In multi node setup how the flow of data starting from login to Orchestrator through load balancer till HAA nodes works

News Knowledge Base
1

In multi node setup how the flow of data starting from login to Orchestrator through load balancer till HAA nodes works?

Below is the high level steps of data flow starting from login to Orchestrator through load balancer till HAA nodes works in multi node setup with HAA-

  • User Login:
    • A user attempts to log in to the Orchestrator portal.
  • Load Balancer:
    • The request from the user hits the load balancer first. The load balancer's role is to distribute incoming requests across multiple Orchestrator nodes in a balanced manner.
    • The load balancer routes the request to one of the Orchestrator nodes in the cluster based on its load balancing algorithm. The choice of node aims to distribute the load evenly across all nodes.
  • Orchestrator Node:
    • The selected Orchestrator node receives the login request.
    • The Orchestrator node processes the login request and interacts with the Orchestrator database to verify user credentials.
  • Session Management:
    • If session persistence (sticky sessions) is configured, the load balancer ensures that subsequent requests from the same user are directed to the same Orchestrator node. This maintains session state for the user's interaction. Uipath recommendation is to not use the sticky session.
  • Data Flow and Communication Between HAA Nodes:
    • Behind the scenes, the Orchestrator nodes in the HAA cluster communicate with each other to synchronize data. This includes information about users, assets, processes, queues, and other Orchestrator-related data.
    • This data synchronization ensures that all nodes have a consistent view of the environment's state. Changes made on one node are propagated to others to maintain data integrity.
  • Failover and Redundancy:
    • If one Orchestrator node becomes unavailable (due to maintenance, failure, etc.), the load balancer detects this and routes traffic to the remaining available nodes.
    • The HAA setup ensures that even if one node goes down, others can still handle requests and maintain the Orchestrator's functionality.
    • The load balancer also monitors the health of Orchestrator nodes and can detect when a node becomes healthy again, allowing it to be included in the load balancing rotation.