and eth3 with name of bond0.
Configure bonding between eth2 and eth3 with name of bond0.
Step 1:
1. Network manager should not be running on the
system, as NIM doesn’t support bonding. So please stop and disable the
networkmanager service.
[root@server2 Desktop]# service NetworkManager status
NetworkManager (pid 2157) is
running...
[root@server2 Desktop]# service NetworkManager stop
Stopping NetworkManager daemon: [
OK ]
[root@server2 Desktop]# service NetworkManager status
NetworkManager is stopped
[root@server2 Desktop]#
|
Step 2:
Add the below line in
/etc/modprobe.conf to load the bonding module in to kernel.
alias bond0
bonding
|
In redhat 6.3, you need to
create new file called “bonding.conf” under /etc/modprobe.d/ with below
mentioned line.
[root@server2
/]# cat > /etc/modprobe.d/bonding.conf
alias bond0
bonding
|
Step 3:
Now time to create a bonding
interface configuration file in /etc/sysconfig/network-scripts/ directory like
the below one.
[root@server2
network-scripts]# pwd
/etc/sysconfig/network-scripts
[root@server2
network-scripts]# cat ifcfg-bond0
#This is
congiguration file for bond0.Used NIC's eth2 & eth3
DEVICE=bond0
IPADDR=192.168.10.25
NETMASK=255.255.255.0
USRCTL=no
ONBOOT=yes
BOOTPRO=none
BONDING_OPTS="mode=0
miimon=100"
|
Step:3
Create a configuration files
under “/etc/sysconfig/network-scripts/” for network interfaces if not exists.If
exists,have a contents like the below one.
[root@server2
network-scripts]# cat ifcfg-eth2
DEVICE=eth2
HWADDR=00:0C:29:C0:16:60
BOOTPRO=none
ONBOOT=yes
MASTER=bond0
SLAVE=yes
USERCTL=no
[root@server2
network-scripts]# cat ifcfg-eth3
DEVICE=eth3
HWADDR=00:0C:29:C0:16:74
BOOTPRO=none
ONBOOT=yes
MASTER=bond0
SLAVE=yes
USERCTL=no
|
Note: Do not copy paste the content
from above output.MAC and DEVICE name will differ for each system.
Step:4
Now restart the network service
to load the configuration.
Note:Do not restart the
network service without server maintenance window.
[root@server2
network-scripts]# service network restart
Shutting down
interface eth2:
/etc/sysconfig/network-scripts/ifdown-eth: line 121:
/sys/class/net/bond0/bonding/slaves: No such file or directory
[ OK ]
Shutting down
interface eth3:
/etc/sysconfig/network-scripts/ifdown-eth: line 121:
/sys/class/net/bond0/bonding/slaves: No such file or directory
[ OK ]
Shutting down
loopback interface:
[ OK ]
Bringing up
loopback interface:
[ OK ]
Bringing up
interface bond0: Determining if ip
address 192.168.10.25 is already in use for device bond0...
[ OK ]
[root@server2
network-scripts]#
|
Step:5
Verify whether “bond0” has come
up with IP or not .
[root@server2
/]# ifconfig -a
bond0 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
inet addr:192.168.10.25 Bcast:192.168.10.255 Mask:255.255.255.0
inet6 addr:
fe80::20c:29ff:fec0:1660/64 Scope:Link
UP BROADCAST RUNNING MASTER MULTICAST MTU:1500
Metric:1
RX packets:142 errors:0 dropped:0
overruns:0 frame:0
TX packets:18 errors:0 dropped:0
overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:8556 (8.3 KiB) TX bytes:1236 (1.2 KiB)
eth2 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
inet6 addr: fe80::20c:29ff:fec0:1660/64
Scope:Link
UP BROADCAST RUNNING SLAVE
MULTICAST MTU:1500 Metric:1
RX packets:1395 errors:0 dropped:0
overruns:0 frame:0
TX packets:19 errors:0 dropped:0
overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:86730 (84.6 KiB) TX bytes:1898 (1.8 KiB)
Interrupt:16 Base address:0x2080
eth3 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
inet6 addr:
fe80::20c:29ff:fec0:1660/64 Scope:Link
UP BROADCAST RUNNING SLAVE
MULTICAST MTU:1500 Metric:1
RX packets:1390 errors:0 dropped:0
overruns:0 frame:0
TX packets:15 errors:0 dropped:0
overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:86384 (84.3 KiB) TX bytes:1578 (1.5 KiB)
Interrupt:18 Base address:0x2480
[root@server2
/]#
|
In the above output,you can see
NIC eth2 and eth3 have flag “SLAVE” and interface “bond0” has flag
MASTER.Another thing you note both, the NIC interface will show same MAC
address.
Step:6
Performing the live test to
ensure bonding is providing the fault tolerance.
First i am removing the LAN
cable from eth3 and let see what happens.
[root@server2
/]# ifconfig -a
bond0 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
inet addr:192.168.10.25 Bcast:192.168.10.255 Mask:255.255.255.0
inet6 addr:
fe80::20c:29ff:fec0:1660/64 Scope:Link
UP BROADCAST RUNNING MASTER
MULTICAST MTU:1500 Metric:1
RX packets:5012 errors:0 dropped:0
overruns:0 frame:0
TX packets:39 errors:0 dropped:0
overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:312100 (304.7 KiB) TX bytes:2682 (2.6 KiB)
eth2 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
inet6 addr: fe80::20c:29ff:fec0:1660/64
Scope:Link
UP BROADCAST RUNNING SLAVE MULTICAST
MTU:1500 Metric:1
RX packets:3830 errors:0 dropped:0
overruns:0 frame:0
TX packets:29 errors:0 dropped:0
overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:238520 (232.9 KiB) TX bytes:2534 (2.4 KiB)
Interrupt:16 Base address:0x2080
eth3 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
BROADCAST SLAVE MULTICAST MTU:1500
Metric:1
RX packets:3825 errors:0 dropped:0
overruns:0 frame:0
TX packets:26 errors:0 dropped:0
overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:238138 (232.5 KiB) TX bytes:2388 (2.3 KiB)
Interrupt:18 Base address:0x2480
[root@server2
/]#
|
Still the bond0 interface is UP
and RUNNING fine.At the same time,”RUNNING” flag has disappear from eth3.
Now i have connected LAN cable back to eth3 and pulling out from eth2.
Now i have connected LAN cable back to eth3 and pulling out from eth2.
[root@server2
/]# ifconfig -a
bond0 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
inet addr:192.168.10.25 Bcast:192.168.10.255 Mask:255.255.255.0
inet6 addr:
fe80::20c:29ff:fec0:1660/64 Scope:Link
UP BROADCAST RUNNING MASTER
MULTICAST MTU:1500 Metric:1
RX packets:5115 errors:0 dropped:0
overruns:0 frame:0
TX packets:39 errors:0 dropped:0
overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:318376 (310.9 KiB) TX bytes:2682 (2.6 KiB)
eth2 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
BROADCAST SLAVE MULTICAST MTU:1500
Metric:1
RX packets:3926 errors:0 dropped:0
overruns:0 frame:0
TX packets:29 errors:0 dropped:0
overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:244376 (238.6 KiB) TX bytes:2534 (2.4 KiB)
Interrupt:16 Base address:0x2080
eth3 Link encap:Ethernet HWaddr 00:0C:29:C0:16:60
UP BROADCAST RUNNING SLAVE
MULTICAST MTU:1500 Metric:1
RX packets:3832 errors:0 dropped:0
overruns:0 frame:0
TX packets:26 errors:0 dropped:0
overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:238558 (232.9 KiB) TX bytes:2388 (2.3 KiB)
Interrupt:18 Base address:0x2480
|
Still “bond0” interface running
with UP & RUNNING flag.So you have successfully configured
bonding on Redhat Linux 6.
To see complete bonding info,use below command.
[root@server2
/]# cat /proc/net/bonding/bond0
Ethernet
Channel Bonding Driver: v3.7.1 (April 27, 2011)
Bonding Mode:
load balancing (round-robin)
MII Status:
up
MII Polling
Interval (ms): 100
Up Delay
(ms): 0
Down Delay
(ms): 0
Slave
Interface: eth2
MII Status:
up
Speed:
Unknown
Duplex:
Unknown
Link Failure
Count: 1
Permanent HW
addr: 00:0c:29:c0:16:60
Slave queue
ID: 0
Slave
Interface: eth3
MII Status:
up
Speed:
Unknown
Duplex:
Unknown
Link Failure
Count: 1
Permanent HW
addr: 00:0c:29:c0:16:74
Slave queue
ID: 0
|
To verify the
current bonding mode,use below command.
[root@server2
/]# cat /sys/class/net/bond0/bonding/mode
balance-rr 0
|
You can modify the bonding mode
by editing “mode” in the ifcfg-bond0 configuration file.
[root@server2
/]# cat /etc/sysconfig/network-scripts/ifcfg-bond0 |grep -i mode
BONDING_OPTS="mode=0
miimon=100"
|
Policy
Name
|
Code
|
Description
|
balance-rr
|
0
|
Round-Robin policy for fault tolerance
|
active-backup
|
1
|
Active-Backup policy for fault tolerance
|
balance-xor
|
2
|
Exclusive-OR policy for fault tolerance
|
broadcast
|
3
|
All tansmissions are sent on all slave
interfaces
|
802.3ad
|
4
|
Dynamic Link Aggregation Policy
|
balance-tlb
|
5
|
Transmit Load Balancing Policy for fault
tolerance
|
balance-alb
|
6
|
Active load Balancing Policy for fault
tolerance
|
To list the currently configured bonds,
[root@server2
/]# cat /sys/class/net/bonding_masters
bond0
|
Thank you for reading.
No comments:
Post a Comment