Replace a Node

Use this procedure if you want to replace a failed node with a new node. The node replacement process removes the failed node from the cluster and inserts the new node in its place.

Note

This procedure applies to multi-node clusters only.

For the failed node (the node that's being replaced), do the following:
1. If you want the new node (the replacement node) to use the same IP settings as the failed node (the node that's being replaced), take a record of the following network settings for the failed node:
  - ICC IP Address—Unlike the other network settings, the new node must use a different ICC IP address than the failed node.
  - Data port IP addresses
  - DNS server addresses
  - NTP server addresses
  Note
  If the failed node is unreachable, you can get the ICC IP address for the failed node from the Cluster Settings page of the other cluster nodes.
2. Remove the failed node from the network so that the node has no connectivity to the other nodes. For example, you could shut the node down or remove the network cables.
For the new node (the replacement node), follow the procedures in the ExtremeCloud Edge - Managed Orchestration Deployment Guide to configure the new node. Do the following:
1. Run the Basic Configuration Wizard on the new node and assign network settings.
  
  Note
  Make sure to assign an ICC IP address that's different than the ICC IP address of the failed node.
2. Configure VRRP on the new node to use a lower priority than the other cluster nodes.
  
  Note
  To ensure that the replacement node successfully joins the cluster, set the VRRP node priority of the replacement node to a value that is lower than the value of the existing nodes. This ensures that the VRRP address is pointing at a working node in the cluster during the joining process. After the replacement node has joined the cluster, you can set the VRRP node priority to first priority if desired, but this is not required.
3. Upgrade the new node to use the same software version that's used by the rest of the cluster.
4. Install the new node on the network so that the node is powered on with connectivity to the rest of the cluster.
From the primary node for the existing cluster, start the node replacement process:
1. Go to Cluster Settings > Node Replacement.
2. Select Next
3. For Select Failed Node, select the ICC IP address of the failed node.
4. For New Node ICC IP Address, enter the ICC IP address of the replacement node (the new node).
5. Select Replace Node.
6. Select OK.