An Approach to Identify Exceptions Propagated by .NET Web Services

High dependability, availability and fault-tolerance are open problems in Service-Oriented Architecture (SOA). Some research suggests the use of exception handling as a powerful mechanism to improve the robustness of SOA. However some works observed that the exception messages and stack traces propagated by exceptions in SOA are not enough to identify the exact cause of an exception and, hence, applying an adequate recovery technique. This paper addresses this gap by proposing an approach to identify exceptions propagated by .NET web services. In this direction, this work: (i) describes an approach to identify exceptions, (ii) implements the approach and (iii) evaluates the proposed approach in a real case study.