The VLAN Trunking Protocol (VTP) is an interesting Layer 2 feature of Ciscos
Catalyst switch line, and one that is definitely useful, especially in large,
switched environments that include multiple Virtual Local Area Networks (VLANs).
In the last article in this series, we took at look at the concept of a VLAN,
along with VLAN tagging protocols like ISL. If you recall, the purpose of configuring
VLAN tagging was to allow traffic from multiple VLANs to cross a trunk link used
to interconnect switches. While this capability is great in large environments,
VLAN tagging does nothing to help ease the burden of configuring individual VLANs
on multiple switches. This is where VTP steps in to help.
The main purpose of VTP is to provide a facility by which individual Cisco
switches can be managed as a group for VLAN configuration purposes. For example,
if VTP is enabled on all of your Cisco switches, the creation of a new VLAN on
one switch makes that VLAN available on all switches with the same VTP management
domain. A VTP management domain is simply a group of switches that participate
in sharing VTP information. A given switch can be part of only one VTP management
domain at a time, and is part of no VTP management domain by default.
It should be immediately obvious why VTP is so beneficial. Imagine an environment
in which a network administrator must manage 20 or more switches. Without VTP,
the creation of a new VLAN would require the administrator to define that new
VLAN on all necessary switches individually, a process that is unnecessarily time-consuming.
Instead, with VTP, the administrator could define that VLAN once, and have VTP
worry about propagating the information to all other switches in the same domain
automatically. The main benefit of VTP is the efficiency that it provides in terms
of adding and deleting VLANs, as well as making changes to VLAN configurations
in large environments.
In general, configuring VTP on a Cisco Catalyst switch is not a difficult task.
In fact, once a VTP management domain name is defined on each switch, the process
by which switches exchange VTP information is automatic and requires no further
configuration or day-to-day management. However, in order to full appreciate how
VTP works within a given VTP domain, you must first understand the different VTP
modes for which a switch can be configured. These are outlined in the next section.
VTP Modes
If you intend to make a switch part of a VTP management domain, each switch
must be configured in one of three possible VTP modes. The VTP mode assigned to
a switch will determine how the switch interacts with other VTP switches in the
management domain. The three VTP modes that can be assigned to a Cisco switch
include server mode, client mode, and transparent mode. Each of these roles is
outlined below:
- Server Mode Once VTP is configured on a Cisco switch, the default mode
used is Server Mode. In any given VTP management domain, at least one switch must
be in Server Mode. When in Server Mode, a switch can be used to add, delete, and
modify VLANs, and this information will be passed to all other switches in the
VTP management domain.
- Client Mode When a switch is configured to use VTP Client Mode, it
is simply the recipient of any VLANs added, deleted, or modified by a switch in
Server Mode within the same management domain. A switch in VTP client mode cannot
make any changes to VLAN information.
- Transparent Mode A switch in VTP Transparent Mode will pass VTP updates
received by switches in Server Mode to other switches in the VTP management domain,
but will not actually process the contents of these messages. When individual
VLANs are added, deleted, or modified on a switch running in transparent mode,
the changes are local to that particular switch only, and are not passed to other
switches in the VTP management domain.
Based on the roles of each VTP mode, the use of each should be more or less
obvious. For example, if you had 15 Cisco switches on your network, you could
configure each of them to be in the same VTP management domain. Although each
could theoretically be left in the default Server Mode, it would probably be easier
to leave only one switch in this configuration, and then configure all remaining
switches for VTP Client Mode. Then, when you need to add, delete, or modify a
VLAN, that change can be carried out on the VTP Server Mode switch and passed
to all Client Mode switches automatically. In cases where you need a switch to
act in a relatively standalone manner, or dont want it to propagate information
about its configured VLANs, use Transparent Mode.
VTP Pruning
Although the configuration of trunk links (using protocols like ISL) allows
traffic from multiple VLANs to travel across a single link, this is not always
optimal. For example, imagine a situation where three switches are connected by
two trunk links, as shown below. In this case, all three switches include ports
that are part of VLAN 1, but only Switches A and B include ports in VLAN 2. In
this case, traffic for VLAN 2 would still be passed to Switch C, even though it
does not have any ports configured for VLAN 2.
When VTP Pruning is implemented in a VTP management domain, traffic for a given
VLAN is only passed to a switch across a trunk link if necessary. In this case,
implementing VTP Pruning in the management domain would ensure that traffic for
VLAN 2 is never passed to Switch C until such time as Switch C actually has VLAN
2 ports configured.
First Appeared at
2000Trainers.com
By Dan
DiNicolo
View
Post
Click
To Comment