One of the main requirements for building an autonomic system is to have a robust monitoring framework. In this paper, a systematic distributed event based (DEB) system monitoring framework “RFDMon” is presented for measuring system variables (CPU utilization, memory utilization, disk utilization, network utilization, etc.), system health (temperature and voltage of Motherboard and CPU) application performance variables (application response time, queue size, and throughput), and scientific application data structures (PBS information and MPI variables) accurately with minimum latency at a specified rate and with controllable resource utilization. This framework is designed to be tolerant to faults in monitoring framework, self-configuring (can start and stop monitoring the nodes and configure monitors for threshold values/changes for publishing the measurements), aware of execution of the framework on multiple nodes through HEARTBEAT messages, extensive (monitors multiple parameters through periodic and aperiodic sensors), resource constrainable (computational resources can be limited for monitors), and expandable for adding extra monitors on the fly. Since RFDMon uses a Data Distribution Services (DDS) middleware, it can be used for deploying in systems with heterogeneous nodes. Additionally, it provides a functionality to limit the maximum cap on resources consumed by monitoring processes such that it reduces the effect on the availability of resources for the applications.