Configuring ISL Trunking On Catalyst 5500/5000 and 6500/6000 Family Switches
Configuring ISL Trunking On Catalyst 5500/5000 and 6500/6000 Family Switches
Related Documents
More...
Related Products/Technology
Related Discussion
Contents
Introduction
Prerequisites
Requirements
Components Used
Conventions
Creation of a Switch-to-Switch ISL Trunk
Tasks
Step-by-Step Instructions for CatOS
Troubleshoot the Result
Step-by-Step Instructions for Cisco IOS Software
Troubleshoot the Result
Cisco Support Community - Featured Conversations
Related Information
Introduction
This document illustrates how to create a switch-to-switch Inter-Switch Link (ISL) trunk. Trunk
ports enable connections between switches to carry traffic from more than one VLAN. If
trunking is not enabled, the link that connects the two switches only carries traffic from the
VLAN that you have configured on the port. Trunking is not necessary in very simple switched
networks with only one VLAN (broadcast domain). In most LANs, a small portion of traffic
consists of special protocols that manage the network. (A few examples are Cisco Discovery
Protocol [CDP], VLAN Trunk Protocol [VTP], Dynamic Trunking Protocol [DTP], Spanning
Tree Protocol [STP], and Port Aggregation Protocol [PAgP].) You also use the management
VLAN when you ping or establish a Telnet directly to or from the switch. (If you use Catalyst
OS [CatOS], you define the VLAN and the IP address of the switch when you configure the sc0
interface. The Step-by-Step Instructions for CatOS section of this document explains this
process.) In a multi-VLAN environment, many network administrators advocate the restriction
of this management traffic to a single VLAN. The VLAN is normally VLAN 1. The
administrators then configure user traffic to flow in VLANs other than this default VLAN. ISL
(Cisco proprietary) is one of two possible trunking protocols for Ethernet. The other protocol is
the IEEE 802.1Q standard.
This document covers the procedure to configure ISL trunking between Catalyst 5500/5000 and
Catalyst 6500/6000 series switches. The CatOS configuration applies to both Catalyst 5500/5000
and 6500/6000 series switches. However, you can only apply the Cisco IOS® Software
configuration to a Catalyst 6500/6000 series switch.
Prerequisites
Requirements
Components Used
The information in this document is based on these software and hardware versions:
Conventions
For more information on document conventions, refer to the Cisco Technical Tips Conventions.
Switches A and B, by default, have the 2/1 ports in VLAN 1. This configuration allows traffic
from other ports in VLAN 1 to flow between the switches without the need to configure
trunking. References to "the management VLAN" apply to VLAN 1.
Note: VLAN 1 is the default VLAN for all types of Ethernet interfaces, as well as FDDI. VLAN
1 is also the default VLAN for the management interface (sc0).
In the example that follows, you have configured ports 2/2 on both switches in VLAN 2. Traffic
from devices that you have attached to the ports in VLAN 2 do not cross over the link between
the switches. Therefore, PCs A and B are not able to communicate.
The solution is to enable ISL trunking on the link between switches A and B. Trunking adds a
VLAN header to each frame during transmit (multiplex) onto a trunk link. This addition allows
the switch at the other end of the link to demultiplex the frames; the switch then forwards the
frames to the appropriate VLAN ports.
Tasks
There are several types of Ethernet interfaces that support ISL trunking. The 10BASE-T
(common Ethernet) ports do not support trunking, while most 100BASE-T (Fast
Ethernet) ports do support trunking.
Note: In this example, the command specifies the port designator 2/1. This limits the
response to the information that is directly applicable.
3. Connect the two switch ports together with the Ethernet crossover cable.
In this example, the Switch A 2/1 port connects to the Switch B 2/1 port.
4. To verify that the ports are operational, issue the show port 2/1 command on Switch A.
5. Switch-A> (enable) show port 2/1
6. Port Name Status VLAN Level Duplex
Speed Type
7. ----- ------------------ ---------- ---------- ------ ------
----- ------------
8. 2/1 connected 1 normal a-full a-
100 10/100BaseTX
9.
Switch-A> (enable)
10. Issue the set interface sc0 172.16.84.17 255.255.255.0 172.16.84.255 command on
Switch A and the set interface sc0 172.16.84.18 255.255.255.0 172.16.84.255 command
on Switch B.
These commands assign IP addresses from the same subnet to the management ports on
both switches. You may also need to specify in the command the VLAN for sc0 (the
management VLAN). Include this VLAN if the VLAN is different than the default
(VLAN 1).
If you have the output of a show interfaces command from your Cisco device, you can
use the Output Interpreter Tool (registered customers only) to display potential issues
and fixes.
11. To verify that the link between switches A and B is not trunking, issue the show trunk
2/1 command on Switch A.
12. Switch-A> (enable) show trunk 2/1
13. Port Mode Encapsulation Status Native
VLAN
14. -------- ----------- ------------- ------------
-----------
15. 2/1 auto isl not-trunking 1
16.
17. Port VLANs allowed on trunk
18. --------
-----------------------------------------------------------------
----
19. 2/1 1-1005
20.
21. Port VLANs allowed and active in management domain
22. --------
-----------------------------------------------------------------
----
23. 2/1 1
24.
25. Port VLANs in spanning tree forwarding state and not
pruned
26. --------
-----------------------------------------------------------------
----
27. 2/1 1
Switch-A> (enable)
Note: The term Native VLAN in this output indicates the VLAN placement of this port
when the port is not in trunking mode. If you have configured the port for 802.1Q
trunking, the Native VLAN field also indicates the VLAN for which the frames have no
tags; all others have tags. (Conversely, with ISL trunking, every data frame has the
appropriate VLAN identifier.)
The trunking status should be not-trunking because the default mode for the DTP is
auto. DTP is the strategic replacement for Dynamic ISL (DISL) because DTP
incorporates support for 802.1Q trunking negotiation. DTP is available in Catalyst
software version 4.x and later, as well as in certain hardware modules. There are five
different modes to which you can configure DTP. Cisco Technical Support recommends
the configuration of desirable trunking mode on the ports of the trunk link. Step 12
discusses this information in more detail.
28. Ping Switch B from Switch A to verify that the switches can communicate over the link.
29. Switch-A> ping 172.16.84.18
30. 172.16.84.18 is alive
Switch-A>
31. To create VLAN 2 in Switch A, issue the set vlan 2 command on Switch A.
Switch B learns about VLAN 2 after the establishment of the VTP domain in Step 11.
32. Move the management interface in switches A and B to VLAN 2, which you created in
Step 8.
To change the interface, issue the set interface sc0 2 command. This output shows the
issue of the command on Switch A:
Issue the show interfaces command to view the change you just made. This output
shows the issue of the command on Switch A. The output shows the new association of
interface sc0 and VLAN 2:
The ping should fail because the management ports are now in VLAN 2 while the link
between the switches is in VLAN 1.
If you have the output of a show vtp domain command from your Cisco device, you can
use the Output Interpreter Tool (registered customers only) to display potential issues
and fixes.
To configure port 2/1 on Switch A for desirable mode, issue the set trunk 2/1 desirable
isl command on Switch A. Switch B is in auto mode. Switch B automatically places the
port 2/1 in trunking mode after completion of the DTP negotiation between the two
switches.
If you have the output of a show trunk command from your Cisco device, you can use
the Output Interpreter Tool (registered customers only) to display potential issues and
fixes.
This list describes the five different states for which you can configure DTP:
o auto: The port listens for DTP frames from the neighbor switch. If the neighbor
switch indicates that the switch would like to be a trunk, or that the switch is a
trunk, auto state creates the trunk with the neighbor switch. The auto state does
not propagate any intent to become a trunk; auto state is solely dependent on the
neighbor switch to make the trunking decision.
o desirable: DTP is spoken to the neighbor switch to which you want to establish
an ISL trunk. The switch with desirable configuration communicates that the
switch is able to be an ISL trunk and wants the neighbor switch to also be an ISL
trunk. Cisco Technical Support recommends the configuration of desirable
trunking mode on the ports of the trunk link.
o on: DTP is spoken to the neighbor switch. The on state automatically enables ISL
trunking on the port, regardless of the state of the neighbor switch. The port
remains an ISL trunk unless the port receives an ISL packet that explicitly
disables the ISL trunk.
o nonegotiate: DTP is not spoken to the neighbor switch. The nonegotiate state
automatically enables ISL trunking on the port, regardless of the state of the
neighbor switch.
o off: There can be no use of ISL on this port, regardless of the configuration of
DTP mode on the other switch port.
This table shows the 15 possible, unique combinations of DTP modes. The table also
shows whether the combinations result in an active bidirectional trunk. While
theoretically you can trunk in one direction on a link and not in the other direction, you
should not perform this kind of trunking.
36. To verify the trunk link, issue the show trunk 2/1 command at the prompt on Switch A.
37. Switch-A> (enable) show trunk 2/1
38. Port Mode Encapsulation Status Native
VLAN
39. -------- ----------- ------------- ------------
-----------
40. 2/1 desirable isl trunking 1
41.
42. Port VLANs allowed on trunk
43. --------
-----------------------------------------------------------------
----
44. 2/1 2,1002-1005
45.
46. Port VLANs allowed and active in management domain
47. --------
-----------------------------------------------------------------
----
48. 2/1 2,1002-1005
49.
50. Port VLANs in spanning tree forwarding state and not
pruned
51. --------
-----------------------------------------------------------------
----
52. 2/1 2,1002-1005
Switch-A> (enable)
You should now see that trunking is operational.
Note: VLANs 1–1005 are allowable on all trunk ports by default. You can clear VLAN 1
from the list of allowable VLANs. If you remove VLAN 1 from a trunk, the trunk
interface continues to send and receive management traffic, for example, CDP, VTP,
PAgP, and DTP in VLAN 1. You cannot remove VLANs 1002–1005. In CatOS versions
earlier than 5.4(x), you cannot remove VLAN 1 from the trunk.
To limit the VLANs on a trunk in CatOS, clear them. Issue the clear trunk 2/1 1-1001
command. To establish the allowable VLANs on the trunk, issue the set trunk 2/1 2
command on Switch A.
In this output, notice the permission of only VLANs 1 and 2 on this trunk link now:
53. Ping Switch B from Switch A to verify that the switches can communicate with each
other over the trunk link.
54. Switch-A> ping 172.16.84.18
55. 172.16.84.18 is alive
Switch-A>
show port capabilities mod/port —To see the physical status of a port and the port
capabilities.
show trunk mod/port —To see the trunking information for a particular port.
show vtp domain—To display VTP information.
show vlan vlan_number —To see information on a particular VLAN.
show spantree vlan_number —To see the status of the spanning tree for a particular
VLAN.
show interfaces—To display the configuration of sc0 and sl0.
ping—To send an Internet Control Message Protocol (ICMP) echo message to another
IP host.
Note: In switches with several interfaces and VLANs, include the module/port or VLAN
number with the show command to restrict the command output. Use of the ? argument with
show commands displays which commands allow use of the mod/port argument. For example,
the show trunk ? command indicates that show trunk mod/port is allowable.
There is currently no command available to view the module or port capabilities in Cisco
IOS Software. All 10/100BASE-T, 1000BASE-TX, 100BASE-FX, and 1000BASE-
SX/LX/ZX-type Ethernet modules for the Catalyst 6500/6000 series switches support
ISL trunking.
3. Connect the two switch ports together with the Ethernet crossover cable.
In this example, the Switch A 2/1 port connects to the Switch B 2/1 port.
4. To verify that the ports are operational, issue the show interfaces fastethernet 2/1
status command.
In the global configuration mode, issue the command interface vlan 1. When you
configure the IP address, issue the command ip address 172.16.84.17 255.255.255.0 on
Switch A and ip address 172.16.84.17 255.255.255.0 on Switch B. To configure the
switchport 2/1 as a Layer 2 (L2) interface and a member of VLAN 1, issue the
commands interface fastethernet 3/1, switchport, and switchport access vlan 1. For
more information on the configuration of L2 interfaces in Cisco IOS Software, refer to
this document:
6. To verify that the link between switches A and B is not trunking, issue the command
show interfaces fastethernet 2/1 trunk.
7. Switch-A# show interfaces fastethernet 2/1 trunk
8.
9. Port Mode Encapsulation Status Native VLAN
10. Fa2/1 desirable negotiate not-trunking 1
11.
12. Port VLANs allowed on trunk
13. Fa2/1 none
14.
15. Port VLANs allowed and active in management domain
16. Fa2/1 none
17.
18. Port VLANs in spanning tree forwarding state and not
pruned
19. Fa2/1 none
20. Switch-A#
21.
Note: The term Native VLAN in this output indicates the VLAN placement of this port
when the port is not in trunking mode. If you have configured the port for 802.1Q
trunking, the Native VLAN field also indicates the VLAN for which the frames have no
tags; all others have tags. (Conversely, with ISL trunking, every data frame has the
appropriate VLAN identifier.)
22. Ping Switch B from Switch A to verify that the switches can communicate over the link.
23. Switch-A> ping 66.123.210.122
24. Type escape sequence to abort.
25. Sending 5, 100-byte ICMP Echos to 172.16.84.18, timeout is 2
seconds:
26. !!!!!
27. Success rate is 100 percent (5/5), round-trip min/avg/max =
4/21/92 ms
Switch-A>
28. To create VLAN 2 in Switch A, issue the vlan database command and the vlan 2
command on Switch A.
Switch B learns about VLAN 2 after the establishment of the VTP domain in Step 11.
For more information on the configuration of VLANs in Cisco IOS Software code, refer
to this document:
o Configuring VLANs
o Switch-A# vlan database
o Switch-A(vlan)# vlan 2
o VLAN 2 added:
o Name: VLAN0002
o Switch-A(vlan)#exit
o APPLY completed.
o Exiting....
Switch-A#
29. Move the management interface on switches A and B to VLAN 2, which you created in
Step 8.
In Cisco IOS Software mode, there is no sc0 management interface. Therefore, configure
the ports, which connect to host A on Switch A and host B on Switch B, to VLAN 2.
Issue the commands switchport and switchport access vlan 2. Then, use the hosts to
perform the ping tests. For this example, configure IP addresses 172.16.85.1/24 on host
A and 172.16.85.2/24 on host B.
Issue the show interfaces command to view the change you just made. This output
shows the issue of the command on Switch A. The output shows the new association of
interface 2/2 and VLAN 2:
30. Attempt to ping the host that connects to the alternate switch.
The ping should fail because the ports are now in VLAN 2 while the link between the
switches is in VLAN 1.
C:\>
o Configuring VTP
o Switch-A(config)# vtp domain Cookbook
o Changing VTP domain name from NULL to cookbook
o Switch-A(config)#
o
32. Turn on trunking between the switches.
Configure port 2/1 on Switch A for desirable mode. To configure the encapsulation type
and the mode, issue the switchport trunk encapsulation isl command and the
switchport mode dynamic desirable command. Switch B is in auto mode. Switch B
automatically places the port 2/1 in trunking mode after completion of the DTP
negotiation between the two switches.
This list describes the five different states for which you can configure DTP:
o auto: The port listens for DTP frames from the neighbor switch. If the neighbor
switch indicates that the switch would like to be a trunk, or that the switch is a
trunk, auto state creates the trunk with the neighbor switch. The auto state does
not propagate any intent to become a trunk; auto state is solely dependent on the
neighbor switch to make the trunking decision.
o desirable: DTP is spoken to the neighbor switch to which you want to establish
an ISL trunk. The switch with desirable configuration communicates that the
switch is able to be an ISL trunk and wants the neighbor switch to also be an ISL
trunk.
o on: DTP is spoken to the neighbor switch. The on state automatically enables ISL
trunking on the port, regardless of the state of the neighbor switch. The port
remains an ISL trunk unless the port receives an ISL packet that explicitly
disables the ISL trunk. Cisco Technical Support recommends the configuration of
desirable trunking mode on the ports.
o nonegotiate: DTP is not spoken to the neighbor switch. The nonegotiate state
automatically enables ISL trunking on the port, regardless of the state of the
neighbor switch.
o off: There can be no use of ISL on this port, regardless of the configuration of
DTP mode on the other switch port.
This table shows the 15 possible, unique combinations of DTP modes. The table also
shows whether the combinations result in an active bidirectional trunk. While
theoretically you can trunk in one direction on a link and not in the other direction, you
should not perform this kind of trunking.
Note: VLANs 1–1005 are allowable on all trunk ports by default. You can clear VLAN 1
from the list of allowable VLANs. If you remove VLAN 1 from a trunk, the trunk
interface continues to send and receive management traffic, for example, CDP, VTP,
PAgP, and DTP in VLAN 1.
To limit or remove VLANs on a trunk, issue the switchport trunk allowed vlan remove
1-1001 command in interface configuration mode. The command removes all removable
VLANs from the trunk. Issue the switchport trunk allowed vlan add 2 command to add
VLAN 2 to the list of allowable VLANs on the trunk.
In this output, notice the permission of only VLANs 2 and 1002–1005 on the trunk:
48. To verify that the trunk link is operational, attempt to ping across the trunk link.
C:\>
show interfaces fastethernet port/mod trunk—To see the trunking information for a
particular interface.
show vtp status—To display VTP information.
show vlan vlan_number —To see information on a particular VLAN.
show spanning-tree vlan vlan_number —To see the status of the spanning tree for a
particular VLAN.
show interfaces fastethernet port/mod switchport —To see L2 interface information.
ping —To send an ICMP echo message to another IP host.
Related Documents
Best Practices for Catalyst 4500/4000, 5500/5000, and 6500/6000 Series Switches
Running CatOS Configuration and Management
Best Practices for Catalyst 6500/6000 Series and Catalyst 4500/4000 Series Switches
Running Cisco IOS Software
Port Vlans allowed on trunk
-------- ---------------------------------------------------------------
2/24 1-1005
You can see that the trunk did not come up. When you see that kind of issue, check the VTP
domain that is configured on the switches. Issue the show vtp domain command:
Now put switch Sa in VTP domain "cisco" with use of the set vtp domain cisco command.
After a few seconds, the trunk is negotiated and up again:
If you want to keep different VTP domains but still create a trunk between two switches, you
must hard code trunking on each side of the trunk (with use of nonegotiate/on).
When you try to delete the extended-range VLANs from a trunk port with use of the clear trunk
command, this error is sometimes shown on the switch console:
Failed to clear vlans in the extended range Maximum of 64 trunks can have
non-default extended range vlan configuration. Use the 'set trunk' command to
restore
some existing entries to the default value.
Note: The term extended range includes any VLAN from 1025 to 4094. The term default
extended range includes all VLANs from 1025 to 4094. If you try to clear any VLAN in the
range from 1025 to 4094, the VLAN becomes non-default extended range. The maximum
number of trunks which pass non-default extended range is 64. This includes both inactive and
active trunks.
This error and the limitation of 64 trunks come from the NVRAM block which is used to store
nondefault configurations for extended-range VLANs. If you issue the show trunk extended-
range command, you can see all the trunks that are configured with nondefault extended ranges.
By default, the entire configuration is stored in NVRAM. NVRAM has different "blocks" for
saving the nondefault configurations. The blocks are placed into different categories, such as
global or module. The block that holds the nondefault configuration for extended ranges has a
limitation of 64 trunks.
There are two workarounds to reduce the number of nondefault extended-range trunks. The first
method is to set any of the nonactive/unused trunk ports back to the default allowed VLANs.
Use the set trunk mod/port 1025-4094 command. Then the clear trunk mod/port 1025-4094
command should work for the extended VLANs. The second workaround is to change the
configuration mode from binary (default) to text mode. Use the set config mode text command
in order to change the configuration mode to text mode. Text mode typically uses less NVRAM
or Flash memory space than binary configuration mode uses.
Note: When operating in text file configuration mode, most user settings are not immediately
saved to NVRAM; configuration changes are only written to DRAM. You must issue the write
memory command in order to store the configuration in nonvolatile storage. Use the set config
mode text auto-save command in order to save the text configuration in NVRAM
automatically.
This is a common issue that began to be raised to Cisco Technical Support when the first
modules that were able to support both 802.1Q and ISL shipped. People were used to the
configuration of a trunk with use of the set trunk module/port on command or the set trunk
module/port nonegotiate command. The problem is that, by default, the encapsulation type is set
to negotiate. The negotiate encapsulation type is only supported by auto or desirable trunking
modes. The on and nonegotiate encapsulation types do not perform any negotiations between
switches and must be hard set to ISL or 802.1Q encapsulation when they are configured. Here is
a log of what happens on the switch in this case:
This makes sense because if you do not negotiate with the remote, how would you know which
kind of encapsulation (802.1Q or ISL) to use in order to bring up the trunk? There are two
possibilities:
Use the desirable mode. In this case, you negotiate the encapsulation mode with the
remote:
Sa> (enable) set trunk 5/24 desirable
Port(s) 5/24 trunk mode set to desirable.
Sa> (enable) 1997 May 09 17:49:19 %DTP-5-TRUNKPORTON:Port 5/24
has become
isl trunk
ping
set interface
set trunk
set vlan
set vtp domain
show interface
show port
show port capabilities
show trunk
show vtp domain
Related Documents
Best Practices for Catalyst 4500/4000, 5500/5000, and 6500/6000 Series Switches
Running CatOS Configuration and Management
Best Practices for Catalyst 6500/6000 Series and Catalyst 4500/4000 Series Switches
Running Cisco IOS Software
Troubleshooting Catalyst 6500/6000 Series Switches Running CatOS on the Supervisor
Engine and Cisco IOS on the MSFC
Configure EtherChannel Between Catalyst 4500/4000, 5500/5000, and 6500/6000
Switches That Run CatOS System Software
Configuring EtherChannel and 802.1Q Trunking Between Catalyst L2 Fixed
Configuration Switches and Catalyst Switches Running CatOS.
More...
Related Products/Technology
Related Discussion