LangChain 和 LangGraph 应用程序支持 基于 OpenTelemetry 的追踪。您可以将追踪数据通过您控制的 OpenTelemetry 收集器进行路由，应用脱敏规则去除敏感字段，然后将清理后的追踪数据转发到 LangSmith，而无需直接发送到 LangSmith。

Traces flow from your application to the collector over OTLP/HTTP. The collector runs a transform processor that redacts sensitive span attributes, such as prompt inputs and model completions, before forwarding the sanitized spans to the LangSmith API.

flowchart TD
    A["Application<br/>(LangChain / LangGraph)"]

    subgraph collector[":4318"]
        B["Receiver<br/>OTLP/HTTP"]
        C["Transform Processor<br/>PII Redaction<br/>(email, phone, SSN, CC)"]
        D["OTLP/HTTP Exporter"]
        B --> C --> D
    end

    E["LangSmith API<br/>api.smith.langchain.com"]

    A -->|"OTLP/HTTP"| B
    D -->|"OTLP/HTTP"| E

前提条件

以下两种方法都需要以下环境变量。将 OTEL_EXPORTER_OTLP_ENDPOINT 设置为您的收集器地址：

LANGSMITH_OTEL_ENABLED="true"
LANGSMITH_TRACING="true"
LANGSMITH_OTEL_ONLY="true"
LANGSMITH_PROJECT="my-project"
OTEL_EXPORTER_OTLP_ENDPOINT="http://<my-otel-collector-endpoint>:4318"

有关 LANGSMITH_PROJECT的更多信息，请参阅 将追踪数据记录到特定项目.

配置收集器

两种方法都需要一个 OpenTelemetry 收集器作为您的应用程序和 LangSmith 之间的中介。以下配置设置了一个在端口 4318上的 OTLP 接收器、一个用于脱敏 gen_ai.prompt 和 gen_ai.completion span 属性的转换处理器，以及一个将清理后的追踪数据转发到 LangSmith API 的导出器：

receivers:
  otlp:
    protocols:
      http:
        endpoint: 0.0.0.0:4318


processors:
  transform/redact:
    error_mode: ignore
    trace_statements:
      - context: span
        statements:
          - replace_pattern(attributes["gen_ai.completion"], "[\\s\\S]*", "[REDACTED]")
          - replace_pattern(attributes["gen_ai.prompt"], "[\\s\\S]*", "[REDACTED]")

exporters:
  otlphttp/langsmith:
    traces_endpoint: "https://api.smith.langchain.com/otel/v1/traces"
    headers:
      x-api-key: "${env:LANGSMITH_API_KEY}"
      Langsmith-Project: "${env:LANGSMITH_PROJECT}"


service:
  pipelines:
    traces:
      receivers: [otlp]
      processors: [transform/redact]
      exporters: [otlphttp/langsmith]

使用 LangChain 或 LangGraph 进行追踪

如果您的应用程序已经使用 LangChain or LangGraph，请使用此方法。追踪集成会根据您的环境变量自动处理 span 创建，因此无需额外的插桩代码：

from langchain.agents import create_agent
from langchain.tools import tool
from langchain_openai import ChatOpenAI


@tool
def tell_joke(topic: str) -> str:
   llm = ChatOpenAI()
   response = llm.invoke(f"Tell me a short, funny joke about {topic}.")
   return response.content


agent = create_agent(
   model=ChatOpenAI(),
   tools=[tell_joke],
   system_prompt="When the user asks for jokes, use the tell_joke tool for each topic.",
)


topics = ["programming", "python", "kubernetes", "machine learning"]


result = agent.invoke(
   {"messages": [{"role": "user", "content": f"Tell me jokes about these topics: {', '.join(topics)}"}]}
)


print(result["messages"][-1].content)

直接使用 OpenTelemetry SDK 进行追踪

如果您需要对追踪提供程序和导出器进行编程控制，请使用此方法。例如，在运行时设置每个请求的项目名称或配置自定义标头。您需要在代码中显式配置提供程序，而不是仅依赖环境变量：

from langchain.prompts import ChatPromptTemplate
from langchain_openai import ChatOpenAI
from opentelemetry import trace
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import BatchSpanProcessor


project_name = os.environ["LANGSMITH_PROJECT"]
otlp_endpoint = os.environ["OTEL_EXPORTER_OTLP_ENDPOINT"]


provider = TracerProvider()
provider.add_span_processor(
   BatchSpanProcessor(
       OTLPSpanExporter(
           endpoint=otlp_endpoint+"/v1/traces",
           headers={"Langsmith-Project": project_name},
       )
   )
)
trace.set_tracer_provider(provider)


chain = ChatPromptTemplate.from_template("Tell me a joke about {topic}") | ChatOpenAI()


for topic in ["programming", "python", "databases", "kubernetes", "machine learning"]:
   print(f"Asking about {topic}...")
   result = chain.invoke({"topic": topic})
   print(f"  {result.content[:100]}\n")


provider.force_flush()
provider.shutdown()