Replace ChromaDB with pgvector for vector storage

Consolidate onto PostgreSQL by using pgvector instead of a separate
ChromaDB instance. This removes a Docker volume, a large dependency,
and simplifies the stack without meaningful performance impact at
our document scale.

- Swap langchain-chroma for langchain-postgres (PGVector)
- Use pgvector/pgvector:pg16 Docker image with init script
- Lazy-initialize vector store to avoid eager DB connections
- Add SQL helpers for stats/delete/list (replacing _collection access)
- Remove legacy main.py, chunker, petmd scraper, and /api/query endpoint

Re-index required after deploy (POST /api/rag/index + /index-obsidian).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
2026-04-24 08:43:52 -04:00
parent 9ed4ca126a
commit 438399646f
19 changed files with 241 additions and 1690 deletions
+2 -4
View File
@@ -2,7 +2,7 @@ version: "3.8"
services:
postgres:
image: postgres:16-alpine
image: pgvector/pgvector:pg16
ports:
- "5432:5432"
environment:
@@ -11,6 +11,7 @@ services:
- POSTGRES_DB=${POSTGRES_DB:-raggr}
volumes:
- postgres_data:/var/lib/postgresql/data
- ./docker/init-pgvector.sql:/docker-entrypoint-initdb.d/init-pgvector.sql
healthcheck:
test: ["CMD-SHELL", "pg_isready -U ${POSTGRES_USER:-raggr}"]
interval: 10s
@@ -29,7 +30,6 @@ services:
- PAPERLESS_TOKEN=${PAPERLESS_TOKEN}
- BASE_URL=${BASE_URL}
- OLLAMA_URL=${OLLAMA_URL:-http://localhost:11434}
- CHROMADB_PATH=/app/data/chromadb
- OPENAI_API_KEY=${OPENAI_API_KEY}
- JWT_SECRET_KEY=${JWT_SECRET_KEY}
- LLAMA_SERVER_URL=${LLAMA_SERVER_URL}
@@ -66,10 +66,8 @@ services:
postgres:
condition: service_healthy
volumes:
- chromadb_data:/app/data/chromadb
- ./obvault:/app/data/obsidian
restart: unless-stopped
volumes:
chromadb_data:
postgres_data: