Nowadays, the consequences of failure and downtime of distributed systems have become more and more severe. As an obvious solution, these systems incorporate protection mechanisms to tolerate faults that could cause systems failures and system dependability must be validated to ensure that protection mechanisms have been implemented correctly and the system will provide the desired level of reliable service. This paper presents a systematic approach for identifying (1) characteristic sets of critical system elements for dependability testing (single points of failure and recovery groups) based on the concept of layered networks; and (2) the most important combinations of components from each recovery group based on a combinatorial technique. Based on these combinations, we determine a set of test templates to be performed to demonstrate system dependability.