Developers and testers constantly use log files and metrics to find and troubleshoot failures. But unfortunately, getting at useful often involves data wrangling, regexes and parsing scripts. Now, let’s say you want to build a report that shows which services exited, and for what reasons.