How do we do it? Testing and Verifying bugs for released software
summary
Defines verification as reproduce collect document then automate: convert every reproducible failure into an MTR test so fixes can be regression checked. Uses load tools like sysbench and mysqlslap plus a homegrown random loader to populate real tables. Spins complex topologies in minutes with divi player to reproduce replication and cluster bugs. Analyzes crashes with gdb/perf and core dumps. Prioritizes by impact not source, keeps ticket timeouts after a few unanswered probes, and turns fixes into automated tests.
AI-written summary · about
chapters (5)
- 0:00Introduction to bug verification
- 12:20Classifying and prioritizing bugs
- 23:16Tools for bug verification and testing
- 31:02Reducing post-release bug inflow
- 35:00Q&A session
related talks
38mThe Verification of a Distributed System
35mEffective Debugging of Non-Reproducible Bugs
10mHow to test software without writing tests
36mHow to defuse a bomb Wait, I mean a bug
18mPreview: Verification (Correctness & Debugging)
50mAutomate Finding & Fixing Bugs with Replay QA
23mBeyond Tests What to Verify in AI Generated Code
43mCan you Test it