Operations guide

This topic helps you keep your NeoLoad environment healthy after deployment. It focuses on the operational side: what to monitor, when to update, and what to do when something breaks. For day-to-day admin tasks, see the admin guide. For test creation and execution, see performance testing in NeoLoad Web.

Monitor system health

Regular monitoring helps you catch problems before they affect test runs. The following table summarizes what to watch for each component:

Component

Role

What to monitor

Controller

Orchestrates test runs, manages connections to load generators, and collects test results.

CPU, memory, and disk space. Low resources can cause data loss or test failures. Verify that the controller can reach NeoLoad Web to publish results.

Load generators

Simulate virtual users during a test. Confirm they're online in Infrastructure in NeoLoad Web.

CPU, memory, and network bandwidth. A saturated load generator produces inaccurate response times that reflect its own limits, not your application's performance.

NeoLoad Web (on-premises only)

Stores results, manages infrastructure, and provides dashboards for analysis. You manage the platform if you host it yourself.

Web interface responsiveness, database connection pool use, query response times, and disk space. Review retention policies before storage becomes a problem.

Network

Connects all components. The controller must reach load generators, and load generators must reach the application under test.

Connectivity between components. For cloud and hybrid deployments, also confirm outbound HTTPS access to the Tricentis NeoLoad cloud.

Update components

NeoLoad releases updates regularly with new features, bug fixes, and security patches. Review the release notes, back up your configurations and database, and schedule a maintenance window when no tests are active.

The recommended update order is:

  1. Update the controller.

  2. Update the load generator agents.

  3. Update the server (on-premises only). For cloud deployments, Tricentis NeoLoad updates the server automatically.

  4. Run a short test to confirm that everything works.

Troubleshooting

Choose the issue that matches your situation:

A test that won't start usually points to an infrastructure or configuration issue:

  • Confirm that your load generators are online and available in the selected zone.

  • Verify that the controller can communicate with all load generators.

  • Confirm that your license has enough capacity for the number of virtual users you've configured.

The most common cause is a connection issue between the controller and NeoLoad Web:

  • Verify the API token and network access.

  • Confirm that you selected the correct workspace for the results.

Storage limits can also prevent results from appearing:

  • For on-premises deployments, check disk space.

  • For cloud deployments, check your subscription limits.

Unexpected response times often point to resource issues on the load generators:

  • Check CPU and memory on each load generator. A saturated load generator produces response times that reflect its own limits, not your application's performance.

  • Check network bandwidth between the load generators and the application.

  • Confirm that the application under test has enough resources (CPU, memory, database connections) to handle the load.

If users can't sign in or the interface won't load:

  • Check their accounts and permissions.

  • Confirm they're assigned to the correct workspace.

  • For on-premises deployments, verify that the NeoLoad Web service is running and that firewall rules allow access to the server.

What's next

Here's what to explore after you're familiar with the operations basics: