Troubleshooting Passive Agents

Because passive agents must wait for a request from the JT Harness harness before running tests, the port that the passive agent uses must be the same as that used by the JT Harness harness to send requests.

Errors in configuring, synchronizing, or implementing the connection between the agent and the JT Harness harness are the most probable causes of failure.

Use the Agent Monitor window, the JT Harness harness Test Manager window, the agent GUI, and the following list of actions as a guide when troubleshooting problems running passive agents:

  1. Verify that the agent was started before the JT Harness harness started the test run. If not, repeat the test run.
  2. Verify that the port value used when starting the agent matches the port value used by the JT Harness harness to send requests.
  3. Check the physical connection between the JT Harness harness platform and the test platform.
  4. Use the Configuration Editor in the Test Manager window to verify that the harness is configured to use agents when running tests. See the JT Harness User's Guide: Graphical User Interface for detailed description of the Configuration Editor and the Test Manager window.
  5. If you are running the tests using multiple Java Virtual Machines, use the Configuration Editor window to verify that the path you provided in the Java Launcher question is the path of the launcher for the agent running tests.
  6. If tests are failing or have errors, check the error messages displayed in the Test Manager window. If the error indicates that tests are failing because of missing classes, perform the following actions:
    1. Verify that the class paths used to start the agent are correct.
    2. Use the Configuration Editor window to verify that the harness is correctly configured to use the agent on the test system.
    3. Run the agent using the -trace option to verify that the paths in the stream messages for the test are correct. If the paths are not correct for the test system, creae a map file for the agent to use in translating host-specific values into values that the agent can use. See Creating a Map File
    4. If a map file was used to run the test, use the Test Run Messages pane to verify that the -mapArgs command is present in the stream messages. If the -mapArgs command is not present, verify that both the agent and the harness are configured to use the map file. Use the Configuration Editor window to verify that the harness is configured to use the agent map file.

Copyright © 2001, 2011, Oracle and/or its affiliates. All rights reserved.