For very large existing services, it would seem that interfaces would need to be refactored to include space for a trace_contect (method1, calls method2, calls method3, which enqueues task to ThreadPool...). That sound correct?
Hi guys! Thank for the video, it was really useful. By the way, can you share some materials on how to leverage context propagation to FastAPI auto-instrumentation?
The standard for HTTP is by associating the trace id to the headers as in the W3C Trace Context spec. For gRPC, this trace id should be sent in the metadata, as far as I know.
Yes as someone who wants to contribute to otel this video helps me alot to understand how it works
This was very helpful, thank you very much!
For very large existing services, it would seem that interfaces would need to be refactored to include space for a trace_contect (method1, calls method2, calls method3, which enqueues task to ThreadPool...). That sound correct?
Hi guys! Thank for the video, it was really useful.
By the way, can you share some materials on how to leverage context propagation to FastAPI auto-instrumentation?
I am struggling how we can propagate trace across grpc services. Can you please provide some direction?
The standard for HTTP is by associating the trace id to the headers as in the W3C Trace Context spec. For gRPC, this trace id should be sent in the metadata, as far as I know.