[Trace]add trace when fd start (#3174)

* add opentelemetry

* add opentelemetry

* add opentelemetry on dequeue

* add opentelemetry on dequeue

* add opentelemetry on dequeue

* fix annotation

* fix annotation when add opentelemetry

* fix opentelemetry-instrumentation-fastapi

* fix pentelemetry-bootstrap

* fix opentelemetry can not work in uvicorn

* move conf to env

* fd start add trace

* fix pre-commit

* fix pre-commit

* change FD_JOB_ID

---------

Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>
Co-authored-by: shige <shige@baidu.com>

fastdeploy/entrypoints/openai/api_server.py

@@ -45,7 +45,7 @@ from fastdeploy.metrics.metrics import (
     get_filtered_metrics,
     main_process_metrics,
 )
-from fastdeploy.metrics.trace_util import inject_to_metadata, instrument
+from fastdeploy.metrics.trace_util import fd_start_span, inject_to_metadata, instrument
 from fastdeploy.plugins.model_register import load_model_register_plugins
 from fastdeploy.utils import (
     FlexibleArgumentParser,
@@ -272,6 +272,7 @@ def launch_api_server() -> None:
 
     api_server_logger.info(f"launch Fastdeploy api server... port: {args.port}")
     api_server_logger.info(f"args: {args.__dict__}")
+    fd_start_span("FD_START")
 
     try:
         uvicorn.run(

fastdeploy/envs.py

@@ -82,6 +82,8 @@ environment_variables: dict[str, Callable[[], Any]] = {
     "ENABLE_V1_KVCACHE_SCHEDULER": lambda: int(os.getenv("ENABLE_V1_KVCACHE_SCHEDULER", "0")),
     # Whether to use PLUGINS.
     "FD_PLUGINS": lambda: None if "FD_PLUGINS" not in os.environ else os.environ["FD_PLUGINS"].split(","),
+    # set trace attribute job_id.
+    "FD_JOB_ID": lambda: os.getenv("FD_JOB_ID"),
 }
 
 

fastdeploy/metrics/trace_util.py

@@ -1,4 +1,5 @@
 import json
+import os
 
 from fastapi import FastAPI
 from opentelemetry import trace
@@ -176,7 +177,22 @@ def start_span(span_name, request, kind=trace.SpanKind.CLIENT):
             return
         # extract Trace context from request.metadata.trace_carrier
         ctx = extract_from_metadata(request)
-        with tracer.start_as_current_span(span_name, context=ctx, kind=kind):
+        with tracer.start_as_current_span(span_name, context=ctx, kind=kind) as span:
+            span.set_attribute("job_id", os.getenv("FD_JOB_ID", default="null"))
+            pass
+    except:
+        pass
+
+
+def fd_start_span(span_name, kind=trace.SpanKind.CLIENT):
+    """
+    when fd start, start a new span show start success
+    """
+    try:
+        if not traces_enable:
+            return
+        with tracer.start_as_current_span(span_name, kind=kind) as span:
+            span.set_attribute("job_id", os.getenv("FD_JOB_ID", default="null"))
             pass
     except:
         pass
@@ -191,7 +207,8 @@ def start_span_request(span_name, request, kind=trace.SpanKind.CLIENT):
             return
         # extract Trace context from request.metadata.trace_carrier
         ctx = extract_from_request(request)
-        with tracer.start_as_current_span(span_name, context=ctx, kind=kind):
+        with tracer.start_as_current_span(span_name, context=ctx, kind=kind) as span:
+            span.set_attribute("job_id", os.getenv("FD_JOB_ID", default="null"))
             pass
     except:
         pass