Different approaches to data storage systems software work and debugging tools creation are presented in this work. Approaches to non-invasive profiling of Linux dynamic libraries. Also, the algorithm of cluster resources update which allows to update the cluster resources without cluster consistency disruption (i.e. forced nodes stop) is considered. Moreover, the methods for log-file view automation, allowing the optimization of test results analysis.