Haystack 的可观测性

本 cookbook 演示如何使用 Litefuse 为你的 Haystack 应用提供实时可观测性。

什么是 Haystack？ Haystack 是由 deepset 开发的开源 Python 框架。其模块化设计允许用户实现自定义流水线来构建生产就绪的 LLM 应用，例如 RAG 流水线和先进的搜索系统。它与 Hugging Face Transformers、Elasticsearch、OpenSearch、OpenAI、Cohere、Anthropic 等都有集成，是各种规模团队中极受欢迎的框架。

什么是 Litefuse？ Litefuse 是开源的 AI Agent 可观测性与评估平台。它帮助团队协作管理 prompt、trace 应用、调试问题，并在生产环境中评估 LLM 系统。

快速开始

我们将通过一个简单示例演示如何使用 Haystack 并将其与 Litefuse 集成。

第 1 步：安装依赖

%pip install haystack-ai langfuse openinference-instrumentation-haystack

第 2 步：设置环境变量

配置你的 Litefuse API Key。可以通过注册 Litefuse Cloud 或自托管 Litefuse 获取。

import os
 
# Get keys for your project from the project settings page: https://litefuse.cloud
os.environ["LANGFUSE_PUBLIC_KEY"] = "pk-lf-..." 
os.environ["LANGFUSE_SECRET_KEY"] = "sk-lf-..." 
os.environ["LANGFUSE_BASE_URL"] = "https://litefuse.cloud"
 
# Your openai and serperdev api key
os.environ["OPENAI_API_KEY"] = "sk-proj-..."
os.environ["SERPERDEV_API_KEY"] = "..."

设置好环境变量后，我们就可以初始化 Langfuse 客户端了。get_client() 会使用环境变量中提供的凭证来初始化 Langfuse 客户端。

from langfuse import get_client
 
langfuse = get_client()
 
# Verify connection
if langfuse.auth_check():
    print("Langfuse client is authenticated and ready!")
else:
    print("Authentication failed. Please check your credentials and host.")

第 3 步：初始化 Haystack Instrumentation

现在我们初始化 OpenInference Haystack instrumentation。这个第三方 instrumentation 会自动捕获 Haystack 的操作并将 OpenTelemetry (OTel) span 导出到 Litefuse。

from openinference.instrumentation.haystack import HaystackInstrumentor
 
# Instrument the Haystack application
HaystackInstrumentor().instrument()

第 4 步：创建一个简单的 Haystack 应用

现在我们使用 OpenAI 模型和 SerperDev 搜索 API 创建一个简单的 Haystack 应用。

import os
 
from haystack.components.agents import Agent
from haystack.components.generators.chat import OpenAIChatGenerator
from haystack.dataclasses import ChatMessage
from haystack.tools import ComponentTool
from haystack.components.websearch import SerperDevWebSearch
 
search_tool = ComponentTool(component=SerperDevWebSearch())
 
basic_agent = Agent(
    chat_generator=OpenAIChatGenerator(model="gpt-4o-mini"),
    system_prompt="You are a helpful web agent.",
    tools=[search_tool],
)
 
result = basic_agent.run(messages=[ChatMessage.from_user("When was the first version of Haystack released?")])
 
print(result['last_message'].text)
 
langfuse.flush()

第 5 步：在 Litefuse 中查看 Trace

运行你的工作流后，登录 Litefuse 即可查看生成的 trace。你将看到每个工作流步骤的日志，以及 token 数、延迟和执行路径等指标。

Litefuse Trace 示例

Litefuse 中的公开示例 trace

Interoperability with the Python SDK

You can use this integration together with the Litefuse SDKs to add additional attributes to the observation.

The @observe() decorator provides a convenient way to automatically wrap your instrumented code and add additional attributes to the observation.

from langfuse import observe, propagate_attributes, get_client
 
langfuse = get_client()
 
@observe()
def my_llm_pipeline(input):
    # Add additional attributes (user_id, session_id, metadata, version, tags) to all spans created within this execution scope
    with propagate_attributes(
        user_id="user_123",
        session_id="session_abc",
        tags=["agent", "my-observation"],
        metadata={"email": "user@litefuse.ai"},
        version="1.0.0"
    ):
 
        # YOUR APPLICATION CODE HERE
        result = call_llm(input)
 
        return result
 
# Run the function
my_llm_pipeline("Hi")

Learn more about using the Decorator in the Langfuse SDK instrumentation docs.

Troubleshooting

No observations appearing

First, enable debug mode in the Python SDK:

export LANGFUSE_DEBUG="True"

Then run your application and check the debug logs:

OTel observations appear in the logs: Your application is instrumented correctly but observations are not reaching Litefuse. To resolve this:
1. Call langfuse.flush() at the end of your application to ensure all observations are exported.
2. Verify that you are using the correct API keys and base URL.
No OTel spans in the logs: Your application is not instrumented correctly. Make sure the instrumentation runs before your application code.

Unwanted observations in Litefuse

The Langfuse SDK is based on OpenTelemetry. Other libraries in your application may emit OTel spans that are not relevant to you. These still count toward your billable units, so you should filter them out. See Unwanted spans in Litefuse for details.

Missing attributes

Some attributes may be stored in the metadata object of the observation rather than being mapped to the Litefuse data model. If a mapping or integration does not work as expected, please raise an issue on GitHub.

Next Steps

Once you have instrumented your code, you can manage, evaluate and debug your application:

Manage prompts in Litefuse Add evaluation scores Run LLM-as-a-judge Evaluators Create datasets Create custom dashboards Test queries in the Playground

Google ADK Instructor

这个页面对你有帮助吗？

支持