In contrast to the monolithic system, microservice architecture decouples the applications into independent service nodes. It enables rapid and reliable delivery among each other. However, transferring data among the service mesh is unavoidable. In this distributed architecture, user privacy data might flow through several service nodes to other ones without the user's awareness and acknowledgment. Hence, it is crucial to trace each service request and be aware of the extent to which personal data is exposed when services are exchanging messages. In this paper, we address the need for identifying any privacy-related concerns within the microservices architecture by building a Privacy and Security Risk Detection (PSRD) model that is based on distributed tracing. Further, we optimize the number of features to successfully classify the risk level of a particular request path correctly. We employ a fully connected neural network (FCNN) to classify the extent of the vulnerabilities that may occur within the path of a discovered service request. Through this model, it is then possible to detect any risks associated with personal data exposed to service providers or through a network of service nodes. Experimental and validation results from using the proposed approach demonstrate the usefulness of the PSRD model.