Category Archives: CCIE Data Center

CCIE DC – 1.1.a Link Aggregation – Configure Virtual Port Channels

CCIE Data Center, ,

This post examines the limitations you should be aware of when configuring virtual port channels. It also examines the configuration steps. This is part of the link aggregation section of the exam requirements in CCIE Data Center.
Aggregation

Our previous post on this subject focused on the control plane. You can find it here: https://www.ajsnetworking.com/ccie-dc-written-1-1/

Limitations

Before you dive right into the configuration steps, you should be aware of the limitations when it comes to your virtual port channels. These include:

  • In the “standard” virtual port channel – you use exactly two vPC peer switches – no more than that are supported. A later post here will cover the extended vPC technology, but this post does not consider that.
  • It is not possible for one of your peer vPC devices to participate in more than one vPC.
  • All ports for a given vPC must exist in the same VDC (Virtual Device Context). You cannot span a vPC across multiple VDCs.
  • A 10 Gbps link is required for the vPC peer link. Typically, you should use at least two links in a port channel for this component.
  • A vPC is a Layer 2 construct. There is no support for the creation of Layer 3 vPCs.

The Configuration Steps

As you might guess, you should double-check all of your physical connections before launching into the vPC configuration. Ensure that your physical link(s) for the peer link are in place and healthy. Also, keep in mind that you make these configurations on each of the vPC peer devices.

Step 1. First, you must enable the vPC feature – you do this with the command: feature vpc

Step 2. Create the vPC using the vpc domain <domain_id> command

Step 3. Next, specify your peer keepalive link – this link can be in any VRF (including Management); use the command – peer-keepalive destination <remote_peer_ip> source <local_peer_ip> vrf <vrf_name>

Step 4. Configure your peer link. This is typically done with a port channel (as described above) as follows:

interface port-channel <port_channel_id>
switchport mode trunk
vpc peer-link

Step 5. Configure a port channel that leads to the downstream device; use the command vpc <domain_id> under the port channel

In the next post, I will walk you through a configuration on live equipment. We will also walk through the important vPC verifications you would want to perform at that time. As always, thanks for reading.

For even more information – check out this Cisco documentation at https://www.cisco.com/c/en/us/products/collateral/switches/nexus-5000-series-switches/design_guide_c07-625857.html
Pearson Education (InformIT)

CCIE DC – 1.1.a Link Aggregation – vPC Control Plane

CCIE Data Center

Enjoy this latest post on the control plane functions of the vPC. This is part of the link aggregation section of the exam requirements in CCIE Data Center.
Aggregation

Our previous post on this subject focused on the data plane. You can find it here: https://www.ajsnetworking.com/ccie-dc-written-1-1-a-link-aggregation-vpc-data-plane/ 

The vPC Control Plane

The vPC peer link between your vPC peers is the workhorse for the control plane messages (BPDUs, LACP, CFSoE) that must flow between them. The control plane is responsible for the following:

  • Exchange of Layer 2 forwarding tables for the vPC peers – MAC addresses learned on one switch are instantly synced to the other switch using CFSoE; this reduces traffic on the vPC peer link
  • Consistency and compatibility checks – CFSoE exchanges information to ensure both peer support vPC and that he member ports are in a compatible state
  • Sync IGMP snooping information – inbound IGMP traffic triggers synchronization much like MAC address learning
  • Monitor the status of vPC member ports – orphan port behavior may occur due to failure conditions
  • Sync the ARP tables of the vPC peers – again CFSoE is used to keep these important tables in sync; this speeds convergence if one of the vPC peers must reload
  • Determine the primary and secondary vPC peer devices – the primary is responsible for the processing of BPDUs; the election is non-preemptive
  • Agree on LACP and STP parameters – this is needed because the vPC announces itself to the downstream switch as a single logical switch

For even more information – check out this Cisco documentation at https://www.cisco.com/c/en/us/products/collateral/switches/nexus-5000-series-switches/design_guide_c07-625857.html
Pearson Education (InformIT)