A state table includes a plurality of possible states of a computer system and the corresponding actions which produce transitions between source and target states. A set of test programs is stored, each test program performing an action in the state table. A test selects an action corresponding to the current state of the computer system; executes the test program which performs the selected action; determines the state of the computer system after the test program has executed; and compares the determined state to the state indicated in the state table as the target state to the selected action on the source state. When an error is found, instead of stopping execution, the test operations to be performed are dynamically reconfigured. Weightings are dynamically allocated to actions in the state table to create a weighted set, and selection of the next test is carried out using random selection over the weighted set. Thus, the continued testing is biased over time towards particular transitions and/or states near to the error location. This enables a tester to discover any other bugs in the same area and also to obtain further diagnostic data on the failure.