Understanding Sun Cluster HA for Siebel Fault Monitor

Use the information in this section to understand Sun Cluster HA for Siebel Fault Monitors.

This section provides the following information.

• Description of Sun Cluster HA for Siebel Fault Monitor's probing algorithm or functionality.

• Conditions, messages, and recovery actions associated with unsuccessful probing.

• Conditions and messages associated with successful probing.

Sun Cluster HA for Siebel supplies two Fault Monitors, the Gateway Fault Monitor and the Server Fault Monitor. Each of these Fault Monitors is described in the following sections.

Extension Properties

Sun Cluster HA for Siebel Fault Monitor uses the following extension properties. You might be able to tune these extension properties. See Sun Cluster HA for Siebel Extension Properties to review or set extension properties.

Table 1–3 Sun Cluster HA for Siebel (SUNW.sblgtwy) Extension Properties

Name/Data Type	Description
Confdir_list(stringarray)	The Siebel gateway root directory location  Default: None Tunable: At creation

Table 1–4 Sun Cluster HA for Siebel (SUNW.sblsrvr) Extension Properties

Name/Data Type	Description
Confdir_list(stringarray)	The Siebel server root directory location  Default: None Tunable: At creation
siebel_enterprise (string)	The Siebel Enterprise name  Default: None Tunable: At creation
siebel_serves (string)	The Siebel server name  Default: None Tunable: At creation

Probing Algorithm and Functionality

Siebel Gateway Fault Monitor

The Siebel Gateway Fault Monitor monitors the Siebel gateway process. If the Siebel gateway process dies, the Fault Monitor restarts it, or fails it over to another node.

Siebel Server Fault Monitor

The Siebel Server Fault Monitor does the following every thorough_probe_interval seconds.

Monitors the Siebel database

If the Siebel database fails, the status of the Siebel server is marked as DEGRADED. When the Siebel database restarts again, the Siebel server resource probe tries to verify that the Siebel server is functioning. When this test fails, the Siebel server is restarted or failed over to another node.

The Fault Monitor also starts the Siebel server when the Siebel database that was not available when the START method ran then becomes available.

Monitors Siebel gateway

If the Siebel gateway fails, the status of the Siebel server is marked as DEGRADED. When the Siebel gateway restarts again, the Siebel server resource probe tries to verify that the Siebel server is functioning. When this test fails, the Siebel server is restarted or failed over to another node.

The Fault Monitor also starts the Siebel server when the Siebel gateway that was not available when the START method ran then becomes available.

Monitors the Siebel server and all its enabled components

If the Siebel server fails, it is restarted or failed over. If any of the Siebel components fail, a partial (10%) failure is reported to the framework. Currently, component failures may be detected only in English language installations of Siebel.