KERNEL DEBUGGING IN A CLUSTER COMPUTING SYSTEM

摘要

An embodiment of a method of maintaining operation of a cluster of computing devices includes an initial step of detecting a suspended kernel process on a first of the computing devices. In addition to the step of detecting the suspended kernel process the method includes the step of issuing a first signal that causes a cluster management process to disregard the first of the computing devices when determining if there exists a quorum of the computing devices.