Importing Existing VXLAN Fabric with NDFC

Hello and welcome to importing an existing fabric video. In this video, you will learn how to import an existing VXLAN fabric using Nexus dashboard fabric controller. You will begin with create the fabric on the MDFC and then discover the existing switches in your deployment. So let's get started. Your company has deployed a data center using Nexus nine k switches. Your team has been configuring the network infrastructure using CLI, which is complex and time consuming. A decision has been made to import the network into the Nexus dashboard fabric controller to provide a single point of provisioning and monitoring. You as a network administrator has been tasked to use the fabric controller to import the existing fabric. You will review the existing fabric configuration using the CLI. Then you will create a fabric on the NDFC followed by discovering and adding switches to the fabric. Set the switch roles, then you recalculate, preview, and deploy any new configuration if needed. And finally, check the imported configuration on the fabric controller. You will start by reviewing the existing configuration on the switches. Use the terminal of your choice to log in to the switches in your fabric. You can start by checking the existing BGP configuration. Issue the command show run section BGP. Notice BGP is configured with autonomous system number of sixty five thousand and two. You also have VXLAN pre configured with BGP EVPN as the control plane. To check that, you can issue the command show run section in VE. Next, let's verify that the fabric is healthy by testing the connectivity between the endpoints. In our case, we have server two connected to leaf two and server three connected to leaf three. Each is configured with a different network. You can verify the layer three connectivity between the servers by using the ping command. And as you can see, there is a layer three connectivity, which means the fabric is healthy. Finally, you can check that the BGP EVPN control plane is working as expected by issuing the command show BGP layer two VPN EVPN. Notice that the address of both servers are learned successfully on the leaf switches. So the fabric was configured via CLI. However, it's a time consuming process to do all of these configs, especially if you are looking to scale in the future. That's why you can use the Nexus dashboard fabric controller to centralize the provisioning and monitoring process by importing the existing fabric. You are now on the NexSys dashboard fabric controller user interface. To import the switches to the NDFC, first, you will need to create the fabric. To do that, navigate to the LAN, then select fabrics, and from the actions drop down menu, select create fabric. Enter the fabric name. Let's call it existing site one. Then click on choose a fabric and from the drop down list, select data center VXLAN EVPN fabric using Nexus nine ks or three ks switches, whether it's a new deployment or an existing one. The fabric settings for creating a standalone fabric will appear. You will need to enter the parameters that match the existing fabric configuration. For example, the BGP autonomous system number that is configured on the fabric is sixty five thousand and two. So under the general parameters in the BGP ASN field, you will need to enter the same. Same idea applies to the underlay subnet ranges used in your existing deployment. To modify the underlay subnets, navigate to resources where you can edit the subnets as per your need. In our case, we're using subnet twenty in the underlay, so I will adjust the configuration to match what I have in my existing deployment. Once you're done, click on save. Next, you will need to discover switches and add them to the fabric. To do that, double click the fabric name you created to display the fabric details. In the fabric overview page, click on the actions, then from the drop down menu, select add switches. In the add switch dialog, in the seed IP field, enter the IP of the seed switch, which is one of the switches that will act as a starting point to discover the rest of the fabric. Provide the credentials used to authenticate with your switch. These credentials are pre created locally on the switch itself. You can specify the maximum number of hops between the seed and the near switches that will be discovered as well. And finally, make sure that the PreserveConfigToggle is set to Yes. This will make sure the existing configuration will remain and the NDFC will not erase the switch config. Then click on discover switches. After a few seconds, you get a list of the discovered switches using the provided parameters. Select the switches to be added to the fabric, then click on add switches. Once the discovery is completed, make sure that the status is switch added, Then click on Close. Next, you need to assign the switch roles. By default, the Fabric Controller requires to know which switches are supposed to be Leafs, Spines, or Border Gateways, And that's why you need to assign the Switch roles. To do that, check the switches, then from the Actions dropdown menu, select Set Row. From the Row list, choose the role as per your own needs, then click on select. When you receive a warning message about performing a recalculate config, click on okay. Repeat the same process for the rest of the switches as per your design, and you can verify the current switch role by looking at the role column. Notice that the switches are in the migration mode, which means the NDFC is in the process right now of reading the current configuration on the switches. After selecting the switch roles, you need to recalculate the new configuration. This will help the NDFC to identify the changes you performed and generate the CLI config that will be pushed down to the switches. To do that, from the actions drop down list, select recalculate config. In a few moments, the deploy configuration window will appear. You can see that some of the switches are out of sync, which means there is a pending configuration to be provisioned to these switches. You can click on the pending config to review the configuration. Once you are done with the review process, click on close, then choose deploy to start the provisioning process. When the deployment completes, notice the status is success. Then click on close. Now your fabric is successfully imported to the NDFC. Note that when you import an existing fabric, the NDFC will also import the networks and the VRFs. To verify that, you can navigate to the VRFs. Notice a VRF has been detected. And also you can navigate to the networks where you can also see the existing networks created on the fabric. Finally, you can test the connectivity once again between the servers to make sure that nothing is affected. As you can see, layer three connectivity remains functional after the importing process. This concludes our Cisco technical education video on importing an existing VXLAN fabric with NDFC. Thank you for watching.