Library built on top of Confluent Kafka Python adding abstractions for consuming and producing messages in a more Pythonic way.
Make sure you have opentelemetry-sdk installed.
Add this code to your applications entry point as early as possible, e.g:
from opentelemetry import trace
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import BatchSpanProcessor, ConsoleSpanExporter
provider = TracerProvider()
processor = BatchSpanProcessor(ConsoleSpanExporter())
provider.add_span_processor(processor)
trace.set_tracer_provider(provider)
...
consumer = init_app()
consumer.consume()All spans will now be printed to the std out instead of getting exported.
NOTE! This will most probably conflict if you use another SDK, e.g: ddtrace with OTEL enabled
(DD_TRACE_OTEL_ENABLED=true), since it will set up it's own read-only tracer provider. This is mainly for debugging
the generated spans from an OTEL point of view. Once verified, you should also test with your preferred SDK/exporter.