When Linux Memory Accounting Goes Wrong
Offered By: USENIX via YouTube
Course Description
Overview
Dive into a comprehensive exploration of Linux memory accounting challenges in this 27-minute conference talk from SREcon21. Uncover the intricacies of debugging memory-related issues where hosts become inaccessible despite cgroup limitations. Explore the complexities of memory accounting in cgroups and learn how improper accounting can lead to catastrophic consequences for cohosted applications and host systems. Follow the speaker's journey through initial debugging, issue recurrence, and in-depth troubleshooting. Gain insights into memory accounting validation, specific case analysis, and effective mitigation strategies. Enhance your understanding of Linux system resource management and improve your ability to handle critical memory-related problems in production environments.
Syllabus
Intro
Search at LinkedIn
Concepts
The Problem
Initial Debugging and Resolution (conto)
Issue recurrence
Troubleshooting (contd)
Memory accounting in cgroups
Validation
What happened in our Case
Mitigation
Conclusion
Taught by
USENIX
Related Courses
How to Not Destroy Your Production Kubernetes ClustersUSENIX via YouTube SRE and ML - Why It Matters
USENIX via YouTube Knowledge and Power - A Sociotechnical Systems Discussion on the Future of SRE
USENIX via YouTube Tracing Bare Metal with OpenTelemetry
USENIX via YouTube Improving How We Observe Our Observability Data - Techniques for SREs
USENIX via YouTube