Automating Insight: Using JSL to Transform Console Logs (2025-US-30MP-2309)

Any production environment can generate vast amounts of process logs, whether in manufacturing, software development, sales, or finance. Buried in these plain-text logs is valuable information that can be used to monitor system health, identify failure points, and drive continuous improvement. This presentation describes how the JMP DevOps team built an automated pipeline to convert unstructured console logs into JMP data tables and publish daily diagnostic visualizations to JMP Live.

Using a JSL script with HTTP Requests and JSON parsing, we collect log data from our Jenkins task scheduling environment and use JMP’s regex capabilities to extract key details such as timestamps, progress messages, and error indicators. We used column tagging – a new data table feature in JMP 19 – to streamline the scripting and publishing process, which enables dynamic handling of new errors without manual intervention. The automated system runs each morning, parsing the previous night’s unit test logs and updating interactive graphs on JMP Live. This process has significantly improved our ability to detect, diagnose, and resolve recurring issues, ultimately leading to a more stable and transparent development pipeline.