Troubleshooting secondary node problems

In a cluster configuration, when a primary node is stopped, and awareness or instant messaging is not working when relying on a secondary node, check to make sure there are virtual host definitions created for each cluster member for the configured ports.

About this task

View the log file profile_root\logs\SystemOut.log for this error:
Mapping for non-existing siplet: Method=INVITE, Application: Unknown.
If you find this error, the virtual hosts definitions must be updated to have host aliases defined for the SIP ports configured on each cluster member.

For additional information about default paths, see Directory conventions.

Procedure

  1. In the WebSphere® Integrated Solutions Console, gather the port configurations for each defined cluster member by selecting Servers > Application Servers.
  2. Select a cluster member.
  3. In the "Communications" section, click Ports.
  4. Record both the SIP_DEFAULTHOST and SIP_DEFAULTHOST_SECURE port numbers.
  5. Repeat the preceding steps for each cluster member.
  6. Select Environment > Virtual Hosts.
  7. Select default_host
  8. In the "Additional Properties" section, click Host Aliases.
  9. For each of the defined cluster members, ensure the ports that you recorded are present in the definitions.

    For example, if the secondary node is defined on server1.example.com at ports 5062 and 5063, ensure the host aliases are defined as *:5062 and *:5063.

  10. To create a new virtual host definition:
    1. Click New.
    2. Type the Fully qualified domain name of the host.
    3. Type the port number.
    4. Click OK and Save.
    5. Repeat for each port that requires a virtual host.