Progress
Version 9
Product Update Bulletin
Fail Over Policies
The fail over policies for Clusters are as follows:
- IsAliveInterval — The polling interval in between calls to the IsAlive method to determine if the application is still alive. The IsAliveInterval is 60 seconds.
- LooksAliveInterval — The polling interval in between calls to the LooksAlive method to determine if an application appears alive. The LooksAliveInterval is 5 seconds.
- CheckInterval — The amount of time to attempt to restart a database on its current node. The CheckInterval period is 900 seconds.
- FailureReportThreshold — The number of unsuccessful restarts to allow during a CheckInterval period before signaling a fail over request. The FailureReportThreshold is 3.
- Retries — The number of attempts to start the application before initiating failover. Retries is 3.
- Delay — The number of seconds to wait prior to initiating fail over. Delay allows the operating system time to perform clean up or other transfer related actions. Delay is 0 seconds.
- FailBack — Indicates if the application is to be transferred back to the primary node upon its return to operation. FailBack is False (Windows).
- AutoStart — Indicates to the cluster software that the application should automatically start without operator intervention. AutoStart is True.(Windows).
Copyright © 2004 Progress Software Corporation www.progress.com Voice: (781) 280-4000 Fax: (781) 280-4095 |