Duplex mode may be set to full or half duplex. If the media type does not allow duplex mode to be set, this is indicated by an onscreen message. If media type is set to auto, or if setting duplex mode is not supported, the duplex setting is not saved to
bigip.conf.
Use the following syntax to set the duplex mode:
b interface duplex full | half | auto
Trunks
Link aggregation is the grouping of links (individual physical interfaces) to form a trunk. Link aggregation increases the bandwidth of the individual links in an additive manner. Thus four fast Ethernet links, if aggregated, create a single 400 Mbps link. The other advantage of link aggregation is link fail-over. If one link in a trunk goes down, traffic is simply redistributed over the remaining links.
A trunk must have a controlling link and acquires all the attributes of that controlling link from Layer 2 and above. Thus the trunk automatically acquires the VLAN membership of the controlling link but does not acquire its media type and speed. Outbound packets to the controlling link are load balanced across all of the known-good links in the trunk. Inbound packets from any link in the trunk are treated as if they came from the controlling link. A maximum of eight links may be aggregated. For optimal performance, links should be aggregated in powers of two. Thus ideally you will aggregate two, four, or eight links.
To configure a trunk using the Configuration utility
1. In the navigation pane, click Network. The VLANs screen opens.
2. Click the Trunks tab. The Trunks screen opens.
3. On the Trunks screen, click the Add button. The Add Trunk screen opens.
4. Select the link that is to be the controlling link from the Available Interfaces list and click controlling >>. The interface appears at the top of the Aggregated Interfaces list.
5. Select the remaining link(s) from the Available Interfaces list and click aggregated >>.
The interface(s) appears in the Aggregated Interfaces list below the controlling link.
6. Click Done.
To configure a trunk from the command line
Use the following syntax to configure a trunk from the command line:
b trunk <controlling_if> define <if_list>
Interfaces are specified using the s.p convention, where s is slot number and p is port number. An <if_list> is one or more such interfaces, with multiple interfaces separated by spaces.
For more information on interface naming, refer to Interface naming convention on page 1-115.
Health monitors
Health monitors verify connections and services on nodes that are members of load balancing pools. The monitor checks the node at a set interval. If the node does not respond within a specified timeout period, the node is marked down and traffic is no longer directed to it.
By default, an icmp (Internet Control Message Protocol) monitor is associated with every node that is a member of a load balancing pool. This monitor is of the simplest type, checking only the node address and checking only for a ping response. To change the interval and timeout values of this default check, or to check specific services on a node, you need to configure a custom monitor or monitors to add to the default monitor. The BIG-IP Controller provides a variety of service-specific monitors in template form. Some of these monitors are usable as is (assuming their default values are acceptable) and may be put in service simply by associating them with the nodes to be monitored. In most cases, however, the template is used purely as a template for configuring custom monitors. Configuring custom monitors and placing them in service is a three-step process:
• Selecting the template
• Configuring the monitor from the template • Associating the monitor with the node or nodes
For example, for the default icmp monitor, we selected the icmp
monitor template, as shown in Figure 1.28.
The icmp monitor template has three attributes, interval, timeout, and dest, each with a default value. (All monitor templates have these three basic attributes. As will be seen, other monitor
monitor type icmp { interval 5 timeout 16 dest * }
templates have additional attributes as required by the service type.) These attributes are inherited by the custom monitor when it is configured and can be left at their default values or assigned new values as required.
For the default monitor, template icmp is used as is, that is, as monitor icmp with its default attribute values. To change any of these default values, you would need to create a custom monitor based upon icmp, for example, my_icmp. Only the values that are actually to be changed would need to be specified in the definition of the custom monitor. Therefore, if you wanted to change the timeout values only, you would define the custom monitor as follows:
b monitor my_icmp ’{ use icmp timeout 20 }’
This would create a new monitor in /config/bigip.conf, as shown in Figure 1.29. You can display this monitor using the command b monitor my_icmp show.
Once the custom monitor exists, you associate it with a node or nodes using the Configuration utility or the bigpipe node
command as follows.
b node 11.11.11.1 11.11.11.2 11.11.11.3 monitor use my_icmp
Note
The nodes are identified by IP address only. icmp can ping addresses only, not specific ports on addresses. This creates three instances of monitor my_icmp, one for each address. You can display the instances using the command b node monitor my_icmp show. monitor my_icmp{ #type icmp use "icmp" interval 5 timeout 20 }
Note that each instance takes as its destination the selfsame node it is associated with. This is because the dest value in my_icmp was left at the default *, which tells the instance to use the associated node as its destination. Assigning a specific address to dest, such as 11.11.11.1, would cause the monitor to verify all three addresses by checking that one address, making 11.11.11.2 and 11.11.11.3
dependent on 11.11.11.1.