Abstract: An embodiment of the present invention is a technique for testing core functionality of a clustered system having a plurality of nodes. A cluster population test that includes populating the clustered system with cluster-aware applications is performed. A crash reboot test is performed to stress core operations running within the clustered system; the crash reboot test includes executing a series of crashes or reboots on a subset of the nodes. A move test that includes moving resources to a selected node of the clustered system and verifying that the resources remain on-line at the selected node is performed. A stop test that includes stopping cluster service on a selected subset of nodes of the clustered system and restarting the cluster service on the selected subset of nodes is performed. A Blue Screen Of Death (BSOD) test that includes crashing in BSOD mode one of the nodes of the clustered system and testing reservation release on the crashed node is performed.