A million ways to die from a data race in Go
Explores common data race patterns in Go through real-world examples, including implicit variable capture in closures, concurrent modification of http.Client fields, misaligned mutex and data lifetimes, and unsynchronized access to shared data structures. Each example demonstrates how seemingly correct code can violate Go's memory model, leading to issues ranging from subtle bugs to memory corruption. Provides fixes for each pattern and concludes with recommendations for avoiding data races through better language features, tooling, and coding practices like using the race detector and achieving high test coverage.
