Method and appartus for detecting an application process failure
View Patent ↗A method and apparatus for detecting an application process failure is described. In one embodiment, a process membership manager is opened with a first process. This action by the first process causes an instance indicia associated with the process membership manager to be assigned a first predefined value. The first process (i.e., parent process) then forks (or creates) a second process (i.e., child process). Once the second process is created, the instance indicia is changed to a second predefined value. In the event the second process fails, the second predefined value will change to reflect the process failure. Consequently, this change of the second predefined value causes a message, which provides notice of the second process failure, to be sent to a process membership manager.
1. A method for initiating a failover process, comprising:
receiving a message from an operating system in response to a failed application process, the failed application process having been forked from a first process, the first process being one of a plurality of administrative members and the failed application process being a regular member in a regular membership group; and
sending a notification of said failed application process to all of the plurality of administrative members in response to receiving said message, wherein said first process commences said failover process in response to said notification.
2. The method of claim 1 , wherein said message is a close entry point call.
3. The method of claim 1 , wherein said first process and said failed application process are a parent process and a child process, respectively.
4. The method of claim 1 , wherein said first process is a high availability daemon.
5. The method of claim 1 , further comprising initiating a failover in response to said message.
6. An apparatus for initiating a failover process, comprising:
means for receiving a message from an operating system in response to a failed application process, the failed application process having been forked from a first process, the first process being one of a plurality of administrative members and the failed application process being a regular member in a regular membership group; and
means for sending a notification of said failed application process to all of the plurality of administrative members in response to receiving said message, wherein said first process commences said failover process in response to said notification.
7. The apparatus of claim 6 , wherein said message is a close entry point call.
8. The apparatus of claim 6 , wherein said first process and said failed application process are a parent process and a child process, respectively.
9. The method of claim 6 , wherein said first process is a high availability daemon.
10. The method of claim 6 , further comprising initiating a failover in response to said message.