This module implements a context API on top of CLS/continuation-local-storage.
The primary objective of CLS is to implement a transparent context API, that is, you don't need to pass around a
variable everywhere in your application code.
const CLSContext = ;const tracer =ctxImpl: 'zipkin'recorder // typically HTTP or KafkalocalServiceName: 'service-a' // name of this application;
A note on CLS context vs. explicit context
There are known issues and limitations with CLS, so some people might prefer to use
the drawback then is that you have to pass around a context object manually.
A note on CLS context and Promises
By default, this package is not suitable if your code inside the context uses promises, however you can enable an experimental feature for async/await support by using cls_hooked library which underneath uses async_hooks.
const CLSContext = ;const tracer =ctxImpl: 'zipkin' truerecorderlocalServiceName: 'service-a';
At the time of this writing,
async_hooks have some performance implications whose effect may vary depending on the node version.
The underneath implementation for async_hooks may change, that is why we hide that with the opt-in parameter.
A note on the workings of CLS context
The package will create a namespace called 'zipkin' by default, if it does not exist yet. In this namespace the code sets the context with the key 'zipkin'. This does not mean that the context is overwritten at every request. The namespace is tied to the call-chain. Data stored within that namespace is unique to that request and namespace. For reference see: here.
If you are using the
Express framework with
body-parser middleware may interfere and lose the context. To prevent that, make sure that the
body-parser middleware is registered before
zipkin in the middleware chain. If you are not explicitly registering
body-parser, please do so by adding
app.use(bodyParser.json()); above the Zipkin middleware registration line.