Cloud monitoring is an established support mechanism for securing Cloud services. Over the last few years, various Cloud monitoring mechanisms have been proposed with different level of monitoring efficacy. To achieve effective monitoring, the monitoring mechanism is required to address the basic challenge that the information of the target-to-monitor may be inaccessible due to (i) access controls, (ii) privacy protection, or (iii) technical difficulties. Therefore, the indirect monitoring mechanism is particularly proposed for addressing the challenge. Specifically, the indirect mechanism makes inferences about the inaccessible information of monitoring targets with the help of the 'monitoring path' that contains a special set of accessible monitoring data underpinning the inference task. However, the process to effectively discover the monitoring path is an open issue. To address this problem, we present our preliminary studies in this paper. Firstly, we review existing work and reveal the limitations for discovering monitoring paths. Secondly, we analyze real cases to provide insights for designing monitoring paths. Finally, we propose a framework and planned research tasks for developing a novel monitoring path discovery mechanism that facilitates performing indirect Cloud monitoring as the expected contribution of our research. © 2018 IEEE.