faust-streaming
Migrate from opentracing to opentelemetry
The opentracing project is deprecated in favor of opentelemetry. Due to ongoing flaws with opentracing, we have had difficulty troubleshooting particular bugs that rely on the usage of opentracing.Span objects.
We will need to miigrate opentracing usage over to opentelemetry as documented in https://opentelemetry-python.readthedocs.io/en/stable/shim/opentracing_shim/opentracing_shim.html.
👋 (I am one of the maintainers of OpenTelemetry).
We will need to miigrate opentracing usage over to opentelemetry as documented in https://opentelemetry-python.readthedocs.io/en/stable/shim/opentracing_shim/opentracing_shim.html.
Shim is aimed at users who want to use OTLP protocol while the code is still instrumented with OpenTracing and plan for incremental migration in large codebases. I looked at opentracing usage Faust codebase. I think you are better off truly migrating over to OpenTelemetry by directly instrumenting with opentelemetry-api. Let us know if you need any clarification and help with this.
Actually, one of my targets for mode-ng is to integrate OpenTelemetry protocol. Of course, long way to go 😭
👋 (I am one of the maintainers of OpenTelemetry).
We will need to miigrate opentracing usage over to opentelemetry as documented in https://opentelemetry-python.readthedocs.io/en/stable/shim/opentracing_shim/opentracing_shim.html.
Shim is aimed at users who want to use OTLP protocol while the code is still instrumented with OpenTracing and plan for incremental migration in large codebases. I looked at
opentracingusage Faust codebase. I think you are better off truly migrating over to OpenTelemetry by directly instrumenting withopentelemetry-api. Let us know if you need any clarification and help with this.
Hi @srikanthccv .. if we start using opentelemetry that would be manual instrumentation.. right? since currently it doesnt support automatic instrumentation for faust. My question in other thread was, do we have any method available which can be used as decorator over to span child traces for the same. any lead on this?
