5 ms·

I've always been under the impression that concepts like thread/continuation-local storage were generally discouraged, but I never fully understood the reason.

I've always been under the impression that concepts like thread/continuation-local storage were generally discouraged, but I never fully understood the reason. Obviously, you'd want to avoid coupling your code to an HTTP-specific context if you want to expose a gRPC endpoint in the future. Are there any other downsides to this approach?

gcommer 8y ago
The problem is that all CLS systems in JS are inherently leaky, due to the many games JS libraries/apps can (and do) play with resource pooling. Implementors basically have to play a plug-the-holes game getting all the popular libraries to properly "play along" with the CLS system. For example, domains needed tons of PRs very similar in nature to this: https://github.com/sequelize/sequelize/pull/589/files#diff-1fcbe253f68a98bb45ab62b33825d759R8 https://github.com/sequelize/sequelize/pull/589/files#diff-1... Also, they typically come with a sizable perf hit: https://github.com/bmeurer/async-hooks-performance-impact https://github.com/bmeurer/async-hooks-performance-impact