CCNP Switch Lab 3-2 Modifying Default Spanning Tree Behavior

CCNP Switch Lab 3-2, Modifying Default Spanning Tree Behavior

Topology
ccnp-switch-lab-modifying-default-spanning-tree-behavior

Objective

  •  Observe what happens when the default spanning tree behavior is modified.

Background
Four switches have just been installed. The distribution layer switches are Catalyst 3560s, and the access layer switches are Catalyst 2960s. There are redundant uplinks between the access layer and distribution layer. Because of the possibility of bridging loops, spanning tree logically removes any redundant links. In this lab, you will see what happens when the default spanning tree behavior is modified.

Note: This lab uses Cisco WS-C2960-24TT-L switches with the Cisco IOS image c2960-lanbasek9-mz.122- 46.SE.bin and Catalyst 3560-24PS switches with the Cisco IOS image c3560-advipservicesk9-mz.122- 46.SE.bin. Other switches (such as a 2950 or 3550) and Cisco IOS Software versions can be used if they have comparable capabilities and features. Depending on the switch model and Cisco IOS Software version, the commands available and output produced might vary from what is shown in this lab.

Required Resources

  • 2 switches (Cisco 2960 with the Cisco IOS Release 12.2(46)SE C2960-LANBASEK9-M image or comparable)
  • 2 switches (Cisco 3560 with the Cisco IOS Release 12.2(46)SE C3560-ADVIPSERVICESK9-M image or comparable)
  • 1 PC (optional) attached to switch ALS1.
  • Ethernet and console cables

Note: Configuring PortFast in Step 5 requires a PC attached to one of the access switches.

Step 1 : Prepare the switches for the lab.

a. Delete vlan.dat, erase the startup configuration, and reload all switches. You can find detailed instructions in Lab 1 -1 or 1 -2.

b. Give each switch a hostname according to the topology diagram.

c. Configure ports Fa0/7 through Fa0/12 on all switches to be trunks. On the 3560s, first set the trunk encapsulation to dot1q. On the 2960s, only dot1q is supported, therefore the switchport trunk encapsulation command is unavailable, but the mode still needs to be changed to trunk. If you do not set the mode of the ports to trunk, they will negotiate the operational mode according to their default DTP settings.

Note: The default mode on a 3560 or 2960 is dynamic auto; the default mode on a 3550 or 2950 is dynamic desirable.

DLS1 example:

Step 2: Display default spanning tree information for all switches.

a. Use the show spanning-tree command to check how the non-configured switches created a spanning tree. Verify which switch became the root bridge. In the topology used in this lab, DLS2 is the root bridge.
ccnp-switch-lab-modifying-default-spanning-tree-behavior-2a
ccnp-switch-lab-modifying-default-spanning-tree-behavior-2a-1
ccnp-switch-lab-modifying-default-spanning-tree-behavior-2a-2
ccnp-switch-lab-modifying-default-spanning-tree-behavior-2a-3
ccnp-switch-lab-modifying-default-spanning-tree-behavior-2a-4
b. If you receive the following message “No spanning tree instance exists”, issue the no shutdown command on all interfaces.

Now that the switch is communicating with the other switches in the topology, you should receive spanning tree output.

c. Issue the show interfaces trunk command on DLS1 to verify the trunking mode, encapsulation and status for the trunk links.
ccnp-switch-lab-modifying-default-spanning-tree-behavior-2c
Switches connected with an access link still send and receive BPDU’s. However, VTP does not work unless
trunk links are established. CDP works over access links as well. DTP is sent over dynamic auto, dynamic
desirable and static trunk ports only

Step 3: Configure specific switches to be primary and secondary root.

In this step you configure other switches to be the primary root and secondary root. Because DLS2 is the root switch in this topology, this lab changes DLS1 to be the primary root and ALS1 to be the secondary. Do the same in your topology, regardless of which switch is the initial root. On one of the switches that you are not changing, you can use the debug spanning-tree events command to monitor topology changes. To change the spanning tree root status, use the global configuration commands spanning-tree vlan vlan_number root primary and spanning-tree vlan vlan_number root secondary. On a switch that you are not going to be modifying, issue the debug command and then watch the output.

a. Issue the debug command on DLS2.

DLS2# debug spanning-tree events
Spanning Tree event debugging is on

b. Change DLS1 to be the primary root switch.

DLS1(config)# spanning-tree vlan 1 root primary

c. Change ALS1 to the secondary root.

ALS1(config)# spanning-tree vlan 1 root secondary

You can see the topology changes on the switch that you enabled debugging on (your output may vary depending on your initial topology):

Notice the timestamps on the debugs to see the difference between changes caused by the commands done in both steps.

d. Display the running config on the new root switches, DLS1 and ALS1.

Notice the spanning tree commands in the running configuration. You see a different command than the one you entered. This is because spanning-tree vlan vlan_number root is a command that sets the priority number on that VLAN automatically rather than typing in a specific priority number. The priority number of a VLAN can be between 0 and 61440 in increments of 4096. To manually set the specific priority number, use the spanning-tree vlan vlan_number priority priority_number command.

The command spanning-tree vlan vlan_number root primary sets the priority to 24576 instead of the default (32768). The command spanning-tree vlan vlan_number root secondary sets the priority to 28672. Given this information, would a lower or higher priority number result in a switch becoming the root bridge?
__________________________________________________________________________________
__________________________________________________________________________________
A lower priority would cause a switch to take over as root bridge.

e. You can also view the priority modification with the show spanning-tree command:

ccnp-switch-lab-modifying-default-spanning-tree-behavior-3e

Step 4: Change the root port using the spanning-tree port-priority command.

With spanning tree, you can also modify port priorities to determine which ports are forwarding and which are blocking. To choose which port becomes the root on a non-root switch when faced with equal-cost redundant root paths via the same neighbor, the switch looks at the port priorities first. If the sender port priorities are the same, the switch picks the port that receives BPDUs with the lowest sender port number. On the link between DLS1 and DLS2, the default forwarding port is Fa0/11 because it is lower, and the default blocking port is Fa0/12 because it is higher. The two ports have equal costs because they have the same speed.

a. You can verify this using the show spanning-tree command on the non-root switch, which is DLS2.

ccnp-switch-lab-modifying-default-spanning-tree-behavior-4a

b. For comparison, issue the show spanning-tree command on DLS1. Notice that all ports are forwarding because it is the root switch.
ccnp-switch-lab-modifying-default-spanning-tree-behavior-4b

Port priorities range from 0 to 240, in increments of 16. The default priority is 128, and a lower priority is preferred. To change port priorities, change them on the switch closer to the root.

c. To make DLS2 Fa0/12 the root port, and Fa0/11 block, change the port priority on DLS1 with the interface-level command spanning-tree port-priority priority.

d. Issue the show spanning-tree command to verify which port is blocking on DLS2.

ccnp-switch-lab-modifying-default-spanning-tree-behavior-4d
On DLS2, although the root port has changed, the port priorities have not. On DLS1, you can see the port priorities have changed, although all ports are still forwarding (because it is the root switch).
ccnp-switch-lab-modifying-default-spanning-tree-behavior-4d-1
Using the above output, how does DLS2 know which port to change to the root port, without changing the port priorities on DLS2?

DLS2 receives BPDUs from DLS1 and adjusts accordingly.

Step 5: Configure PortFast on an access port.

a. (Optional) If you have a host attached to ASL1 Fa0/6 you can perform this step. If not, read through the following information to see how a port goes through the spanning tree states with and without PortFast enabled.

Another feature of spanning tree is PortFast. PortFast allows you to bypass the normal states of IEEE 802.1D spanning tree and move a port to the forwarding state as soon as it is turned on. This is useful when connecting hosts to a switch, because they can start communicating on the VLAN instantly rather than waiting for spanning tree. There is no danger of creating a spanning tree loop because you are not connecting another switch. A client that runs DHCP as soon as it starts up benefits, because the DHCP requests could be ignored if the port was not in the spanning tree forwarding state. PortFast must be used carefully to avoid inadvertently creating spanning tree loops.

b. Ensure that the port to which the host is attached (Fa0/6) on ALS1 is shut down initially.

c. Enable spanning tree debugging on ALS1 .

d. Set port Fa0/6 switchport mode to access, enable the port and observe the debug output. Notice what happens when the port is brought up. Your output may vary.

e. Shut down the port for the next part.

f. Activate PortFast on the port with the interface-level command spanning-tree portfast. The switch warns you about the possibility of creating switching loops.

g. Now, bring up the port by issuing the no shutdown command on the interface.

h. Be sure to turn off debugging before continuing:

Why could enabling portfast on redundant switch access links be a bad idea?
__________________________________________________________________________________
__________________________________________________________________________________
A spanning-tree loop could occur. This could cause many problems, including broadcast storms.

Note: The spanning-tree portfast trunk interface-level command can be useful if a trunk is being connected to a router or a server. If RSTP is used, both trunk and access links can be moved to a forwarding state rapidly. The spanning-tree portfast trunk command is to be used only on trunks connected to non-switching devices.

Step 6: Change root port using the spanning-tree cost command.

Another way of changing which port becomes the root is to modify the port costs using the interface command spanning-tree cost cost. The default cost is 4 for a gigabit Ethernet port, 19 for a Fast Ethernet port, and 100 for a 10baseT Ethernet port. Lower cost is preferred.

Note: Each port has a default cost value based on a guideline established as part of IEEE 802.1d. In the original specification, the cost of a port cost is calculated as 1,000 Mbps (1 gigabit per second) divided by the bandwidth at which the port is functioning. A 10 Mbps connection have a cost of (1,000/10) or 100. As the speed of networks has increased beyond gigabit, the standard cost has been modified somewhat. The new cost values are:

Bandwidth STP Cost

Bandwidth STP Cost
4 Mbps 250
10 Mbps 100
16 Mbps 62
45 Mbps 39
100 Mbps 1 9
155 Mbps 14
622 Mbps 6
1 Gbps 4
10 Gbps 2

a. For this scenario, change the cost of port Fa0/1 0 on ALS2. First, look at the current port costs using the show spanning-tree command.

Note: The cost shown here is for the port. The root bridge path cost is the sum of link port costs between a switch and the root bridge. The cost of traversing this path is the sum of the costs of the segments on the path. This determines how far away the root bridge is.

ccnp-switch-lab-modifying-default-spanning-tree-behavior-6a

b. Change the port cost for Fa0/10 on ALS2 to 10 and then issue the show spanning-tree command.
ccnp-switch-lab-modifying-default-spanning-tree-behavior-6b
Device Configurations (Instructor version)

Switch DLS1

Switch DLS2

Switch ALS1

Switch ALS2

More Resources

About the author

Prasanna

Leave a Comment