This section provides the sequence of tasks you can perform to diagnose and solve problems with Oracle Traffic Director.
Verify whether the system configuration is correct.
For information about the supported platforms and operating systems, see the Oracle Fusion Middleware Supported System Configurations at:
http://www.oracle.com/technetwork/middleware/ias/downloads/fusion-certification-100350.html
Look for a solution to the problem in Section 16.2, "Solutions to Common Errors."
Check whether the information in Section 16.3, "Frequently Asked Questions" helps you understand or solve the problem.
Try to diagnose the problem.
Review the messages logged in the server log. Look for messages of type WARNING, ERROR, and INCIDENT_ERROR.
For messages of type WARNING and ERROR, try to solve the problem by following the directions, if any, in the error message.
An INCIDENT_ERROR message indicates a serious problem caused by unknown reasons. You should contact Oracle for support.
Increase the verbosity of the server log, and try to reproduce the problem.
Oracle Traffic Director supports several log levels for the server log, as described in Table 12-1, "Server Log Levels". The default log level is NOTIFICATION:1. The least verbose log level is INCIDENT_ERROR, at which only serious error messages are logged. At the TRACE:1, TRACE:16, or TRACE:32 levels, the logs are increasingly verbose, but provide more detailed information, which can be useful for diagnosing problems.
Increase the log verbosity and then try to reproduce the problem. When the problem occurs again, review the messages logs for pointers to the cause of the problem.
For information about changing the server log level, see Section 12.3, "Configuring Log Preferences."
Restore the instances to a previous configuration.
When you redeploy a modified configuration to its instances, a backup of the previous configuration is stored in a zip file on the administration server.
Restore the instances to an appropriate previous configuration as described in Section 4.9, "Restoring a Configuration from a Backup," and check whether the problem persists.
If the problem does not occur with the previous configuration, it is clear that the problem is caused by a change made in the current configuration.
Contact Oracle for support, as described in Section 16.4, "Contacting Oracle for Support."
This section provides information about the tasks you can perform to diagnose and solve problems with an Oracle Traffic Director high availability configuration.
Ensure that the interface being used is EoIB (Ethernet over InfiniBand).
The Oracle Traffic Director configuration must be deployed on two nodes. For more information, see Section 14.2, "Creating and Managing Failover Groups."
The router ID for each failover group has to be unique.
Make sure that KeepAlived is installed. In most cases KeepAlived software is installed by default on both the Exalogic compute nodes (or VMs) where Oracle Traffic Director instances are running. To check if KeepAlived is installed, run the following command:
rpm -qa | grep keepalived
If KeepAlived is correctly installed, an output similar to the following is displayed:
keepalived-1.2.2-1.el5
Note that if KeepAlived is not installed, the RPM can be found in the Exalogic software repository. For more information about KeepAlived, see Section 14.2.1, "How Failover Works."
For KeepAlived specific information, check the logs in the /var/log/messages directory.
Make sure to provide the correct VIP address and the appropriate subnet mask (netmask) bit-size for successfully completing the high availability configuration. In addition, ensure that you provide the netmask bits and not the actual netmask value. For more information, see Section 14.2.3, "Creating Failover Groups."
If you are using Oracle Traffic Director 11.1.1.6.0, note that it does not support high availability with more than one Oracle Traffic Director configuration on the same pair of administration nodes.