Building Error-Free Layer 2 EtherChannels

Introduction

Have you ever struggled with an EtherChannel on Cisco switches? It can really test someone’s patience. You might even correct the misconfiguration and reboot each device only to see that physical ports are still error-disabled. This is the type of self-induced troubleshooting that we need to avoid like the plague in the CCIE Lab exam. There is just not going to be time for us to waste wrestling with issues like I have described here.

In this particular post, let me demonstrate my fool proof method and “configuration cadence” for building my Layer 2 EtherChannels in the CCIE lab exam environment.

Questions

Before you dive right in and start playing with your Layer 2 infrastructure with EtherChannels, you want to ensure that you have the answers to the following questions readily available:

• What are the device names involved?
• What are the specific ports in the bundle?
• What is the exact required encapsulation and Dynamic Trunk Port modes for the member ports?
• What is the required aggregation protocol or is this a static EtherChannel configuration?

Once you have this information documented, you are ready to make your configurations. Again, here we are going to slow down a little bit, so that we can speed up later. Take your time to get the EtherChannels correct up front, and save yourself brutal troubleshooting time later on. Also, remember how the upper layers of the network are going to rely on the configurations we make down here at Layer 2.

The Steps

What are my steps then for the Layer 2 EtherChannel? Here they are:

1. On DeviceA, enter interface range configuration mode for the correct ports in the bundle. Remember that with interface range config mode, the ports do not have to be continuous. You can use a space, then a comma, then a space in order to specify discontiguous ports.
2. Issue the command shutdown. This is going to ensure that we do not introduce errors and conflicts in our configuration. It is critical to start this way if you want to ensure there are no problems to troubleshoot.
3. Enter the commands that are required to make the ports “twins”. This is a great time to enter commands like speed and duplex and your trunk encapsulation and DTP mode.
4. Issue the do show ip int brief command (or a command like it) to check for any existing EtherChannels (PortChannel interfaces). You certainly do not want to risk overwriting any existing configuration that needs to be in place in your Layer 2 infrastructure.
5. Use the channel-group X mode {active | on | desirable} command as necessary. The X in this command is a locally significant value, but again, you want to ensure you do not overwrite some existing PortChannel. Notice also how this command will create the PortChannel automatically for you. This logical port will inherit the initial properties assigned to the physical interfaces.
6. Visit DeviceB and put the appropriate “mirrored” commands in place.
7. Revisit DeviceA and issue the no shutdown command for the range of interfaces that are to participate in the EtherChannel.
8. Verification is simple, watch the physical interfaces reach an UP/UP state and watch the PortChannel interface come to an UP/UP state as well.
9. A final verification can be to use show interface trunk to ensure you have the trunk acting appropriately.

In an upcoming post here, we will walk through the steps on a Layer 3 EtherChannel and have some fun with it at the command line as well.

Anthony Sequeira, CCIE, CCSI
Twitter: compsolv
Facebook: http://www.facebook.com/compsolv

Print Friendly

Tags: CCIE, CCIE Routing and Switching, free ccie training