Analyzing HotSpot Crashes: Detection, Classification, and Handling - Java

Explore the intricacies of HotSpot crash analysis in this 46-minute Java conference talk by Volker Simonis, Development Expert at SAP. Uncover the contents of hs_err_pid and replay_pid files generated during Java application crashes. Learn to interpret error files produced by the HotSpot crash handler and leverage this information for crash analysis and reproduction. Gain proficiency in using tools like jcmd and the HotSpot Serviceability Agent to introspect native core files and extract crucial Java information. Delve into topics such as HotSpot VM overview, runtime crashes, interpreter crashes, out-of-memory issues, compiler crashes, and debugging techniques. By the end of this talk, acquire the skills to detect, classify, and handle Java HotSpot VM problems effectively, enhancing your ability to troubleshoot and maintain Java applications.

Intro
HotSpot VM Overview
Crash in the Runtime
Where did we crash?
Crash in the Interpreter
Out of Memory - Serviceability Agent
Crash in the Compiler
Debugging the Crash
Serviceability Agent - Dump Classes from Core
Crash Location and Crash Reason