DB2 10.5 for Linux, UNIX, and Windows

DB2 fault monitor registry file

A fault monitor registry file is created for every DB2® database manager instance on each physical machine when the fault monitor daemon is started. The keywords and values in this file specify the behavior of the fault monitors.

The fault monitor registry file can be found in the /sqllib/ directory and is called fm.machine_name.reg. This file can be altered using the db2fm command.

If the fault monitor registry file does not exist, the default values will be used.

Here is an example of the contents of the fault monitor registry file:

   FM_ON = no 
   FM_ACTIVE = yes 
   START_TIMEOUT = 600 
   STOP_TIMEOUT = 600 
   STATUS_TIMEOUT = 20 
   STATUS_INTERVAL = 20 
   RESTART_RETRIES = 3 
   ACTION_RETRIES = 3 
   NOTIFY_ADDRESS = instance_name@machine_name

Fault monitor registry file keywords

FM_ON: Specifies whether or not the fault monitor should be started. If the value is set to NO, the fault monitor daemon will not be started, or will be turned off if it had already been started. The default value is NO.
FM_ACTIVE: Specifies whether or not the fault monitor is active. The fault monitor will only take action if both FM_ON and FM_ACTIVE are set to YES. If FM_ON is set to YES and FM_ACTIVE is set to NO, the fault monitor daemon will be started, but it will not be active. That means that is will not try to bring DB2 back online if it shuts down. The default value is YES.
START_TIMEOUT: Specifies the amount of time within which the fault monitor must start the service it is monitoring. The default value is 600 seconds.
STOP_TIMEOUT: Specifies the amount of time within which the fault monitor must bring down the service it is monitoring. The default value is 600 seconds.
STATUS_TIMEOUT: Specifies the amount of time within which the fault monitor must get the status of the service it is monitoring. The default value is 20 seconds.
STATUS_INTERVAL: Specifies the minimum time between two consecutive calls to obtain the status of the service that is being monitored. The default value is 20 seconds.
RESTART_RETRIES: Specifies the number of times the fault monitor will try to obtain the status of the service being monitored after a failed attempt. Once this number is reached the fault monitor will take action to bring the service back online. The default value is 3.
ACTION_RETRIES: Specifies the number of times the fault monitor will attempt to bring the service back online. The default value is 3.
NOTIFY_ADDRESS: Specifies the e-mail address to which the fault monitor will send notification messages. The default is instance_name@machine_name).