feat: Full support for Astra DB as a backend

.env.example

@@ -40,6 +40,9 @@ FETCH_OPENRAG_DOCS_AT_STARTUP=false
 LANGFLOW_SECRET_KEY=
 
 # flow ids for chat and ingestion flows
+# OpenRAG automatically switches to the built-in Astra flow IDs when
+# VECTOR_BACKEND=astra. Leave these values alone unless you are replacing
+# the stock OpenSearch-backed flows with your own custom JSON files.
 LANGFLOW_CHAT_FLOW_ID=1098eea1-6649-4e1d-aed1-b77249fb8dd0
 LANGFLOW_INGEST_FLOW_ID=5488df7c-b93f-4f87-a446-b67028bc0813
 LANGFLOW_URL_INGEST_FLOW_ID=72c3d17c-2dac-4a73-b48a-6518473d7830
@@ -93,6 +96,23 @@ OPENSEARCH_USERNAME=admin
 # Change this if you want to use a different index name or avoid conflicts
 OPENSEARCH_INDEX_NAME=documents
 
+# Knowledge backend selection
+# - opensearch: store and search knowledge in OpenSearch (default)
+# - astra: use the built-in Astra-backed Langflow flows
+VECTOR_BACKEND=opensearch
+
+# Astra DB connection for VECTOR_BACKEND=astra
+ASTRA_DB_APPLICATION_TOKEN=
+ASTRA_DB_API_ENDPOINT=
+# Optional Astra DB keyspace/namespace override. Leave empty to use the database default.
+ASTRA_DB_KEYSPACE=
+
+# Optional: override the built-in Astra flow IDs (only needed for custom Astra flows)
+# ASTRA_CHAT_FLOW_ID=6df7ff83-97be-46be-b624-ab40ff53fd7b
+# ASTRA_INGEST_FLOW_ID=4ae67a0f-8f97-4563-ad54-3c18fe07ce30
+# ASTRA_URL_INGEST_FLOW_ID=fbd0aee6-29b1-4a51-a802-27fb3bec9927
+# ASTRA_NUDGES_FLOW_ID=05b262d8-45ba-4574-afad-797e8918defd
+
 # IBM AMS Authentication (IBM Watsonx Data embedded mode)
 # Set IBM_AUTH_ENABLED=true to authenticate via the ibm-openrag-session cookie
 # instead of Google OAuth. The raw IBM JWT is also passed directly to OpenSearch.
@@ -231,7 +251,6 @@ LOG_LEVEL=
 SERVICE_NAME=openrag
 # Secret key for session management (auto-generated if not provided)
 SESSION_SECRET=
-
 # OPTIONAL: JWT signing key for token issuance and verification.
 # If not set, OpenRAG generates an RSA key pair (private_key.pem / public_key.pem)
 # under OPENRAG_KEYS_PATH at startup and uses RS256 signing.

docker-compose.yml

@@ -70,6 +70,7 @@ services:
       - LANGFLOW_CHAT_FLOW_ID=${LANGFLOW_CHAT_FLOW_ID}
       - LANGFLOW_INGEST_FLOW_ID=${LANGFLOW_INGEST_FLOW_ID}
       - LANGFLOW_URL_INGEST_FLOW_ID=${LANGFLOW_URL_INGEST_FLOW_ID}
+      - VECTOR_BACKEND=${VECTOR_BACKEND:-opensearch}
       - DISABLE_INGEST_WITH_LANGFLOW=${DISABLE_INGEST_WITH_LANGFLOW:-false}
       - INGEST_SAMPLE_DATA=${INGEST_SAMPLE_DATA:-true}
       - NUDGES_FLOW_ID=${NUDGES_FLOW_ID}
@@ -100,6 +101,13 @@ services:
       - IBM_COS_HMAC_SECRET_ACCESS_KEY=${IBM_COS_HMAC_SECRET_ACCESS_KEY}
       - IBM_COS_AUTH_ENDPOINT=${IBM_COS_AUTH_ENDPOINT}
       - OPENSEARCH_INDEX_NAME=${OPENSEARCH_INDEX_NAME:-documents}
+      - ASTRA_DB_APPLICATION_TOKEN=${ASTRA_DB_APPLICATION_TOKEN}
+      - ASTRA_DB_API_ENDPOINT=${ASTRA_DB_API_ENDPOINT}
+      - ASTRA_DB_KEYSPACE=${ASTRA_DB_KEYSPACE}
+      - ASTRA_CHAT_FLOW_ID=${ASTRA_CHAT_FLOW_ID}
+      - ASTRA_INGEST_FLOW_ID=${ASTRA_INGEST_FLOW_ID}
+      - ASTRA_URL_INGEST_FLOW_ID=${ASTRA_URL_INGEST_FLOW_ID}
+      - ASTRA_NUDGES_FLOW_ID=${ASTRA_NUDGES_FLOW_ID}
       - LANGFLOW_KEY=${LANGFLOW_KEY}
       - LANGFLOW_KEY_RETRIES=${LANGFLOW_KEY_RETRIES:-15}
       - LANGFLOW_KEY_RETRY_DELAY=${LANGFLOW_KEY_RETRY_DELAY:-2.0}
@@ -185,12 +193,15 @@ services:
       - OPENSEARCH_PORT=${LANGFLOW_OPENSEARCH_PORT:-${OPENSEARCH_PORT:-9200}}
       - OPENSEARCH_URL=https://${LANGFLOW_OPENSEARCH_HOST:-${OPENSEARCH_HOST:-opensearch}}:${LANGFLOW_OPENSEARCH_PORT:-${OPENSEARCH_PORT:-9200}}
       - OPENSEARCH_INDEX_NAME=${OPENSEARCH_INDEX_NAME:-documents}
+      - VECTOR_BACKEND=${VECTOR_BACKEND:-opensearch}
+      - ASTRA_DB_APPLICATION_TOKEN=${ASTRA_DB_APPLICATION_TOKEN}
+      - ASTRA_DB_API_ENDPOINT=${ASTRA_DB_API_ENDPOINT}
       - DOCLING_SERVE_URL=${DOCLING_SERVE_URL:-http://host.docker.internal:5001}
       - FILENAME=None
       - MIMETYPE=None
       - FILESIZE=0
       - SELECTED_EMBEDDING_MODEL=${SELECTED_EMBEDDING_MODEL:-text-embedding-3-small}
-      - LANGFLOW_VARIABLES_TO_GET_FROM_ENVIRONMENT=JWT,OPENRAG-QUERY-FILTER,OPENSEARCH_PASSWORD,OPENSEARCH_URL,DOCLING_SERVE_URL,OWNER,OWNER_NAME,OWNER_EMAIL,CONNECTOR_TYPE,DOCUMENT_ID,SOURCE_URL,ALLOWED_USERS,ALLOWED_GROUPS,FILENAME,MIMETYPE,FILESIZE,SELECTED_EMBEDDING_MODEL,OPENAI_API_KEY,ANTHROPIC_API_KEY,WATSONX_APIKEY,WATSONX_URL,WATSONX_PROJECT_ID,OLLAMA_BASE_URL,OPENSEARCH_INDEX_NAME
+      - LANGFLOW_VARIABLES_TO_GET_FROM_ENVIRONMENT=JWT,OPENRAG-QUERY-FILTER,OPENSEARCH_PASSWORD,OPENSEARCH_URL,DOCLING_SERVE_URL,OWNER,OWNER_NAME,OWNER_EMAIL,CONNECTOR_TYPE,DOCUMENT_ID,SOURCE_URL,ALLOWED_USERS,ALLOWED_GROUPS,FILENAME,MIMETYPE,FILESIZE,SELECTED_EMBEDDING_MODEL,OPENAI_API_KEY,ANTHROPIC_API_KEY,WATSONX_APIKEY,WATSONX_URL,WATSONX_PROJECT_ID,OLLAMA_BASE_URL,OPENSEARCH_INDEX_NAME,ASTRA_DB_APPLICATION_TOKEN,ASTRA_DB_API_ENDPOINT,ASTRA_DB_KEYSPACE
       - LANGFLOW_LOG_LEVEL=DEBUG
       - LANGFLOW_WORKERS=${LANGFLOW_WORKERS:-1}
       - LANGFLOW_AUTO_LOGIN=${LANGFLOW_AUTO_LOGIN}

docs/docs/reference/configuration.mdx

@@ -80,6 +80,25 @@ For Langflow flow IDs and Langflow timeout settings, see [Langflow settings](#la
 | `INGESTION_TIMEOUT` | `3600` | Document ingestion timeout limit in seconds for each file. Increase this value if you experience timeouts when ingesting very large documents. Must be greater than or equal to [`LANGFLOW_TIMEOUT`](#langflow-settings).  |
 | `UPLOAD_BATCH_SIZE` | `25` | When ingesting folders, set the maximum number of files to ingest per batch. Each batch is an [ingestion task](/ingestion#monitor-ingestion). Increase this value to ingest more files per batch. If this value is too high, performance issues can occur. |
 
+## Knowledge backend settings
+
+Select the vector backend that OpenRAG should use for Langflow ingestion and retrieval.
+
+| Variable | Default | Description |
+|----------|---------|-------------|
+| `VECTOR_BACKEND` | `opensearch` | Select the active knowledge backend. Use `opensearch` for the default OpenSearch-backed flows, or `astra` to switch OpenRAG to the built-in Astra DB Langflow flows. |
+| `ASTRA_DB_APPLICATION_TOKEN` | Not set | Required when `VECTOR_BACKEND=astra`. Astra DB application token used by the Astra-backed Langflow vector store component. |
+| `ASTRA_DB_API_ENDPOINT` | Not set | Required when `VECTOR_BACKEND=astra`. Astra DB Data API endpoint used by the Astra-backed Langflow vector store component. |
+| `ASTRA_DB_KEYSPACE` | Not set | Optional when `VECTOR_BACKEND=astra`. Astra DB keyspace/namespace to use instead of the database default. |
+| `ASTRA_CHAT_FLOW_ID` | Built-in | Optional. Override the default Astra chat/retrieval flow ID. Only needed when replacing the built-in Astra flows with custom ones. |
+| `ASTRA_INGEST_FLOW_ID` | Built-in | Optional. Override the default Astra ingestion flow ID. |
+| `ASTRA_URL_INGEST_FLOW_ID` | Built-in | Optional. Override the default Astra URL ingestion flow ID. |
+| `ASTRA_NUDGES_FLOW_ID` | Built-in | Optional. Override the default Astra nudges flow ID. |
+
+:::warning
+The Astra backend stores one vector representation per collection. If you change the embedding provider or model for an existing Astra-backed corpus, reingest the documents so all stored vectors are regenerated with the new embedding configuration.
+:::
+
 ## Langflow settings {#langflow-settings}
 
 Configure the OpenRAG Langflow server's authentication, contact point, and built-in flow definitions.
@@ -103,7 +122,7 @@ For better security, it is recommended to set `LANGFLOW_SUPERUSER_PASSWORD` so t
 | `LANGFLOW_SECRET_KEY` | Automatically generated | Secret encryption key for Langflow internal operations. It is recommended to [generate your own Langflow secret key](https://docs.langflow.org/api-keys-and-authentication#langflow-secret-key) for this variable. If this variable isn't set, then Langflow generates a secret key automatically.  |
 | `LANGFLOW_SUPERUSER` | `admin` | Username for the Langflow administrator user. |
 | `LANGFLOW_SUPERUSER_PASSWORD` | Not set | Langflow administrator password. If this variable isn't set, then the Langflow server starts _without_ authentication enabled. It is recommended to set `LANGFLOW_SUPERUSER_PASSWORD` so the [Langflow server starts with authentication enabled](https://docs.langflow.org/api-keys-and-authentication#start-a-langflow-server-with-authentication-enabled). |
-| `LANGFLOW_CHAT_FLOW_ID`, `LANGFLOW_INGEST_FLOW_ID`, `NUDGES_FLOW_ID`, `LANGFLOW_URL_INGEST_FLOW_ID` | Built-in flow IDs | These variables are set automatically to the IDs of the chat, Docling ingestion, URL ingestion, and nudges [flows](/agents). The default values are found in [`.env.example`](https://github.com/langflow-ai/openrag/blob/main/.env.example). Only change these values if you want to replace a built-in flow with your own custom flow. The flow JSON must be present in your version of the OpenRAG codebase. For example, if you [deploy self-managed services](/docker), you can add the flow JSON to your local clone of the OpenRAG repository before deploying OpenRAG. |
+| `LANGFLOW_CHAT_FLOW_ID`, `LANGFLOW_INGEST_FLOW_ID`, `NUDGES_FLOW_ID`, `LANGFLOW_URL_INGEST_FLOW_ID` | Built-in flow IDs | These variables are set automatically to the IDs of the stock OpenSearch-backed chat, Docling ingestion, URL ingestion, and nudges [flows](/agents). The default values are found in [`.env.example`](https://github.com/langflow-ai/openrag/blob/main/.env.example). Only change these values if you want to replace a built-in flow with your own custom flow. The flow JSON must be present in your version of the OpenRAG codebase. For example, if you [deploy self-managed services](/docker), you can add the flow JSON to your local clone of the OpenRAG repository before deploying OpenRAG. When `VECTOR_BACKEND=astra`, OpenRAG automatically switches to the built-in Astra flow set. |
 | `LANGFUSE_SECRET_KEY` | Not set | Optional Langfuse secret key to enable the [Langflow integration with Langfuse](https://docs.langflow.org/integrations-langfuse). |
 | `LANGFUSE_PUBLIC_KEY` | Not set | Optional Langfuse public key to enable the [Langflow integration with Langfuse](https://docs.langflow.org/integrations-langfuse). |
 | `LANGFUSE_HOST` | Not set | Leave empty for Langfuse Cloud. Required for self-hosted Langfuse deployments if `LANGFUSE_SECRET_KEY` and `LANGFUSE_PUBLIC_KEY` are set. The address must be relative to the OpenRAG container deployment. For example, `http://localhost:3002` or `http://host.docker.internal:3000`. |