Merge pull request #121 from TranslatorSRI/otel-instrumentation

otel stuff in place, helm changes required are just passing the following env vars:
 - SERVER_NAME (defaults to it's infores)
 - JAEGER_HOST
 - JAEGER_PORT 
 - OTEL_ENABLED (set to false by default) 

Closes #105.

README.md

@@ -34,3 +34,16 @@ Helm charts can be found at https://github.com/helxplatform/translator-devops/he
 ```bash
 curl -X POST "http://localhost:6434/lookup?string=oxycod&offset=0&limit=10" -H "accept: application/json"
 ```
+
+## Configuration
+
+NameRes can be configured by setting environmental variables:
+
+* `SOLR_HOST` and `SOLR_PORT`: Hostname and port for the Solr database containing NameRes information.
+* `SERVER_NAME`: The name of this server (defaults to `infores:sri-name-resolver`)
+* `SERVER_ROOT`: The server root (defaults to `/`)
+* `MATURITY_VALUE`: How mature is this NameRes (defaults to `maturity`, e.g. `development`)
+* `LOCATION_VALUE`: Where is this NameRes setup (defaults to `location`, e.g. `RENCI`)
+* `OTEL_ENABLED`: Turn on Open TELemetry (default: `'false'`) -- only `'true'` will turn this on. 
+    * `JAEGER_HOST` and `JAEGER_PORT`: Hostname and port for the Jaegar instance to provide telemetry to.
+    * `JAEGER_SERVICE_NAME`: The name of this service (defaults to the value of `SERVER_NAME`)

api/server.py

@@ -8,7 +8,7 @@
   * Matching names are returned first, followed by non-matching names
 """
 import json
-import logging
+import logging, warnings
 import os
 import re
 from typing import Dict, List, Union, Annotated
@@ -358,3 +358,39 @@ async def lookup(string: str,
 
 # Override open api schema with custom schema
 app.openapi_schema = construct_open_api_schema(app)
+
+# Set up opentelemetry if enabled.
+if os.environ.get('OTEL_ENABLED', 'false') == 'true':
+    from opentelemetry.instrumentation.fastapi import FastAPIInstrumentor
+    from opentelemetry import trace
+    from opentelemetry.exporter.jaeger.thrift import JaegerExporter
+    from opentelemetry.sdk.resources import SERVICE_NAME, Resource
+    from opentelemetry.sdk.trace import TracerProvider
+    from opentelemetry.sdk.trace.export import BatchSpanProcessor
+    # from opentelemetry.sdk.trace.export import ConsoleSpanExporter
+
+    from opentelemetry.instrumentation.httpx import HTTPXClientInstrumentor
+
+    # httpx connections need to be open a little longer by the otel decorators
+    # but some libs display warnings of resource being unclosed.
+    # these supresses such warnings.
+    logging.captureWarnings(capture=True)
+    warnings.filterwarnings("ignore", category=ResourceWarning)
+    plater_service_name = os.environ.get('SERVER_NAME', 'infores:sri-name-resolver')
+    assert plater_service_name and isinstance(plater_service_name, str)
+
+    jaeger_exporter = JaegerExporter(
+        agent_host_name=os.environ.get("JAEGER_HOST", "localhost"),
+        agent_port=int(os.environ.get("JAEGER_PORT", "6831")),
+    )
+    resource = Resource(attributes={
+        SERVICE_NAME: os.environ.get("JAEGER_SERVICE_NAME", plater_service_name),
+    })
+    provider = TracerProvider(resource=resource)
+    # processor = BatchSpanProcessor(ConsoleSpanExporter())
+    processor = BatchSpanProcessor(jaeger_exporter)
+    provider.add_span_processor(processor)
+    trace.set_tracer_provider(provider)
+    FastAPIInstrumentor.instrument_app(app, tracer_provider=provider, excluded_urls=
+                                       "docs,openapi.json")    
+    HTTPXClientInstrumentor().instrument()

requirements.txt

@@ -7,3 +7,7 @@ jsonlines
 
 # For testing
 pytest
+opentelemetry-sdk==1.21.0
+opentelemetry-exporter-jaeger==1.21.0
+opentelemetry-instrumentation-fastapi==0.42b0
+opentelemetry-instrumentation-httpx==0.42b0