You can change the IP address of an NSX Manager in an NSX Manager cluster. This section describes several approaches.

For example, if you have a cluster consisting of Manager A, Manager B, and Manager C, you can change the IP address of one or more of the managers in the following ways:
  • Scenario A:
    • Manager A has IP address 172.16.1.11.
    • Manager B has IP address 172.16.1.12.
    • Manager C has IP address 172.16.1.13.
    • Add Manager D with a new IP address, for example, 192.168.55.11.
    • Remove Manager A.
    • Add Manager E with a new IP address, for example, 192.168.55.12.
    • Remove Manager B.
    • Add Manager F with a new IP address, for example, 192.168.55.13.
    • Remove Manager C.
  • Scenario B:
    • Manager A has IP address 172.16.1.11.
    • Manager B has IP address 172.16.1.12.
    • Manager C has IP address 172.16.1.13.
    • Add Manager D with a new IP address, for example, 192.168.55.11.
    • Add Manager E with a new IP address, for example, 192.168.55.12.
    • Add Manager F with a new IP address, for example, 192.168.55.13.
    • Remove Manager A, Manager B, and Manager C.
  • Scenario C:
    • Manager A has IP address 172.16.1.11.
    • Manager B has IP address 172.16.1.12.
    • Manager C has IP address 172.16.1.13.
    • Remove Manager A.
    • Add Manager D with a new IP address, for example, 192.168.55.11.
    • Remove Manager B.
    • Add Manager E with a new IP address, for example, 192.168.55.12.
    • Remove Manager C.
    • Add Manager F with a new IP address, for example, 192.168.55.13.

The first two scenarios require additional virtual RAM, CPU, and disk for the additional NSX Managers during this IP address change.

Scenario C is not recommended because it temporarily reduces the number of NSX Managers and a loss of one of the two active managers during the IP address change will have an impact on the operations of NSX. This scenario is for a situation where additional virtual RAM, CPU and disk are not available and an IP address change is required.

Note: If you are using the cluster VIP feature, you must either use the same subnet for the new IP addresses or deactivate the cluster VIP during the IP address changes because the cluster VIP requires all NSX Managers to be in the same subnet.

Prerequisites

Familiarize yourself with how to deploy an NSX Manager into a cluster. For more information, see the NSX Installation Guide.

Procedure

  1. If the NSX Manager you want to remove was deployed manually, perform the following steps.
    1. Run the following CLI command to detach the NSX Manager from the cluster.
      detach node <node-id>
    2. Delete the NSX Manager VM.
  2. If the NSX Manager you want to delete was deployed automatically through the NSX Manager UI, perform the following steps.
    1. From your browser, log in with administrator privileges to an NSX Manager at https://nsx-manager-ip-address.
      This NSX Manager must not be the one that you want to delete.
    2. From the Systems tab, click NSX Management Nodes.
      The status of the NSX Manager cluster is displayed.
    3. For the NSX Manager that you want to delete, click the gear icon and select Delete.
  3. Deploy a new NSX Manager.
  4. If you have an NSX Manager cluster that is onboarded to the NSX+ Intelligence service or the NSX+ NDR service, activate the maintenance mode for the NSX+ Intelligence and the NSX+ NDR agents.
    1. To check whether the NSX Manager cluster is onboarded to either the NSX+ Intelligence service or the NSX+ NDR service, use the following API request using the IP address of the restored NSX Manager. 
      GET https://nsx-manager-ip-address/policy/api/v1/infra/sites/agents/intelligence/maintenance
      If the NSX Manager site is not onboarded, the API request returns the following message. In this case, no further action is required. 
      {
"enable": true.
"agent_error_message"; "Site is not onboarded with Saas. Invalid operation."
}
      If the NSX Manager site is onboarded, the API request returns the following message. Continue with the next steps below. 
      {
"enable": false
}
    2. After you complete the process to change the IP address of the NSX Manager node for all three nodes in the NSX Manager cluster and you confirmed that the NSX Manager site is onboarded, activate the maintenance mode for the NSX+ Intelligence and the NSX+ NDR agents using the following API request. 
      PUT https://nsx-manager-ip-address/policy/api/v1/infra/sites/agents/intelligence/maintenance
{
"enable": true
}
    3. Wait for all the NSX Manager node objects to be in the REALIZED state. Use the following API call to check. 
      GET https://nsx-manager-ip-address/policy/api/v1/infra/realized-state/realized-entities?intent_path=/infra/sites/agents/intelligence
      In the API call output, ensure that the output "state": "REALIZED" exists for all the objects in the list.
    4. Log in to any of the NSX Manager in the cluster and from the command line, clear the NsxiAgentDockerConfig table using the following command. 
      /opt/vmware/bin/corfu_tool_runner.py -n nsx -o clearTable -t NsxiAgentDockerConfig
    5. Deactivate the maintenance mode using the following API request. 
      PUT https://nsx-manager-ip-address/policy/api/v1/infra/sites/agents/intelligence/maintenance
{
"enable": false
}