If any of the node in the cluster is not working or is down, it is recommended to replace the node with a new node. The goal here is to replace the node while other nodes in the cluster are still processing the web-service requests and maintain the cluster along with keeping the same software stack.

PREREQUISITES

1. Ability to switch traffic in Load Balancer if the node being replaced is actively processing web-service requests.
2. Networking related files (IP address, FQDN, /etc/hosts) configured.
3. Credentials for the strongkey user in the Node to be replaced.
4. Credentials for the root and strongkey user in the new Node.
5. Credentials for MariaDB users (such as “skfsdbuser”) with access to the “skfs” database in the new and old Node.
   
6. Credentials for a pinguser to verify the functionality.
7. Ability to transfer files between old and new servers (either using USB flash drive or scp/sftp).