Dependable communications is becoming a critical factor due to the pervasive usage of networked embedded systems that increasingly interact with human lives in one way or the other in many real-time applications. Though many smaller systems are providing dependable services employing uniprocesssor solutions, stringent fault containment strategies etc., these practices are fast becoming inadequate due to the prominence of COTS in hardware and component based development(CBD) in software as well as the increased focus on building 'system of systems'. Hence the repertoire of design paradigms, methods and tools available to the developers of distributed real-time systems needs to be enhanced in multiple directions and dimensions. In future scenarios, potentially a network needs to cater to messages of multiple criticality levels (and hence varied redundancy requirements) and scheduling them in a fault tolerant manner becomes an important research issue. We address this problem in the context of Controller Area Network (CAN), which is widely used in automotive and automation domains, and describe a methodology which enables the provision of appropriate scheduling guarantees. The proposed approach involves definition of fault-tolerant windows of execution for critical messages and the derivation of message priorities based on earliest deadline first (EDF).