Ethernet Parameters

Using Multiple Ethernet Cards

It is possible to use multiple Ethernet cards on a single machine. For each card there must be an alias and, possibly, options lines for each card in /etc/modprobe.conf.

For additional information about using multiple Ethernet cards, refer to the Linux Ethernet-HOWTO online at http://www.redhat.com/mirrors/LDP/HOWTO/Ethernet-HOWTO.html.

The Channel Bonding Module

Community Enterprise Linux allows administrators to bind NICs together into a single channel using the bonding kernel module and a special network interface, called a channel bonding interface. Channel bonding enables two or more network interfaces to act as one, simultaneously increasing the bandwidth and providing redundancy.

To channel bond multiple network interfaces, the administrator must perform the following steps:

alias bond<N> bonding

Replace <N> with the interface number, such as 0. For each configured channel bonding interface, there must be a corresponding entry in /etc/modprobe.conf.

• Configure a channel bonding interface as outlined in "Channel Bonding Interfaces".
• To enhance performance, adjust available module options to ascertain what combination works best. Pay particular attention to the miimon or arp_interval and the arp_ip_target parameters. Refer to "bonding Module Directives" for a list of available options and how to quickly determine the best ones for your bonded interface.

bonding Module Directives

It is a good idea to test which channel bonding module parameters work best for your bonded interfaces before adding them to the BONDING_OPTS="<bonding parameters>" directive in your bonding interface configuration file (ifcfg-bond0 for example). Parameters to bonded interfaces can be configured without unloading (and reloading) the bonding module by manipulating files in the sysfs file system.

sysfs is a virtual file system that represents kernel objects as directories, files and symbolic links. sysfs can be used to query for information about kernel objects, and can also manipulate those objects through the use of normal file system commands. The sysfs virtual file system has a line in /etc/fstab, and is mounted under /sys. All bonded interfaces can be configured dynamically by interacting with and manipulating files under the /sys/class/net/ directory.

After you have created a channel bonding interface file such as ifcfg-bond0 and inserted SLAVE=yes and MASTER=bond0 directives in the bonded interfaces following the instructions in "Channel Bonding Interfaces", you can proceed to testing and determining the best parameters for your bonded interface.

First, bring up the bond you created by running ifconfig bond<N> up as root:

ifconfig bond0 up

If you have correctly created the ifcfg-bond0 bonding interface file, you will be able to see bond0 listed in the output of running ifconfig (without any options):

~]# ifconfig
bond0     Link encap:Ethernet  HWaddr 00:00:00:00:00:00
          UP BROADCAST RUNNING MASTER MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:0 (0.0 b)  TX bytes:0 (0.0 b)
eth0      Link encap:Ethernet  HWaddr 52:54:00:26:9E:F1
          inet addr:192.168.122.251  Bcast:192.168.122.255  Mask:255.255.255.0
          inet6 addr: fe80::5054:ff:fe26:9ef1/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:207 errors:0 dropped:0 overruns:0 frame:0
          TX packets:205 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:70374 (68.7 KiB)  TX bytes:25298 (24.7 KiB)
[output truncated]

To view all existing bonds, even if they are not up, run:

~]# cat /sys/class/net/bonding_masters
bond0

You can configure each bond individually by manipulating the files located in the /sys/class/net/bond<N>/bonding/ directory. First, the bond you are configuring must be taken down:

ifconfig bond0 down

As an example, to enable MII monitoring on bond0 with a 1 second interval, you could run (as root):

echo 1000 > /sys/class/net/bond0/bonding/miimon

To configure bond0 for balance-alb mode, you could run either:

echo 6 > /sys/class/net/bond0/bonding/mode

...or, using the name of the mode:

echo balance-alb > /sys/class/net/bond0/bonding/mode

After configuring some options for the bond in question, you can bring it up and test it by running ifconfig bond<N> up . If you decide to change the options, take the interface down, modify its parameters using sysfs, bring it back up, and re-test.

Once you have determined the best set of parameters for your bond, add those parameters as a space-separated list to the BONDING_OPTS= directive of the /etc/sysconfig/network-scripts/ifcfg-bond<N> file for the bonded interface you are configuring. Whenever that bond is brought up (for example, by the system during the boot sequence if the ONBOOT=yes directive is set), the bonding options specified in the BONDING_OPTS will take effect for that bond. For more information on configuring bonded interfaces (and BONDING_OPTS), refer to "Channel Bonding Interfaces".

The following is a list of available channel bonding module parameters for the bonding module. For more in-depth information on configuring channel bonding and the exhaustive list of bonding module parameters, install the kernel-doc package and then locating and opening the included bonding.txt file:

yum -y install kernel-doc
nano -w $(rpm -ql kernel-doc | grep bonding.txt)

Bonding Interface Parameters

arp_interval=<time_in_milliseconds>

Specifies (in milliseconds) how often ARP monitoring occurs.

It is essential that both arp_interval and arp_ip_target parameters are specified, or, alternatively, the miimon parameter is specified. Failure to do so can cause degradation of network performance in the event that a link fails.

If using this setting while in mode=0 or mode=1 (the two load-balancing modes), the network switch must be configured to distribute packets evenly across the NICs. For more information on how to accomplish this, refer to /usr/share/doc/kernel-doc-<kernel_version>/Documentation/networking/bonding.txt

The value is set to 0 by default, which disables it.

arp_ip_target=<ip_address> [,<ip_address_2>,...<ip_address_16> ]

Specifies the target IP address of ARP requests when the arp_interval parameter is enabled. Up to 16 IP addresses can be specified in a comma separated list.

arp_validate=<value>

Validate source/distribution of ARP probes; default is none. Other valid values are active, backup, and all.

debug=<number>

Enables debug messages. Possible values are:

• 0 - Debug messages are disabled. This is the default.

• 1 - Debug messages are enabled.

downdelay=<time_in_milliseconds>

Specifies (in milliseconds) how long to wait after link failure before disabling the link. The value must be a multiple of the value specified in the miimon parameter. The value is set to 0 by default, which disables it.

lacp_rate=<value>

Specifies the rate at which link partners should transmit LACPDU packets in 802.3ad mode. Possible values are:

• slow or 0 - Default setting. This specifies that partners should transmit LACPDUs every 30 seconds.

• fast or 1 - Specifies that partners should transmit LACPDUs every 1 second.

miimon=<time_in_milliseconds>

Specifies (in milliseconds) how often MII link monitoring occurs. This is useful if high availability is required because MII is used to verify that the NIC is active. To verify that the driver for a particular NIC supports the MII tool, type the following command as root:

ethtool <interface_name> | grep "Link detected:"

In this command, replace <interface_name> with the name of the device interface, such as eth0, not the bond interface. If MII is supported, the command returns:

Link detected: yes

If using a bonded interface for high availability, the module for each NIC must support MII. Setting the value to 0 (the default), turns this feature off. When configuring this setting, a good starting point for this parameter is 100.

It is essential that both arp_interval and arp_ip_target parameters are specified, or, alternatively, the miimon parameter is specified. Failure to do so can cause degradation of network performance in the event that a link fails.

mode=<value>

...where <value> is one of:

• balance-rr or 0 - Sets a round-robin policy for fault tolerance and load balancing. Transmissions are received and sent out sequentially on each bonded slave interface beginning with the first one available.

• active-backup or 1 - Sets an active-backup policy for fault tolerance. Transmissions are received and sent out via the first available bonded slave interface. Another bonded slave interface is only used if the active bonded slave interface fails.
• balance-xor or 2 - Sets an XOR (exclusive-or) policy for fault tolerance and load balancing. Using this method, the interface matches up the incoming request's MAC address with the MAC address for one of the slave NICs. Once this link is established, transmissions are sent out sequentially beginning with the first available interface.
• broadcast or 3 - Sets a broadcast policy for fault tolerance. All transmissions are sent on all slave interfaces.
• 802.3ad or 4 - Sets an IEEE 802.3ad dynamic link aggregation policy. Creates aggregation groups that share the same speed and duplex settings. Transmits and receives on all slaves in the active aggregator. Requires a switch that is 802.3ad compliant.
• balance-tlb or 5 - Sets a Transmit Load Balancing (TLB) policy for fault tolerance and load balancing. The outgoing traffic is distributed according to the current load on each slave interface. Incoming traffic is received by the current slave. If the receiving slave fails, another slave takes over the MAC address of the failed slave.
• balance-alb or 6 - Sets an Active Load Balancing (ALB) policy for fault tolerance and load balancing. Includes transmit and receive load balancing for IPV4 traffic. Receive load balancing is achieved through ARP negotiation.

num_unsol_na=<number>

Specifies the number of unsolicited IPv6 Neighbor Advertisements to be issued after a failover event. One unsolicited NA is issued immediately after the failover.

The valid range is 0 - 255; the default value is 1. This option affects only the active-backup mode.

primary=<interface_name>

Specifies the interface name, such as eth0, of the primary device. The primary device is the first of the bonding interfaces to be used and is not abandoned unless it fails. This setting is particularly useful when one NIC in the bonding interface is faster and, therefore, able to handle a bigger load.

This setting is only valid when the bonding interface is in active-backup mode. Refer to /usr/share/doc/kernel-doc-<kernel-version>/Documentation/networking/bonding.txt for more information.

primary_reselect=<value>

Specifies the reselection policy for the primary slave. This affects how the primary slave is chosen to become the active slave when failure of the active slave or recovery of the primary slave occurs. This option is designed to prevent flip-flopping between the primary slave and other slaves. Possible values are:

• always or 0 (default) - The primary slave becomes the active slave whenever it comes back up.

• better or 1 - The primary slave becomes the active slave when it comes back up, if the speed and duplex of the primary slave is better than the speed and duplex of the current active slave.
• failure or 2 - The primary slave becomes the active slave only if the current active slave fails and the primary slave is up.

The primary_reselect setting is ignored in two cases:

• If no slaves are active, the first slave to recover is made the active slave.

• When initially enslaved, the primary slave is always made the active slave.

Changing the primary_reselect policy via sysfs will cause an immediate selection of the best active slave according to the new policy. This may or may not result in a change of the active slave, depending upon the circumstances

updelay=<time_in_milliseconds>

Specifies (in milliseconds) how long to wait before enabling a link. The value must be a multiple of the value specified in the miimon parameter. The value is set to 0 by default, which disables it.

use_carrier=<number>

Specifies whether or not miimon should use MII/ETHTOOL ioctls or netif_carrier_ok() to determine the link state. The netif_carrier_ok() function relies on the device driver to maintains its state with netif_carrier_on/off ; most device drivers support this function.

The MII/ETHROOL ioctls tools utilize a deprecated calling sequence within the kernel. However, this is still configurable in case your device driver does not support netif_carrier_on/off .

Valid values are:

• 1 - Default setting. Enables the use of netif_carrier_ok().

• 0 - Enables the use of MII/ETHTOOL ioctls.

If the bonding interface insists that the link is up when it should not be, it is possible that your network device driver does not support netif_carrier_on/off .

xmit_hash_policy=<value>

Selects the transmit hash policy used for slave selection in balance-xor and 802.3ad modes. Possible values are:

• 0 or layer2 - Default setting. This option uses the XOR of hardware MAC addresses to generate the hash. The formula used is:

(<source_MAC_address> XOR <destination_MAC>) MODULO <slave_count>

This algorithm will place all traffic to a particular network peer on the same slave, and is 802.3ad compliant.

• 1 or layer3+4 - Uses upper layer protocol information (when available) to generate the hash. This allows for traffic to a particular network peer to span multiple slaves, although a single connection will not span multiple slaves.

  The formula for unfragmented TCP and UDP packets used is:

  ((<source_port> XOR <dest_port>) XOR
    ((<source_IP> XOR <dest_IP>) AND 0xffff)
      MODULO <slave_count>

  For fragmented TCP or UDP packets and all other IP protocol traffic, the source and destination port information is omitted. For non-IP traffic, the formula is the same as the layer2 transmit hash policy.

  This policy intends to mimic the behavior of certain switches; particularly, Cisco switches with PFC2 as well as some Foundry and IBM products.

  The algorithm used by this policy is not 802.3ad compliant.

• 2 or layer2+3 - Uses a combination of layer2 and layer3 protocol information to generate the hash.

  Uses XOR of hardware MAC addresses and IP addresses to generate the hash. The formula is:

  (((<source_IP> XOR <dest_IP>) AND 0xffff) XOR
    ( <source_MAC> XOR <destination_MAC> ))
      MODULO <slave_count>

  This algorithm will place all traffic to a particular network peer on the same slave. For non-IP traffic, the formula is the same as for the layer2 transmit hash policy.

  This policy is intended to provide a more balanced distribution of traffic than layer2 alone, especially in environments where a layer3 gateway device is required to reach most destinations.

  This algorithm is 802.3ad compliant.