Final Thesis: Enhancing Observability in the JValue Project: Logging for a Microservice-based System
Abstract: This thesis discusses the integration of a log-management and request tracing system into the JValue microservice application using industry standards and open source solutions. For the log-management Fluent Bit and Data Prepper are used to collect and aggregate the logs, while OpenSearch is storing them as well as providing the UI for querying said logs. Through the use of pipelines the logs are enriched and then separated by the kind of application they were generated from. The request tracing is reusing Data Prepper and OpenSearch for aggregation, storage and UI, but uses OpenTelemetry for trace generation and collection. OpenTelemetry is injected into the JValue microservices using its zero-code instrumentation approach in order to avoid any code modification. All of these services are deployed to a pre-existing Kubernetes cluster using Helm.
Keywords: microservices, observability, logging, log aggregation, OpenSearch, OpenTelemetry, JValue, open source
PDF: Master Thesis
Reference: Felix Müller. Enhancing Observability in the JValue Project: Logging for a Microservice-based System. Friedrich-Alexander-Universität Erlangen-Nürnberg: 2024.