IBM Cloud Docs
Configuring your network to use the customer-managed NSX edge cluster with your VMs

Configuring your network to use the customer-managed NSX edge cluster with your VMs

Configure the network for your virtual machines (VMs) so you can take advantage of the VMware NSX Edge™ cluster that is deployed in your VMware Cloud Foundation for Classic - Automated instances. For more information about the security measures that are in place to help minimize security risk, see Does the management services NSX Edge pose a security risk?

VMware NSX® is a network virtualization platform that allows the virtualization of isolated networks and provides several networking services such as switches, routing, and firewalls. For more information about NSX, see Overview of NSX for vSphere.

As part of the ordering process for your Automated instance, the following actions are completed on your behalf:

  • A private customer subnet is ordered to be used by your VMs to access the IBM Cloud® infrastructure private network.
  • A public customer subnet is ordered to allow your VMs to access the internet.
  • NSX is deployed and configured in your Automated instance.
  • A sample NSX Logical Switch is deployed to be used by the customer workload VMs.
  • A sample router is deployed for potential east-west communication between local workloads that are connected to layer 2 (L2) networks. This router is a NSX-T™ Tier 1 Router for NSX-T and an NSX Distributed Logical Router for NSX-V.
  • An NSX Edge appliance is deployed and configured to perform network address translation (NAT). NAT is done from the range of IP addresses of the workload logical switch to a public IP address on the NAT rules.
  • (NSX-V only) If you installed the Veeam® service, the NSX Manager is configured to do a daily backup of the NSX configurations. For more information, see Veeam 9.5.

Configuring the networking settings for your VMs

To take advantage of NSX for your workload VMs, you must configure a number of settings by completing the following steps when you create your VMs:

  1. Set the network adapter of the VM to the workload logical switch:

    1. On the New Virtual Machine dialog box, click the Customize Hardware tab.
    2. On the new device menu, select Network and click Add.
    3. On the newly added network adapter, select the workload logical switch from the menu. The following examples show the name of the workload logical switch:
      • For NSX-T - overlay-ls
      • For NSX-V - vxw-dvs-17-virtualwire-1-sid-6000-Workload

    Ensure that you do not select the Workload Transit switch.

  2. Identify an available IP address for the VM:

    • The IP address must be in the range of 192.168.10.0/24. The IP address 192.168.10.1 is reserved (see Step 3).
    • When you configure the networking of the operating system that runs on the VM, use the selected IP address and the netmask 255.255.255.0.

    You are responsible for managing the range of IP addresses to which you assigned your VMs.

  3. Assign the default gateway of the VM as 192.168.10.1. For NSX-T, this address is the IP address of the downlink router port on the customer Tier 1 logical router. The address is connected to the same overlay logical switch as the workload VMs. For NSX-V, this address is the IP address of the NSX DLR on the same logical switch as the workload VMs.

Enabling the SNAT rule for NSX-T

NSX-T enables the SNAT rule by default. For more information about modifying the existing rules, see Configure source and destination NAT on a Tier-0 logical router.

Enabling the firewall rule for NSX-T

NSX-T disables all traffic through the workload NSX edge rule by default. To allow VMs to use the SNAT rule described in the previous section, you must create a firewall policy and rule to define and allow the traffic. For more information about modifying the existing rules, see Add a Gateway Firewall Policy and Rule.

Enabling the SNAT rule for NSX-V

If you want your workload VMs to have outbound access to the internet, you must enable the associated SNAT (Source Network Address Translation) rule. Enabling the SNAT rule allows internet access from your VMs to be translated to a single public IP address. Complete the following steps in the VMware vSphere® Web Client:

  1. Click Home > Networking & Security.
  2. On the navigator pane, click NSX Edges and double-click the edge that is named customer-nsx-edge.
  3. Click Management > NAT to open the NAT tab.
  4. Select the default SNAT rule in the table and click the green checkmark next to the table to enable the rule.

For more information about NSX Edge NAT rules, see Managing NAT rules.

Identifying customer subnets details for NSX-T

The logical routers <instance_name>-<podname>-<cluster_name>-T1-workload and <instance_name>-<podname>-<cluster_name>-T0-workload, and the edges cust-edge0 and cust-edge1 are intended for your own usage. You can modify them to define more NAT rules for inbound or outbound traffic. These rules must use only the IP addresses on the public or private customer subnets that are ordered on your behalf.

To identify the details for the customer subnets so you can use the IP addresses ordered, complete the following steps in the NSX-T Web Client:

  1. From the VMware Solutions console, click Resources > VCF for Classic from the left navigation pane.
  2. In the VMware Cloud Foundation for Classic table, locate and select the instance.
  3. Click the Infrastructure tab and select the management cluster.
  4. Go to the Network interface section on the cluster page.
  5. Collapse the Public VLAN and Private VLAN options. The public and private customer subnets are the ones that have customer workload edge in the Description field.

Additionally, you can find more details about the customer subnets by completing the following steps in the IBM Cloud infrastructure customer portal:

  1. Click Networking > IP Management > Subnets.

  2. Click the filter menu and in the Subnet field enter the identifier as seen in the description of cust-edge0 in the NSX-T Web Client.

  3. Review the notes that are shown for the IP addresses. These notes identify which of the subnets and IP addresses are ordered and used during the initial setup.

    Do not use the IP addresses that are ordered and used during the initial setup. However, you can use other IP addresses on these subnets according to your requirements. To set up more network address translation rules, see Managing NAT rules.

Identifying customer subnets details for NSX-V

The edge customer-nsx-edge is intended for your own usage, so you can modify it to define more NAT rules for inbound or outbound traffic. These rules must use only the IP addresses on the public or private customer subnets that are ordered on your behalf.

To identify the details for the customer subnets so you can use the IP addresses ordered, complete the following steps in the VMware vSphere Web Client:

  1. Click Home > Networking & Security.
  2. On the navigator pane, click NSX Edges and locate customer-nsx-edge edge in the list of edges in the right pane.
  3. In the Description column, hover over the edge description for customer-nsx-edge to see the subnet identifiers for both the private and public customer subnets.

Additionally, you can find more details about the customer subnets by completing the following steps in the IBM Cloud infrastructure customer portal:

  1. Click Networking > IP Management > Subnets.

  2. Click the filter menu and in the Subnet field enter the identifier as seen in the description of customer-nsx-edge in the VMware vSphere Web Client.

  3. Review the notes that are shown for the IP addresses. These notes identify which of the subnets and IP addresses are ordered and used during the initial setup.

    Do not use the IP addresses that are ordered and used during the initial setup. However, you can use other IP addresses on these subnets according to your requirements. To set up more network address translation rules, see Managing NAT rules.

Ordering more subnets

You can order more subnets if the provided subnets do not fulfill your requirements for IP addresses.

  • For NSX-T, you can order more public or private static subnets and configure them with the Tier-0 VIP for the subnet's endpoint.
  • For NSX-V, you can order more public or private portable subnets.