Flagrant Bugs and How to Avoid Them: The ClickHouse Example
summary
Shows real-world bugs encountered when running ClickHouse at large scale, especially processing internet-traffic data, and explains their root causes across application code, compiler, OS, and hardware. Describes concrete avoidance strategies and best practices for writing resilient code and operating scaled ClickHouse clusters.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 3:11Configuration errors
- 7:55Replication issues
- 11:50Code and hardware bugs
- 17:18Hardware failures and effects
- 27:28Bug pattern detection
- 32:37Hardware diagnostics
- 38:01Error messages and users
- 40:50Binary data analysis
- 43:49Conclusion
related talks
59mEffective Use of ClickHouse
53mClickHouse Is Slowing Down
49mSolving real business cases with ClickHouse
37mFinding Bottlenecks in ClickHouse: Profiling, Heatmaps & Flamegraphs
47mClickHouse Without Pain: Our Experience Refactoring a Legacy Cluster and the Impact of Shard Groups
51mWhat You Need to Know About ClickHouse Architecture
47mTesting ClickHouse We Deserve
42mOptimizing ClickHouse query performance
