Configuring Connection Limits on NSX Native Load Balancer Virtual Servers

Products

• VMware NSX

Issue/Introduction

This article provides guidance on controlling traffic flow on NSX Native Load Balancer Virtual Servers. While the NSX Native Load Balancer does not provide granular L7 request-per-second (RPS) rate limiting, it allows for the configuration of connection-level limits that protect backend servers from saturation and burst traffic.

Environment

VMware NSX

Resolution

To prevent backend server exhaustion and manage traffic load, administrators can configure connection limits on the Virtual Server object. These parameters apply to both L4 (TCP/UDP) and L7 (HTTP/HTTPS) Virtual Servers.

Available Parameters

• Max Concurrent Connections: Defines the maximum number of simultaneous active connections that the Virtual Server will accept. Once this limit is reached, the Load Balancer will reject or queue new incoming connections, preventing the backend pool members from becoming overwhelmed.
• Max New Connection Rate: Defines the maximum number of new connections that the Virtual Server will accept per second. This is an effective tool to mitigate burst traffic or connection-flood attacks.

Configuration Steps

1. Log in to the NSX Manager UI.
2. Navigate to Networking > Load Balancing > Load Balancers.
3. Select the relevant Load Balancer instance.
4. Click the Virtual Servers tab and select the specific Virtual Server you wish to configure.
5. Click Edit.
6. Expand the Advanced Settings
7. Locate the following fields and define your desired thresholds:
   • Max Concurrent Connections
   • Max New Connection Rate
8. Click Save to apply the configuration.

Additional Information

Setting up load balancer components