The Bottleneck: When Monitoring Only Warns You After the Fact
As a leading European platform for legally binding e-signatures with locations in Austria, Germany, and Switzerland, MOXIS is built on maximum availability and uncompromising data privacy. These aren't optional extras—they are the foundation of the company's entire business model.
Until recently, infrastructure monitoring followed a traditional approach, which created distinct day-to-day challenges:
-
Reactive operations: Performance drops and technical issues were usually only detected after they had already occurred.
-
Gradual performance degradation: Subtle, slow-burning system declines often went unnoticed, buried in everyday background noise.
-
Time-consuming troubleshooting: When an issue did arise, finding the root cause required tedious, manual analysis.
For a company handling highly sensitive customer data, this was not a sustainable long-term approach. The goal was clear: detect and resolve issues before they ever impact the end user.
The Architecture: All Data in One Central Location
Together, MOXIS and EBCONT built a centralized observability solution powered by the Elastic Stack. Log data from all 400+ systems now flows into a high-performance Elastic cluster. Instead of isolated, siloed solutions, there is now a single, consolidated view of the entire infrastructure's health.
Additionally, Java-based applications are monitored using Application Performance Monitoring (APM). This surfaces bottlenecks right at the source, long before the system noticeably slows down. The project was executed through a structured, four-phase approach: from the initial assessment and building a robust cluster to the gradual integration of the first dashboards and alerts. Since launching in 2023, the system has been continuously expanded to include new data sources.
Early Detection: Automated Analysis Instead of Manual Troubleshooting
With this consolidated data foundation in place, automated machine learning processes take over the heavy lifting. They continuously analyze performance patterns, autonomously correlate log data, and immediately flag anomalous behavior.
As soon as irregularities occur, automated alerting pipelines kick in. Integrated SLA monitoring is deeply woven into daily operations, allowing the operations team to step in early and nip potential IT issues in the bud. This approach has noticeably driven down the number of reactive support tickets.
"With Elastic Observability, we have successfully shifted to proactive operations. Today, we catch bugs and performance bottlenecks before our customers even have to open a ticket. It has become a foundational pillar for the quality of our SaaS operations."
— Stephan Kochauf, COO of MOXIS
Uncompromising Data Privacy: Intelligent Analysis Entirely On-Premises
One critical factor sets this project apart from standard cloud solutions: the entire infrastructure runs fully on-premises and containerized via Docker in an Austrian data center. No sensitive signature or log data ever leaves the environment for external clouds. The project proves that cutting-edge early detection and strict European data privacy standards can go hand in hand perfectly.
Key Takeaways for SaaS Providers
The MOXIS use case clearly demonstrates that proactive monitoring isn't a question of company size, but of the right architecture. For those looking to keep complex software structures stable while maintaining full data sovereignty, this on-premises architecture delivers the ideal solution.
Ready to shift your IT monitoring from reactive to proactive? Let's start an informal conversation about what a modern observability infrastructure could look like for your systems.
Get in touch directly with our expert Christoph Eder.