VLANs
VLANs
VLANs
This chapter describes how to configure Virtual LANs (VLANs) on a ServerIron ADX.
The Overview section provides basic information about VLAN options available on a ServerIron ADX. Following
this section, other sections provide configuration procedures and examples.
To display configuration information for VLANs, see Displaying VLAN Information on page 5-36.
Overview
This section describes the ServerIron ADX VLAN features. Configuration procedures and examples appear in
later sections of this chapter.
Layer 2 port-based VLAN a set of physical ports that share a common, exclusive Layer 2 broadcast domain
IP subnet VLANs a subset of ports in a port-based VLAN that share a common, exclusive subnet broadcast
domain for a specified IP subnet
When a ServerIron ADX device receives a packet on a port that is a member of a VLAN, the device forwards the
packet based on the following VLAN hierarchy:
If the port belongs to an IP subnet VLAN and the packet belongs to the corresponding IP subnet, the device
forwards the packet to all the ports within that IP subnet VLAN.
If the packet cannot be forwarded based on either of the VLAN membership types listed above, but the packet
can be forwarded at Layer 2, the device forwards the packet on all the ports within the receiving ports portbased VLAN.
June, 2009
5-1
Figure 5.1 on page 5-2 shows an example of a ServerIron ADX on which a Layer 2 port-based VLAN has been
configured.
Figure 5.1
DEFAULT-VLAN
VLAN ID = 1
Layer 2 Port-based VLAN
A port can belong to only one port-based VLAN, unless you apply 802.1q tagging to the port. 802.1q tagging
allows the port to add a four-byte tag field, which contains the VLAN ID, to each packet sent on the port. You also
can configure port-based VLANs that span multiple devices by tagging the ports within the VLAN. The tag
enables each device that receives the packet to determine the VLAN the packet belongs to. 802.1q tagging
applies only to Layer 2 VLANs, not to Layer 3 VLANs.
Since each port-based VLAN is a separate Layer 2 broadcast domain, by default each VLAN runs a separate
instance of the Spanning Tree Protocol (STP).
Layer 2 traffic is bridged within a port-based VLAN and Layer 2 broadcasts are sent to all the ports within the
VLAN.
5-2
June, 2009
parameters. You configure a separate virtual routing interface on each VLAN that you want to be able to route
from or to. For example, if you configure two IP subnet VLANs on a Layer 3 Switch, you can configure a virtual
routing interface on each VLAN, then configure IP routing parameters for the subnets. Thus, the Layer 3 Switch
forwards IP subnet broadcasts within each VLAN at Layer 2 but routes Layer 3 traffic between the VLANs using
the virtual routing interfaces.
NOTE: The Layer 3 Switch uses the lowest MAC address on the device (the MAC address of port 1 or 1/1) as the
MAC address for all ports within all virtual routing interfaces you configure on the device.
The routing parameters and the syntax for configuring them are the same as when you configure a physical
interface for routing.
All the ports within an IP Subnet VLAN must be in the same port-based VLAN. The IP Subnet VLAN cannot have
ports in multiple port-based VLANs, unless the ports in the port-based VLAN to which you add the IP Subnet
VLAN are 802.1q tagged.
You can configure multiple IP Subnet VLANs within the same port-based VLAN. In addition, a port within a portbased VLAN can belong to multiple VLANs. For example, if you have a port-based VLAN that contains ports 1
10, you can configure port 5 as a member of more than one IP Subnet.
IP Subnet VLANs
For IP, you can provide more granular broadcast control by instead creating the following types of VLAN:
The ServerIron ADX routes packets between VLANs at Layer 3. To configure an IP subnet VLAN to route, you
must add a virtual routing interface to the VLAN, then configure the appropriate routing parameters on the virtual
routing interface.
NOTE: The Layer 3 Switch routes packets between VLANs of the same protocol. The Layer 3 Switch cannot
route from one protocol to another.
Default VLAN
By default, all the ports on a ServerIron ADX are in a single port-based VLAN. This VLAN is called DEFAULTVLAN and is VLAN number 1.
June, 2009
5-3
Figure 5.2 on page 5-4 shows an example of the default Layer 2 port-based VLAN.
Figure 5.2
DEFAULT-VLAN
VLAN ID = 1
Layer 2 Port-based VLAN
When you configure a port-based VLAN, one of the configuration items you provide is the ports that are in the
VLAN. When you configure the VLAN, the ServerIron ADX automatically removes the ports that you place in the
VLAN from DEFAULT-VLAN. By removing the ports from the default VLAN, the ServerIron ADX ensures that each
port resides in only one Layer 2 broadcast domain.
NOTE: Information for the default VLAN is available only after you define another VLAN.
Some network configurations may require that a port be able to reside in two or more Layer 2 broadcast domains
(port-based VLANs). In this case, you can enable a port to reside in multiple port-based VLANs by tagging the
port. See the following section.
If your network requires that you use VLAN ID 1 for a user-configured VLAN, you can reassign the default VLAN to
another valid VLAN ID. See Assigning a Different VLAN ID to the Default VLAN on page 5-11.
802.1q Tagging
802.1q tagging is an IEEE standard that allows a networking device to add information to a Layer 2 packet in order
to identify the VLAN membership of the packet. A ServerIron ADX tags a packet by adding a four-byte tag to the
packet. The tag contains the tag value, which identifies the data as a tag, and also contains the VLAN ID of the
VLAN from which the packet is sent.
5-4
The default tag value is 8100 (hexadecimal). This value comes from the 802.1q specification. You can
change this tag value on a global basis on a ServerIron ADX if needed to be compatible with other vendors
equipment.
June, 2009
The VLAN ID is determined by the VLAN on which the packet is being forwarded.
Figure 5.3 on page 5-5 shows the format of packets with and without the 802.1q tag. The tag format is vendorspecific. To use the tag for VLANs configured across multiple devices, make sure all the devices support the same
tag format.
Figure 5.3
6 bytes
2 bytes
Up to 1500 bytes
4 bytes
Destination
Address
Source
Address
Type
Field
Data
Field
CRC
6 bytes
6 bytes
2 bytes
Up to 1496 bytes
4 bytes
Destination
Address
Source
Address
Length
Field
Data
Field
CRC
Ethernet II
IEEE 802.3
6 bytes
2 bytes
4 bytes
2 bytes
Destination
Address
Source
Address
Type
802.1q
Field
Tag
Type
Field
Data
Field
6 bytes
6 bytes
2 bytes
4 bytes
2 bytes
Destination
Address
Source
Address
Length
802.1q
Field
Tag
Length
Field
Octet 1
Octet 2
Up to 1500 bytes
Up to 1500 bytes
Up to 1496 bytes
Up to 1496 bytes
1 2 3 4 5 6 7 8
Data
Field
Data
Field
Data
Field
4 bytes
CRC
4 bytes
4 bytes
4 bytes
CRC
CRC
IEEE
802.3
Octet 4
(3 bits)
NOTE: You cannot configure a port to be a member of the default port-based VLAN and another port-based
VLAN at the same time. Once you add a port to a port-based VLAN, the port is no longer a member of the default
VLAN. The port returns to the default VLAN only if you delete the other VLAN(s) that contains the port.
If you configure a VLAN that spans multiple devices, you need to use tagging only if a port connecting one of the
devices to the other is a member of more than one port-based VLAN. If a port connecting one device to the other
is a member of only a single port-based VLAN, tagging is not required.
If you use tagging on multiple devices, each device must be configured for tagging and must use the same tag
value. In addition, the implementation of tagging must be compatible on the devices. The tagging on all Brocade
devices is compatible with other Brocade devices.
June, 2009
5-5
Figure 5.4 on page 5-6 shows an example of two devices that have the same Layer 2 port-based VLANs
configured across them. Notice that only one of the VLANs requires tagging.
Figure 5.4
Segment 1
Segment 2
Segment 1
Segment 2
Also by default, each port-based VLAN has a separate instance of STP. Thus, when STP is globally enabled,
each port-based VLAN on the device runs a separate spanning tree.
You can enable or disable STP on the following levels:
Port-based VLAN Affects all ports within the specified port-based VLAN.
STP is a Layer 2 protocol. Thus, you cannot enable or disable STP for individual protocol VLANs or for IP subnet
VLANs. The STP state of a port-based VLAN containing these other types of VLANs determines the STP state for
all the Layer 2 broadcasts within the port-based VLAN. This is true even though Layer 3 protocol broadcasts are
sent on Layer 2 within the VLAN.
It is possible that STP will block one or more ports in a IP subnet VLAN that uses a virtual routing interface to route
to other VLANs. For IP subnet VLANs, even though some of the physical ports of the virtual routing interface are
blocked, the virtual routing interface can still route so long as at least one port in the virtual routing interfaces
protocol VLAN is not blocked by STP.
If you enable Single STP (SSTP) on the device, the ports in all VLANs on which STP is enabled become members
of a single spanning tree. The ports in VLANs on which STP is disabled are excluded from the single spanning
tree.
5-6
June, 2009
For more information, see Configuring Spanning Tree Protocol (STP) and IronSpan Features on page 6-1.
VE = virtual interface
(VE stands for Virtual Ethernet)
VE 3
VE 1
VE 4
VE 2
June, 2009
5-7
For configuration information, see Configuring VLAN Groups and Virtual Routing Interface Groups on page 5-24.
Dynamic ports
Static ports
Dynamic Ports
Dynamic ports are added to a VLAN when you create the VLAN. However, if a dynamically added port does not
receive any traffic for the VLANs IP subnet within ten minutes, the port is removed from the VLAN. However, the
port remains a candidate for port membership. Thus, if the port receives traffic for the VLANs IP subnet, the
ServerIron ADX adds the port back to the VLAN.
After the port is added back to the VLAN, the port can remain an active member of the VLAN up to 20 minutes
without receiving traffic for the VLANs protocol. If the port ages out, it remains a candidate for VLAN membership
and is added back to the VLAN when the VLAN receives protocol traffic. At this point, the port can remain in the
VLAN up to 20 minutes without receiving traffic for the VLANs IP subnet, and so on.
Unless you explicitly add a port statically or exclude a port, the port is a dynamic port and thus can be an active
member of the VLAN, depending on the traffic it receives.
Figure 5.6 on page 5-8 shows an example of a VLAN with dynamic ports. Dynamic ports not only join and leave
the VLAN according to traffic, but also allow some broadcast packets of the specific protocol to leak through the
VLAN. See Broadcast Leaks on page 5-9.
Figure 5.6
VLAN with dynamic portsall ports are active when you create the VLAN
A = active port
C = candidate port
When you add ports dynamically,
all the ports are added when you add
the VLAN.
5-8
June, 2009
Ports in a new IP subnet VLAN that do not receive traffic for the VLANs protocol age out after 20 minutes and
become candidate ports. Figure 5.7 on page 5-9 shows what happens if a candidate port receives traffic for the
VLANs protocol.
Figure 5.7
VLAN with dynamic portscandidate ports become active again if they receive protocol traffic
Static Ports
Static ports are permanent members of the IP subnet VLAN. The ports remain active members of the VLAN
regardless of whether the ports receive traffic for the VLANs protocol. You must explicitly identify the port as a
static port when you add it to the VLAN. Otherwise, the port is dynamic and is subject to aging out.
Excluded Ports
If you want to prevent a port in a port-based VLAN from ever becoming a member of an IP subnet VLAN
configured in the port-based VLAN, you can explicitly exclude the port. You exclude the port when you configure
the IP subnet VLAN.
Excluded ports do not leak broadcast packets. See Broadcast Leaks on page 5-9.
Broadcast Leaks
A dynamic port becomes a member of a Layer 3 protocol VLAN when traffic from the VLAN's protocol is received
on the port. After this point, the port remains an active member of the protocol VLAN, unless the port does not
receive traffic from the VLAN's protocol for 20 minutes. If the port does not receive traffic for the VLAN's protocol
for 20 minutes, the port ages out and is no longer an active member of the VLAN.
To enable a host that has been silent for awhile to send and receive packets, the dynamic ports that are currently
members of the Layer 3 protocol VLAN "leak" Layer 3 broadcast packets to the ports that have aged out. When a
host connected to one of the aged out ports responds to a leaked broadcast, the port is added to the protocol
VLAN again.
To "leak" Layer 3 broadcast traffic, an active port sends 1/8th of the Layer 3 broadcast traffic to the inactive (aged
out) ports.
Static ports do not age out and do not leak broadcast packets.
June, 2009
5-9
As a ServerIron ADX receives packets, the VLAN classification starts from the highest level VLAN first. Therefore,
if an interface is configured as a member of both a port-based VLAN and an IP subnet VLAN, IP packets coming
into the interface are classified as members of the IP subnet VLAN because that VLAN is higher in the VLAN
hierarchy.
A port can belong to multiple, unique, overlapping IP subnet VLANs without VLAN tagging.
A port can belong to multiple, overlapping Layer 2 port-based VLANs only if the port is a tagged port. Packets
sent out of a tagged port use an 802.1q-tagged frame.
When both port and IP subnet VLANs are configured on a given device, all IP subnet VLANs must be strictly
contained within a port-based VLAN. An IP subnet VLAN cannot include ports from multiple port-based
VLANs. This rule is required to ensure that port-based VLANs remain loop-free Layer 2 broadcast domains.
Multiple IP subnet VLANs are configurable within each port-based VLAN on the Layer 2 Switch.
Removing a configured port-based VLAN from a Foundry Networks Layer 2 Switch or Layer 3 Switch
automatically removes any protocol-based VLAN, IP subnet VLAN, AppleTalk cable VLAN, or IPX network
VLAN, or any Virtual Ethernet router interfaces defined within the Port-based VLAN.
June, 2009
5 - 11
To reassign the default VLAN to a different VLAN ID, enter the following command:
ServerIron(config)# default-vlan-id 4095
Syntax: [no] default-vlan-d <vlan-id>
You must specify a valid VLAN ID that is not already in use. For example, if you have already defined VLAN 10, do
not try to use 10 as the new VLAN ID for the default VLAN. Valid VLAN IDs are numbers from 1 4096.
NOTE: Changing the default VLAN name does not change the properties of the default VLAN. Changing the
name allows you to use the VLAN ID 1 as a configurable VLAN.
NOTE: VLAN ID 4094 is reserved for use by Single STP.
Create a VLAN.
Delete a VLAN.
Modify a VLAN.
EXAMPLE:
Figure 5.8 on page 5-13 shows a simple port-based VLAN configuration using a single Brocade Layer 2 Switch.
All ports within each VLAN are untagged. One untagged port within each VLAN is used to connect the Layer 2
Switch to a Layer 3 Switch (in this example, a NetIron) for Layer 3 connectivity between the two port-based
VLANs.
5 - 12
June, 2009
Figure 5.8
ServerIron ADX
interface e 1
IP Subnet 1
VLAN 222
Ports 1 - 8
Port 1
interface e 2
IP Subnet 2
Port 9
VLAN 333
Ports 9 - 16
ServerIron
ADX
Ports 2 - 8
IP Subnet 1
Ports 9 - 16
IP Subnet 2
To create the two port-based VLANs shown in Figure 5.8 on page 5-13, use the following method.
ServerIron(config)# vlan 222
ServerIron(config-vlan-222)#
ServerIron(config-vlan-222)#
ServerIron(config-vlan-333)#
by port
untag e 1 to 8
vlan 333 by port
untag e 9 to 16
June, 2009
5 - 13
Figure 5.9
Router
IP Subnet1
IP Subnet2
Port 17
Port 18
IP Subnet3
Port 19
IP Subnet4
Port 20
ServerIron-A
ROOT BRIDGE
FOR
VLAN - BLUE
VLAN - RED
ServerIron-B
ROOT BRIDGE
FOR
VLAN - BROWN
VLAN - GREEN
ServerIron
FastIron Workgroup
Link
Activity
Link
Activity
Console
Power
11
13
15
10
12
14
16
FDX
100
Link / Act
FDX
100
Link / Act
FDX
100
Link / Act
FDX
100
Link / Act
FDX
100
Link / Act
FDX
100
Link / Act
17
19
21
23
18
20
22
24
Router
To configure the Port-based VLANs on the ServerIron ADX Layer 2 Switches in Figure 5.9 on page 5-14, use the
following method.
5 - 14
June, 2009
June, 2009
5 - 15
Access the global CONFIG level of the CLI on by entering the following commands:
ServerIron-A> enable
No password has been assigned yet...
ServerIron-A# configure terminal
ServerIron-A(config)#
2.
3.
Enter the following commands to exit the CONFIG level and save the configuration to the system-config file on
flash memory:
ServerIron-A(config)#
ServerIron-A(config)# end
ServerIron-A# write memory
FastIron-A#
4.
Access the global CONFIG level of the CLI on ServerIron-A by entering the following command:
ServerIron-A> enable
No password has been assigned yet...
ServerIron-A# configure terminal
ServerIron-A(config)#
2.
Access the level of the CLI for configuring port-based VLAN 4 by entering the following command:
ServerIron-A(config)#
ServerIron-A(config)# vlan 4
ServerIron-A(config-vlan-4)#
3.
4.
Enter the following commands to exit the VLAN CONFIG mode and save the configuration to the systemconfig file on flash memory:
ServerIron-A(config-vlan-4)#
ServerIron-A(config-vlan-4)# end
ServerIron-A# write memory
ServerIron-A#
5 - 16
June, 2009
Access the global CONFIG level of the CLI on ServerIron-A by entering the following commands:
ServerIron-A> enable
No password has been assigned yet...
ServerIron-A# configure terminal
ServerIron-A(config)#
2.
Access the level of the CLI for configuring port-based VLAN 3 by entering the following command:
ServerIron-A(config)#
ServerIron-A(config)# vlan 3
ServerIron-A(config-vlan-3)#
3.
From VLAN 3s configuration level of the CLI, enter the following command to enable STP on all tagged and
untagged ports associated with VLAN 3.
ServerIron-B(config-vlan-3)#
ServerIron-B(config-vlan-3)# spanning-tree
ServerIron-B(config-vlan-3)#
4.
Enter the following commands to exit the VLAN CONFIG mode and save the configuration to the systemconfig file on flash memory:
ServerIron-B(config-vlan-3)#
ServerIron-B(config-vlan-3)# end
ServerIron-B# write memory
ServerIron-B#
5.
Repeat steps 1 4.
NOTE: You do not need to configure values for the STP parameters. All parameters have default values as
noted below. Additionally, all values will be globally applied to all ports on the system or on the port-based VLAN
for which they are defined.
To configure a specific path-cost or priority value for a given port, enter those values using the key words in the
brackets [ ] shown in the syntax summary below. If you do not want to specify values for any given port, this
portion of the command is not required.
Syntax: vlan <vlan-id> by port
Syntax: [no] spanning-tree
Syntax: spanning-tree [ethernet <portnum> path-cost <value> priority <value>] forward-delay <value>
hello-time <value> maximum-age <time> priority <value>
Bridge STP Parameters (applied to all ports within a VLAN)
Forward Delay the period of time a bridge will wait (the listen and learn period) before forwarding data
packets. Possible values: 4 30 seconds. Default is 15.
Maximum Age the interval a bridge will wait for receipt of a hello packet before initiating a topology change.
Possible values: 6 40 seconds. Default is 20.
Hello Time the interval of time between each configuration BPDU sent by the root bridge. Possible values:
1 10 seconds. Default is 2.
Priority a parameter used to identify the root bridge in a network. The bridge with the lowest value has the
June, 2009
5 - 17
highest priority and is the root. Possible values: 1 65,535. Default is 32,678.
Port Parameters (applied to a specified port within a VLAN)
Path Cost a parameter used to assign a higher or lower path cost to a port. Possible values: 1 65535.
Default is (1000/Port Speed) for Half-Duplex ports and is (1000/Port Speed)/2 for Full-Duplex ports.
Priority value determines when a port will be rerouted in relation to other ports. Possible values: 0 255.
Default is 128.
Configuration Example
Suppose you want to create three separate Layer 3 broadcast domains within a single Layer 2 STP broadcast
domain:
Also suppose you want a single router interface to be present within all of these separate broadcast domains,
without using IEEE 802.1q VLAN tagging or any proprietary form of VLAN tagging.
Figure 5.10 on page 5-18 shows this configuration.
Figure 5.10
ServerIron ADX
Port 25
IP-Subnet1
IP-Subnet 2
IP Subnet 3
Port 25
ServerIron ADX
Ports 17-25
IP-Subnet 3
To configure the VLANs shown in Figure 5.10 on page 5-18, use the following procedure.
1.
To permanently assign ports 1 8 and port 25 to IP subnet VLAN 1.1.1.0, enter the following commands:
ServerIron> en
5 - 18
June, 2009
To permanently assign ports 9 16 and port 25 to IP subnet VLAN 1.1.2.0, enter the following commands:
ServerIron(config-ip-subnet)# ip-subnet 1.1.2.0/24 name Yellow
ServerIron(config-ip-subnet)# no dynamic
ServerIron(config-ip-subnet)# static ethernet 9 to 16 ethernet 25
3.
To permanently assign ports 17 25 to IP subnet VLAN 1.1.3.0, enter the following commands:
ServerIron(config-ip-subnet)# ip-subnet 1.1.3.0/24 name Brown
ServerIron(config-ip-subnet)# no dynamic
ServerIron(config-ip-subnet)# static ethernet 17 to 25
June, 2009
5 - 19
address to each virtual routing interface. The IP address on each of the virtual routing interfaces must be in a
separate subnet. The Brocade device routes Layer 3 traffic between the subnets using the subnet addresses.
NOTE: Before using the method described in this section, see Configuring VLAN Groups and Virtual Routing
Interface Groups on page 5-24. You might be able to achieve the results you want using the methods in that
section instead.
Figure 5.11 on page 5-20 shows an example of this type of configuration.
Figure 5.11
VLAN 2
VLAN 3
VLAN 4
BigIron
Switching Router
VLAN 2
VE 1
-IP 10.0.0.1/24
VLAN 3
VE 2
-IP 10.0.1.1/24
VLAN 4
VE 3
-IP 10.0.2.1/24
As shown in this example, each VLAN has a separate IP subnet address. If you need to conserve IP subnet
addresses, you can configure multiple VLANs with the same IP subnet address, as shown in Figure 5.12 on
page 5-21.
5 - 20
June, 2009
Figure 5.12
VLAN 2
VLAN 3
VLAN 4
BigIron
Switching Router
VLAN 2
VE 1
-IP 10.0.0.1/24
VLAN 3
VE 2
-Follow VE 1
VLAN 4
VE 3
-Follow VE 1
Each VLAN still requires a separate virtual routing interface. However, all three VLANs now use the same IP
subnet address.
In addition to conserving IP subnet addresses, this feature allows containment of Layer 2 broadcasts to segments
within an IP subnet. For ISP environments where the same IP subnet is allocated to different customers, placing
each customer in a separate VLAN allows all customers to share the IP subnet address, while at the same time
isolating them from one anothers Layer 2 broadcasts.
NOTE: You can provide redundancy to an IP subnet address that contains multiple VLANs using a pair of
Brocade Layer 3 Switches configured for Brocades VRRP (Virtual Router Redundancy Protocol).
The Brocade device performs proxy Address Resolution Protocol (ARP) for hosts that want to send IP traffic to
hosts in other VLANs that are sharing the same IP subnet address. If the source and destination hosts are in the
same VLAN, the Brocade device does not need to use ARP.
If a host attached to one VLAN sends an ARP message for the MAC address of a host in one of the other
VLANs using the same IP subnet address, the Brocade device performs a proxy ARP on behalf of the other
host. The Brocade device then replies to the ARP by sending the virtual routing interface MAC address. The
Brocade device uses the same MAC address for all virtual routing interfaces.
When the host that sent the ARP then sends a unicast packet addressed to the virtual routing interfaces MAC
address, the device switches the packet on Layer 3 to the destination host on the VLAN.
June, 2009
5 - 21
NOTE: If the Brocade devices ARP table does not contain the requested host, the Brocade device forwards
the ARP request on Layer 2 to the same VLAN as the one that received the ARP request. Then the device
sends an ARP for the destination to the other VLANs that are using the same IP subnet address.
If the destination is in the same VLAN as the source, the Brocade device does not need to perform a proxy
ARP.
Configure a separate virtual routing interface for each VLAN, but do not add an IP subnet address to more
than one of the virtual routing interfaces.
Configure the virtual routing interfaces that do not have the IP subnet address to follow the virtual routing
interface that does have the address.
To configure the VLANs shown in Figure 5.12 on page 5-21, you could enter the following commands.
ServerIron(config)# vlan 1
ServerIron(config-vlan-1)#
ServerIron(config-vlan-1)#
ServerIron(config-vlan-1)#
by port
untag ethernet 1/1
tag ethernet 1/8
router-interface ve 1
vlan 2 by port
untag ethernet 1/2
tag ethernet 1/8
router-interface ve 2
vlan 3 by port
untag ethernet 1/5 to 1/6
tag ethernet 1/8
router-interface ve 3
interface
ip follow
interface
ip follow
ve
ve
ve
ve
2
1
3
1
5 - 22
June, 2009
The IP follower feature allows multiple virtual routing interfaces to share the same IP address. One virtual routing
interface has the IP address and the other virtual routing interfaces are configured to follow the virtual routing
interface that has the address.
By default, the follower interfaces are secured by the ACLs that are applied to the interface that has the address.
In fact, an ACL applied to a follower interface is ignored. For example, if you configure virtual routing interfaces 1,
2, and 3, and configure interfaces 2 and 3 to follow interface 1, then the ACLs applied to interface 1 also apply to
interfaces 2 and 3. Any ACLs applied separately to interface 2 or 3 are ignored.
You can enable a follower virtual routing interface to use the ACLs you apply to it instead of using the ACLs applied
to the interface that has the address. For example, you can enable virtual routing interface 2 to use its own ACLs
instead of using interface 1s ACLs.
To enable a virtual routing interface to use its own ACLs instead of the ACLs of the interface it is following, enter
the following command at the configuration level for the interface:
ServerIron(config-vif-2)# no ip follow acl
Syntax: [no] ip follow acl
The following commands show a complete IP follower configuration. Virtual routing interfaces 2 and 3 have been
configured to share the IP address of virtual routing interface 1, but also have been configured to use their own
ACLs instead of virtual routing interface 1s ACLs.
ServerIron(config)# vlan 1 name primary_vlan
ServerIron(config-vlan-1)# untag ethernet 1/1
ServerIron(config-vlan-1)# tag ethernet 1/8
ServerIron(config-vlan-1)# router-interface ve 1
ServerIron(config-vlan-1)# exit
ServerIron(config)# interface ve 1
ServerIron(config-ve-1)# ip address 10.0.0.1/24
ServerIron(config-ve-1)# ip access-group 1 in
ServerIron(config-ve-1)# exit
ServerIron(config)# vlan 2 name followerA
ServerIron(config-vlan-2)# untag ethernet 1/2
ServerIron(config-vlan-2)# tag ethernet 1/8
ServerIron(config-vlan-2)# router-interface ve 2
ServerIron(config-vlan-2)# exit
ServerIron(config)# interface ve 2
ServerIron(config-ve-2)# ip follow ve 1
ServerIron(config-v2-2)# no ip follow acl
ServerIron(config-ve-2)# ip access-group 2 in
ServerIron(config-ve-2)# exit
ServerIron(config)# vlan 3 name followerB
ServerIron(config-vlan-3)# untag ethernet 1/5 to 1/6
ServerIron(config-vlan-3)# tag ethernet 1/8
ServerIron(config-vlan-3)# router-interface ve 3
ServerIron(config-vlan-3)# exit
ServerIron(config)# interface ve 3
ServerIron(config-ve-3)# ip follow ve 1
ServerIron(config-ve-3)# no ip follow acl
ServerIron(config-ve-3)# ip access-group 3 out
ServerIron(config-ve-3)# exit
5 - 23
other hosts connected to port 4 and port 6 respectively and in IP subnets 10.10.10.0/24 and 20.20.20.0/24
respectively can send IP packets to each other via the virtual routing interfaces VE10 and VE20. In this situation,
the ServerIron ADX is Layer-3 routing the IP packets from one VLAN to another. This example is configured as
described in the following.
The following commands create a port-based VLAN and add two ports as tagged and untagged members
respectively:
ServerIron(config)# vlan 10
ServerIron(config-vlan-10)# untag ethernet 4
ServerIron(config-vlan-10)# tag ethernet 5
The following commands create a virtual routing interface for VLAN 10 and configure an IP address on the virtual
routing interface.
ServerIron(config-vlan-10)# router-interface ve 10
ServerIron(config-vlan-10)# interface ve 10
ServerIron(config-vif-10)# ip address 10.10.10.1/24
The following commands create a second virtual routing interface for VLAN 20.
ServerIron(config)# vlan 20
ServerIron(config-vlan-20)# untag ethernet 6
ServerIron(config-vlan-20)# tag ethernet 5
ServerIron(config-vlan-20)# router-interface ve 20
ServerIron(config-vlan-20)# interface ve 20
ServerIron(config-vif-20)# ip address 20.20.20.1/24
The VLAN group feature allows you to create multiple port-based VLANs with identical port members. Since
the member ports are shared by all the VLANs within the group, you must add the ports as tagged ports. This
feature not only simplifies VLAN configuration but also allows you to have a large number of identically
configured VLANs in a startup-config file on the devices flash memory module. Normally, a startup-config file
with a large number of VLANs might not fit on the flash memory module. By grouping the identically
configured VLANs, you can conserve space in the startup-config file so that it fits on the flash memory
module.
The virtual routing interface group feature is useful when you want to configure the same IP subnet address
on all the port-based VLANs within a VLAN group. You can configure a virtual routing interface group only
after you configure a VLAN group with the same ID. The virtual routing interface group automatically applies
to the VLANs in the VLAN group that has the same ID and cannot be applied to other VLAN groups or to
individual VLANs.
You can create up to 32 VLAN groups and 32 virtual routing interface groups. A virtual routing interface group
always applies only to the VLANs in the VLAN group with the same ID.
NOTE: Depending on the size of the VLAN ID range you want to use for the VLAN group, you might need to
allocate additional memory for VLANs. On Layer 3 Switches, if you allocate additional memory for VLANs, you
also need to allocate the same amount of memory for virtual routing interfaces. This is true regardless of whether
you use the virtual routing interface groups. To allocate additional memory, see Allocating Memory for More
VLANs or Virtual Routing Interfaces on page 5-27.
5 - 24
June, 2009
June, 2009
5 - 25
Displaying the VLAN Group and Virtual Routing Interface Group Information
To verify configuration of VLAN groups and virtual routing interface groups, display the running-config file. If you
have saved the configuration to the startup-config file, you also can verify the configuration by displaying the
startup-config file. The following example shows the running-config information for the VLAN group and virtual
routing interface group configured in the previous examples. The information appears in the same way in the
startup-config file.
ServerIron(config)# show running-config
lines not related to the VLAN group omitted...
vlan-group 1 vlan 2 to 900
add-vlan 1001 to 1002
tagged ethe 1/1 to 1/2
router-interface-group
5 - 26
June, 2009
June, 2009
5 - 27
the end points of the channel appear to each other to be directly attached. The network that connects them is
transparent to the two devices.
You can aggregate up to 4094 VLANs within another VLAN. This provides a total VLAN capacity on one Brocade
device of 16,760,836 channels (4094 * 4094).
The devices connected through the channel are not visible to devices in other channels. Therefore, each client
has a private link to the other side of the channel.
The feature allows point-to-point and point-to-multipoint connections.
Figure 5.13 on page 5-28 shows a conceptual picture of the service that aggregated VLANs provide. Aggregated
VLANs provide a path for multiple client channels. The channels do not receive traffic from other channels. Thus,
each channel is a private link.
Figure 5.13
. . .
Client 3
. . .
Client 5
Client 1
192.168.1.69/24
sub-net
192.168.1.0/24
Each client connected to the edge device is in its own port-based VLAN, which is like an ATM channel. All the
clients VLANs are aggregated by the edge device into a single VLAN for connection to the core. The single VLAN
that aggregates the clients VLANs is like an ATM path.
The device that aggregates the VLANs forwards the aggregated VLAN traffic through the core. The core can
consist of multiple devices that forward the aggregated VLAN traffic. The edge device at the other end of the core
separates the aggregated VLANs into the individual client VLANs before forwarding the traffic. The edge devices
forward the individual client traffic to the clients. For the clients perspective, the channel is a direct point-to-point
link.
Figure 5.14 on page 5-29 shows an example application that uses aggregated VLANs. This configuration
includes the client connections shown in Figure 5.13 on page 5-28.
5 - 28
June, 2009
Figure 5.14
. . .
Client 3
Port 1/3
VLAN 103
. . .
Client 6
Port 1/1
VLAN 101
Client 5
Port 1/5
VLAN 105
Client 1
192.168.1.69/24
. . .
Client 8
Port 1/3
VLAN 103
. . .
Client 10
Port 1/5
VLAN 105
209.157.2.12/24
Ports 1/1 - 1/5
Untagged
Device A
Tag Type 8100
Port 2/1
Tagged
Port 2/1
Tagged
Port 3/1
Untagged
Device B
Tag Type 8100
Port 3/2
Untagged
Device C
Tag Type 9100
VLAN Aggregation
Enabled
Port 4/1
Tagged
Port 4/1
Tagged
Device D
Tag Type 9100
VLAN Aggregation
Enabled
Port 3/1
Untagged
Port 3/2
Untagged
Port 2/1
Tagged
Port 2/1
Tagged
Device E
Tag Type 8100
Device F
Tag Type 8100
192.168.1.129/24
In this example, a collocation service provides private channels for multiple clients. Although the same devices
are used for all the clients, the VLANs ensure that each client receives its own Layer 2 broadcast domain,
separate from the broadcast domains of other clients. For example, client 1 cannot ping client 5.
The clients at each end of a channel appear to each other to be directly connected and thus can be on the same
subnet and use network services that require connection to the same subnet. In this example, client 1 is in subnet
192.168.1.0/24 and so is the device at the other end of client 1s channel.
Since each VLAN configured on the core devices is an aggregate of multiple client VLANs, the aggregated VLANs
greatly increase the number of clients a core device can accommodate.
This example shows a single link between the core devices. However, you can use a trunk group to add link-level
redundancy.
On each edge device, configure a separate port-based VLAN for each client connected to the edge device. In
each client VLAN:
June, 2009
5 - 29
Add the port connected to the core device (the device that will aggregate the VLANs) as a tagged port.
This port must be tagged because all the client VLANs share the port as an uplink to the core device.
Enable VLAN aggregation. This support allows the core device to add an additional tag to each Ethernet
frame that contains a VLAN packet from the edge device. The additional tag identifies the aggregate
VLAN (the path). However, the additional tag can cause the frame to be longer than the maximum
supported frame size. The larger frame support allows Ethernet frames up to 1530 bytes long.
NOTE: Enable the VLAN aggregation option only on the core devices.
Configure a VLAN tag type (tag ID) that is different than the tag type used on the edge devices. If you
use the default tag type (8100) on the edge devices, set the tag type on the core devices to another
value, such as 9100. The tag type must be the same on all the core devices. The edge devices also
must have the same tag type but the type must be different from the tag type on the core devices.
NOTE: You can enable the Spanning Tree Protocol (STP) on the edge devices or the core devices, but not both.
If you enable STP on the edge devices and the core devices, STP will prevent client traffic from travelling through
the core to the other side.
5 - 30
June, 2009
ServerIron(config)# aggregated-vlan
ServerIron(config)# vlan 101 by port
ServerIron(config-vlan-101)# tagged ethernet 4/1
ServerIron(config-vlan-101)# untagged ethernet 3/1
ServerIron(config-vlan-101)# exit
ServerIron(config)# vlan 102 by port
ServerIron(config-vlan-102)# tagged ethernet 4/1
ServerIron(config-vlan-102)# untagged ethernet 3/2
ServerIron(config-vlan-102)# exit
ServerIron(config)# write memory
Syntax: [no] tag-type <num>
Syntax: [no] aggregated-vlan
The <num> parameter specifies the tag type can be a hexadecimal value from 0 ffff. The default is 8100.
5 - 31
5 - 32
June, 2009
June, 2009
5 - 33
Figure 5.15
Untagged
Traffic
Hub
Port 2/11
Tagged, VLAN 20
dual-mode
Port 2/9
Tagged, VLAN 20
VLAN 20
Traffic
Port 2/10
Untagged
Untagged
Traffic
To enable the dual-mode feature on port 2/11 in Figure 5.15 on page 5-34:
ServerIron(config)# vlan 20
ServerIron(config-vlan-20)# tagged e 2/11
ServerIron(config-vlan-20)# tagged e 2/9
ServerIron(config-vlan-20)# int e 2/11
ServerIron(config-if-e100-2/11)# dual-mode
ServerIron(config-if-e100-2/11)# exit
Syntax: [no] dual-mode
You can configure a dual-mode port to transmit traffic for a specified VLAN (other than the DEFAULT-VLAN) as
untagged, while transmitting traffic for other VLANs as tagged. Figure 5.16 on page 5-34 illustrates this
enhancement.
Figure 5.16
VLAN 10
Untagged
Traffic
Port 2/10
Untagged, VLAN 10
Hub
Port 2/9
Tagged, VLAN 20
VLAN 20
Tagged
Traffic
VLAN 20
Tagged
Traffic
In Figure 5.16 on page 5-34, tagged port 2/11 is a dual-mode port belonging to VLANs 10 and 20. The default
VLAN assigned to this dual-mode port is 10. This means that the port transmits tagged traffic on VLAN 20 (and all
other VLANs to which the port belongs) and transmits untagged traffic on VLAN 10.
5 - 34
June, 2009
The dual-mode feature allows tagged traffic for VLAN 20 and untagged traffic for VLAN 10 to go through port 2/11
at the same time. A dual-mode port transmits only untagged traffic on its default VLAN (that is, either VLAN 1, or
a user-specified VLAN ID), and only tagged traffic on all other VLANs.
The following commands configure VLANs 10 and 20 in Figure 5.16 on page 5-34. Tagged port 2/11 is added to
VLANs 10 and 20, then designated a dual-mode port whose specified default VLAN is 10. In this configuration,
port 2/11 transmits only untagged traffic on VLAN 10 and only tagged traffic on VLAN 20.
ServerIron(config)# vlan 10
ServerIron(config-vlan-10)#
ServerIron(config-vlan-10)#
ServerIron(config-vlan-10)#
by port
untagged e 2/10
tagged e 2/11
exit
ServerIron(config)# vlan 20
ServerIron(config-vlan-20)#
ServerIron(config-vlan-20)#
ServerIron(config-vlan-20)#
by port
tagged e 2/9
tagged e 2/11
exit
If you do not specify a <vlan-id> in the dual mode command, the ports default VLAN is set to 1. The port
transmits untagged traffic on the DEFAULT-VLAN.
The dual-mode feature is disabled by default. Only tagged ports can be configured as dual-mode ports.
In trunk group, either all of the ports must be dual-mode, or none of them can be.
The show vlan command displays a separate row for dual-mode ports on each VLAN. For example:
ServerIron(config)# show vlan
Total PORT-VLAN entries: 3
Maximum PORT-VLAN entries: 16
legend: [S=Slot]
PORT-VLAN
Untagged
Untagged
Untagged
Tagged
Uplink
DualMode
PORT-VLAN
Untagged
Tagged
Uplink
DualMode
PORT-VLAN
Untagged
Tagged
Uplink
DualMode
June, 2009
tree Off
16 17 18 19
Off
Off
5 - 35
5 - 36
June, 2009
To display VLAN information for all the VLANs of which port 7/1 is a member, enter the following command:
ServerIron(config)# show vlans e 7/1
Total PORT-VLAN entries: 3
Maximum PORT-VLAN entries: 8
legend: [S=Slot]
PORT-VLAN 100, Name [None], Priority level0, Spanning tree Off
Untagged Ports: (S7) 1 2 3 4
Tagged Ports: None
IP-subnet VLAN 207.95.11.0 255.255.255.0, Dynamic port disabled
Static ports: (S7) 1 2
Exclude ports: None
Dynamic ports: None
Syntax: show vlans [<vlan-id> | ethernet <portnum> ]
The <vlan-id> parameter specifies a VLAN for which you want to display the configuration information.
The ethernet <portnum> parameter specifies a port. If you use this parameter, the command lists all the VLAN
memberships for the port.
June, 2009
5 - 37
5 - 38
June, 2009