Kibana MCP Server

""" Distributed Tracing Configuration Sets up and configures OpenTelemetry for distributed tracing. """ from opentelemetry import trace from opentelemetry.sdk.trace import TracerProvider from opentelemetry.sdk.trace.export import BatchSpanProcessor from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter from opentelemetry.sdk.resources import Resource from loguru import logger from src.core.constants import APP_NAME, APP_VERSION def setup_tracing(): """Initializes the OpenTelemetry tracer.""" # Create a resource to identify our application resource = Resource(attributes={ "service.name": APP_NAME, "service.version": APP_VERSION, }) # Set up a tracer provider provider = TracerProvider(resource=resource) trace.set_tracer_provider(provider) # For development, we can use a console exporter. # For production, this would be configured to point to a real collector like Jaeger or Datadog. # Example using OTLP (OpenTelemetry Protocol) Exporter: # DISABLED: No collector running on localhost:4318 # exporter = OTLPSpanExporter(endpoint="http://localhost:4318/v1/traces") # processor = BatchSpanProcessor(exporter) # trace.get_tracer_provider().add_span_processor(processor) logger.info("OpenTelemetry tracing initialized (exporter disabled).") def get_tracer(name: str): """Gets a tracer instance for a specific module.""" return trace.get_tracer(name)