Skip to main content
GET
/
health
QA Health Check
curl --request GET \
  --url https://api.example.com/health
{
  "ready": true
}

​
QA API Health Check

Checks the readiness status of the LangChain QA API service.

​
Endpoint

GET /health

​
Response

Returns the operational status of the QA API and its components.
​
ready
boolean
required
Overall readiness status. Returns true only when both the retriever and LLM are successfully initialized.Returns false if OPENAI_API_KEY is not configured or initialization failed.

​
Status Codes

  • 200 OK - Health check completed successfully

​
Example Request

cURL
curl -X GET "http://localhost:8001/health" \
  -H "accept: application/json"

​
Example Responses

​
Service Ready

200 OK
{
  "ready": true
}

​
Service Not Ready

200 OK
{
  "ready": false
}

​
Implementation Details

Defined in src/qa_api.py:206-208 Response Model: HealthResponse (src/qa_api.py:47-48) 
class HealthResponse(BaseModel):
    ready: bool
The endpoint checks:
  • retriever: Chroma vector store retriever for document search
  • llm: ChatOpenAI language model instance
Both must be non-None for the service to be considered ready.

​
Initialization Requirements

The service requires these environment variables:

​
Required

  • OPENAI_API_KEY - OpenAI API key for embeddings and chat completions

​
Optional

  • QA_TRANSCRIPT_PDF - Path to course transcript PDF (default: tableau_course_transcript.pdf)
  • OPENAI_CHAT_MODEL - Chat model name (default: gpt-4o-mini)
  • OPENAI_EMBEDDING_MODEL - Embedding model name (default: text-embedding-3-small)
  • QA_CHROMA_COLLECTION - Chroma collection name (default: tableau_qa_collection)

​
Initialization Process

On startup (src/qa_api.py:150-203):
  1. Check for OPENAI_API_KEY
  2. Load course transcript PDF or use fallback sample text
  3. Split transcript into chunks using Markdown headers
  4. Generate embeddings and create Chroma vector store
  5. Initialize retriever with k=4 search results
  6. Create ChatOpenAI instance with temperature=0
  7. Build chat prompt template
If any step fails or OPENAI_API_KEY is missing, the service starts but returns ready: false.

​
Use Cases

  • Container health probes - Kubernetes liveness/readiness checks
  • Load balancer routing - Ensure traffic only goes to initialized instances
  • Service monitoring - Track QA API availability
  • Pre-request validation - Check service is ready before sending questions
  • QA Ask - Submit questions to the QA system
  • QA Stream - Stream answers in real-time

Build docs developers (and LLMs) love

Get started for freeTalk to us