In our previous post, we discussed how developers can quickly provision application infrastructure using instances and images in OpenStack. Today, we’ll discuss an important topic: Networking! How do we configure the networks that OpenStack instances use to communicate with each other and with the outside world?
VMware Integrated OpenStack provides two networking options for your infrastructure:
- VMware NSX networking
- VMware vSphere Distributed Switch (VDS) networking
The VDS option is appropriate for simple networking use cases. That is, your instances only need to communicate on a few VLANs with no need for advanced functionality like overlapping IP addresses, neutron-provided layer 3 routing, etc. The NSX option allows for advanced networking use cases including private networks for tenants, attaching floating IPs to your instances, etc.
For the purposes of this article, we will focus on the VMware NSX option. Configuring VDS networking is a fairly simple process, and we’ll point out the difference in the configuration process where applicable.
The first step in setting up your OpenStack network service is configuring your external, or provider, network. This is the VLAN provisioned for your instances to have access to the outside world. The external network is configured by a user with administrator permissions using either the Horizon GUI or the neutron API/CLI.
When configuring the external network for the VMware NSX networking option, the provider network type is “Port Group”. The physical network is the port group ID (dvportgroup-50110 in my example) for the external network you defined in vSphere. See Figure 1 for a configuration example.
If you are working with the VDS networking option instead of the VMware NSX option, you specify the provider network type as “VLAN” with the physical network labeled simply “dvs”. The VLAN ID is specified in the Segmentation ID textbox. The “Shared” option must be selected so that your tenants can use this network when booting instances (See Figure 2 for a configuration example). VMware Integrated OpenStack will use this information to automatically create a port group on the VDS that you specified when you deployed OpenStack control plane.
Once your external network is defined, you define a subnet the external OpenStack network. Make sure to uncheck the “Enable DHCP” option, to specify the network address and gateway, and to specify the IP allocation range in case the entire subnet isn’t available for use.
Now that your external network configuration is complete, your tenants can allocate IP addresses for their instances using the OpenStack GUI, APIs, or CLIs as seen in our previous blog post.
The following video provides a detailed walkthrough of configuring OpenStack networks.
Stay tuned for next week’s blog post when we discuss the OpenStack storage service! In the meantime, you can learn more on the VMware Product Walkthrough site and on the VMware Integrated OpenStack product page.