From Debugging to Prevention: Smarter Linux Diagnostics with Automation
summary
Demonstrates a working pipeline that automates collection with sos-style tar archives, runs run_analysis.py (Insights‑Core rules) on a cron schedule, logs findings to a JSON log and fires notifier.py to email and optionally run a bash remediation that restarts offending services. A concrete rule flags processes using >50% memory when free RAM falls below 100 MB (used to detect a memory leak), and the demo uses a lightweight tar generator instead of full SOS reports. Notes availability of 200+ parsers.
AI-written summary · about
chapters (10)
- 0:00Smarter Linux diagnostics intro
- 1:03Insight Score framework
- 2:49Reactive vs proactive diagnostics
- 3:32Insight Score pipeline components
- 7:01Source Reports generation
- 9:33Debugging to prevention shift
- 10:56Proactive diagnostics workflow
- 12:21Memory leak detection demo
- 17:02Proactive automation best practices
- 18:45Q&A
related talks
24mInsights: the state analyzer to rule them all
40mOptimizing Linux Systems: Leveraging Guider for Advanced Performance Analysis and...
53mHow to make syzbot reports easier to debug?
36mIntroducing a New Performance Monitoring Daemon
31mFrom Kernel To Cloud: Building a Complete Open Source Observ...
33mDebugging the smart home
39mHow to Use Common Investigation Practices in Troubleshooting
34mChaos Debugging: Finding and Fixing Microservice Weak Points