Communication Workflow with CipherTrust Manager Behind a Load Balancer
When a load balancer is configured in front of the CipherTrust Manager cluster nodes, the following workflow illustrates how requests and responses flow. Consider a 4-node CipherTrust Manager cluster, where a user browses an agent's filesystem via the CipherTrust Manager GUI. The request is routed to Node1 through the load balancer, while the agent is connected to Node2. Here’s how the process unfolds:
Initiating the Request: The user selects the "Enter/Browse Path" option to browse a directory on the agent, and the request is directed to Node1.
Notifying Other Cluster Nodes: Node1 broadcasts the request to all nodes in the cluster, storing the details in the database. This ensures all nodes are aware of the request.
Handling the Request: Each node processes the notification. Only the node connected to the agent (Node2 in this case) takes action; the others disregard the request.
Retrieving the Data: Node2 fetches the requested data from the agent, records it in the database, and notifies all nodes that the request has been completed.
Notifying the User: Node1, which initially received the request, retrieves the data and delivers the results to the user’s interface.
This system ensures that regardless of where a request originates, all nodes work together to fulfill the request. The process remains efficient and seamless with the load balancer directing traffic across the CipherTrust Manager cluster nodes.
