RhesisConnector

Connect your pipelines to Rhesis for OpenTelemetry-based tracing of pipeline runs.

Key Features

• Integrates Rhesis OpenTelemetry tracing into your pipelines without connecting to other components.
• Automatically traces all pipeline operations when added to a pipeline.
• Captures detailed span data for pipeline runs, including inputs, outputs, and component-level timings.
• Supports optional per-run metadata via invocation_context to tag traces with session IDs, test run identifiers, or other context.
• Configurable trace name, project ID, environment, and custom span handlers.
• Returns the trace URL and trace ID for each pipeline run so you can link directly to the trace in the Rhesis UI.

Configuration

1. Drag the RhesisConnector component onto the canvas from the Component Library.
2. Click on the component to open the configuration panel.
3. On the General tab:
   1. Create a secret with your Rhesis API key. Use RHESIS_API_KEY as the secret key. For instructions, see Create Secrets. Get your API key from Rhesis.
   2. Set name to an identifier for the trace shown in the Rhesis UI.
   3. Optionally set RHESIS_BASE_URL to your Rhesis backend URL (defaults to http://localhost:8080).
4. Set HAYSTACK_CONTENT_TRACING_ENABLED=true in your environment before importing Haystack to capture component inputs and outputs on spans.

Connections

RhesisConnector doesn't connect to any other components. Add it to your pipeline and it automatically traces all pipeline operations.

It accepts an optional invocation_context dictionary at runtime for attaching per-run metadata to traces. It outputs name, trace_url, and trace_id values you can use to link to the trace in the Rhesis UI.

Source Code

To check this component's source code, open rhesis_connector.py in the Haystack Core Integrations repository.

Usage Examples

Basic Configuration

  RhesisConnector:
    type: haystack_integrations.components.connectors.rhesis.rhesis_connector.RhesisConnector
    init_parameters:
      name: rag-pipeline-trace
      api_key:
        type: env_var
        env_vars:
        - RHESIS_API_KEY
        strict: false
      base_url:
      project_id:
      environment:
      frontend_url:

Using the Component in a Pipeline

This is an example of a RAG pipeline with Rhesis tracing enabled. The RhesisConnector is added without connecting it to other components — it automatically traces all pipeline operations.

# haystack-pipeline
components:
  retriever:
    type: haystack_integrations.components.retrievers.opensearch.embedding_retriever.OpenSearchEmbeddingRetriever
    init_parameters:
      top_k: 10
      document_store:
        type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
        init_parameters:
          hosts:
          index: ''
          max_chunk_bytes: 104857600
          embedding_dim: 768
          return_embedding: false
          method:
          mappings:
          settings:
          create_index: true
          http_auth:
          use_ssl:
          verify_certs:
          timeout:

  prompt_builder:
    type: haystack.components.builders.prompt_builder.PromptBuilder
    init_parameters:
      required_variables: "*"
      template: |-
        Answer the question based on the provided context.

        Context:
        {% for document in documents %}
        {{ document.content }}
        {% endfor %}

        Question: {{ question }}
        Answer:

  llm:
    type: haystack.components.generators.chat.openai.OpenAIChatGenerator
    init_parameters:
      api_key:
        type: env_var
        env_vars:
        - OPENAI_API_KEY
        strict: false
      model: gpt-4o-mini

  RhesisConnector:
    type: haystack_integrations.components.connectors.rhesis.rhesis_connector.RhesisConnector
    init_parameters:
      name: rag-pipeline-trace
      api_key:
        type: env_var
        env_vars:
        - RHESIS_API_KEY
        strict: false
      base_url:
      project_id:
      environment:
      frontend_url:
      span_handler:

connections:
- sender: retriever.documents
  receiver: prompt_builder.documents
- sender: prompt_builder.prompt
  receiver: llm.messages

max_runs_per_component: 100

metadata: {}

inputs:
  query:
  - retriever.query
  - prompt_builder.question

outputs:
  replies: llm.replies

Parameters

Inputs

Parameter	Type	Description
invocation_context	Optional[Dict[str, Any]]	A dictionary with additional context for the invocation. Useful for tagging traces with session IDs, test run identifiers, or other metadata. These key-value pairs are visible in the Rhesis traces.

Outputs

Parameter	Type	Description
name	str	The name of the tracing component.
trace_url	str	The URL to the trace in the Rhesis UI.
trace_id	str	The ID of the trace.

Init Parameters

These are the parameters you can configure in Pipeline Builder:

Parameter	Type	Default	Description
name	str		The trace name shown in the Rhesis UI.
api_key	Optional[Secret]	Secret.from_env_var('RHESIS_API_KEY')	Rhesis API key for trace ingestion.
base_url	Optional[str]	None	Rhesis backend base URL. Defaults to the RHESIS_BASE_URL environment variable or http://localhost:8080.
project_id	Optional[str]	None	Rhesis project ID. Defaults to the RHESIS_PROJECT_ID environment variable. Resolved from the API key when omitted.
environment	Optional[str]	None	Deployment environment label (for example, production, staging). Defaults to the RHESIS_ENVIRONMENT environment variable or development.
frontend_url	Optional[str]	None	Frontend base URL used to build trace_url deep links. Defaults to the RHESIS_FRONTEND_URL environment variable.
span_handler	Optional[SpanHandler]	None	Optional custom handler for processing spans. Uses DefaultSpanHandler when omitted.

Run Method Parameters

These are the parameters you can configure for the component's run() method. This means you can pass these parameters at query time through the API, in Playground, or when running a job. For details, see Modify Pipeline Parameters at Query Time.

Parameter	Type	Default	Description
invocation_context	Optional[Dict[str, Any]]	None	A dictionary with additional context for this invocation. These key-value pairs are attached to the root trace in Rhesis.

Related Information

• Create Secrets
• Pipeline Components