refactor(phase-9): isolate Ray into services/workers, delete legacy indexer components

.env.ollama

@@ -0,0 +1,49 @@
+# LLM
+BASE_URL=http://host.docker.internal:11434/v1
+API_KEY=ollama
+MODEL=qwen2.5:0.5b
+LLM_SEMAPHORE=1
+TIMEOUT=300
+
+# VLM (Vision support)
+VLM_BASE_URL=http://host.docker.internal:11434/v1
+VLM_API_KEY=ollama
+VLM_MODEL=qwen3-vl:2b
+VLM_SEMAPHORE=1
+
+# EMBEDDER
+EMBEDDER_MODEL_NAME=dengcao/Qwen3-Embedding-0.6B:Q8_0
+EMBEDDER_BASE_URL=http://host.docker.internal:11434/v1
+EMBEDDER_API_KEY=ollama
+MAX_MODEL_LEN=2048
+
+# RAG CONFIG
+RAG_MODE=SimpleRag
+CONTEXTUAL_RETRIEVAL=false
+RERANKER_ENABLED=false
+MAX_OUTPUT_TOKENS=2048
+
+# App Settings
+# Docker Compose expands PWD from the shell; customize if your shell does not set it.
+SHARED_ENV=${PWD}/.env.ollama
+APP_PORT=8002
+CHAINLIT_PORT=8090
+DEFAULT_LANGUAGE=en-US
+INDEXERUI_PORT=8060
+
+# Chainlit conversation history (SQLAlchemy -> existing rdb PostgreSQL)
+CHAINLIT_DATABASE_URL=postgresql+asyncpg://root:root_password@rdb:5432/chainlit
+DATABASE_URL=postgresql://root:root_password@rdb:5432/chainlit
+CHAINLIT_AUTH_SECRET=openrag_local_dev_secret_2026
+
+# RAY & System
+RAY_ENABLE_UV_RUN_RUNTIME_ENV=0
+RAY_NUM_GPUS=0
+RAY_DASHBOARD_PORT=8265
+RAY_memory_usage_threshold=0.99
+RAY_memory_monitor_refresh_ms=0
+SUPER_ADMIN_MODE=true
+AUTH_TOKEN=sk-1234
+SAVE_UPLOADED_FILES=true
+PDFLoader=PyMuPDFLoader
+WHISPER_N_WORKERS=0

.github/workflows/integration_tests.yml

@@ -0,0 +1,84 @@
+name: Integration tests
+
+on:
+  push:
+    branches:
+      - main
+      - dev
+      - "refactor/hexagonal"
+      - "refactor/phase-**"
+  pull_request:
+
+jobs:
+  milvus-integration:
+    runs-on: ubuntu-latest
+    timeout-minutes: 20
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Set up Docker Compose
+        uses: docker/setup-compose-action@v1
+        with:
+          version: v2.34.0
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install uv
+        run: |
+          curl -LsSf https://astral.sh/uv/install.sh | sh
+          echo "$HOME/.local/bin" >> $GITHUB_PATH
+
+      - name: Cache uv
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cache/uv
+          key: uv-${{ runner.os }}-3.12-${{ hashFiles('**/pyproject.toml', '**/uv.lock') }}
+          restore-keys: |
+            uv-${{ runner.os }}-3.12-
+
+      - name: Install dependencies
+        run: uv sync
+
+      - name: Start Milvus stack
+        working-directory: tests/integration
+        run: docker compose up -d --wait --wait-timeout 180
+
+      - name: Verify Milvus reachable from host
+        run: |
+          for i in {1..30}; do
+            if curl -sf http://localhost:9091/healthz 2>/dev/null; then
+              echo "Milvus healthy after $i attempts"
+              break
+            fi
+            echo "Attempt $i/30 - waiting..."
+            sleep 2
+          done
+          curl -sf http://localhost:9091/healthz
+
+      - name: Run integration tests
+        env:
+          OPENRAG_TEST_VDB_HOST: localhost
+          OPENRAG_TEST_VDB_PORT: "19530"
+        run: uv run pytest tests/integration/ -m integration -v --tb=short
+
+      - name: Show logs on failure
+        if: failure()
+        working-directory: tests/integration
+        run: |
+          echo "=== Milvus Logs ==="
+          docker compose logs milvus --tail=200
+          echo "=== etcd Logs ==="
+          docker compose logs etcd --tail=50
+          echo "=== minio Logs ==="
+          docker compose logs minio --tail=50
+
+      - name: Cleanup
+        if: always()
+        working-directory: tests/integration
+        run: docker compose down -v

.github/workflows/layer_guard.yml

@@ -0,0 +1,21 @@
+name: Layer guard
+
+on:
+  push:
+    branches:
+      - "refactor/hexagonal"
+      - "refactor/phase-**"
+  pull_request:
+    branches:
+      - "refactor/hexagonal"
+
+jobs:
+  layer-import-guard:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - name: Run layer guard
+        run: python scripts/check_layer_imports.py

.github/workflows/lint.yml

@@ -2,9 +2,16 @@ name: Linting
 
 on:
   push:
-    branches: [ main, dev ]  
+    branches:
+      - main
+      - dev
+      - "refactor/hexagonal"
+      - "refactor/phase-**"
   pull_request:
-    branches: [ main, dev ]
+    branches:
+      - main
+      - dev
+      - "refactor/hexagonal"
 
 jobs:
   lint:

.github/workflows/unit_tests.yml

@@ -2,7 +2,11 @@ name: Unit tests
 
 on:
   push:
-    branches: [ main, dev ]
+    branches:
+      - main
+      - dev
+      - "refactor/hexagonal"
+      - "refactor/phase-**"
   pull_request:
 
 jobs:

.gitignore

@@ -59,9 +59,9 @@ volumes/*
 !volumes/.gitkeep # Keep the placeholder
 
 # services
-services/
-services/*
-!services/.gitkeep # Keep the placeholder
+/services/
+/services/*
+!/services/.gitkeep # Keep the placeholder
 
 *.csv
 *.pkl

FORWARD_PORT_LOG.md

@@ -0,0 +1,27 @@
+# Forward Port Log
+
+Tracks `dev` changes during MODE 2 isolation (Phases 5-9).
+Each entry: what changed on `dev`, whether it was forward-ported or deferred
+to the cutover re-implementation queue.
+
+> Created retroactively at the start of Phase 5 (2026-04-29). The Phase 0-4
+> Mode 1 work merged from `dev` cleanly so this log starts empty.
+
+---
+
+## Forward-ported (critical)
+
+_Security fixes, data-loss bugs, production outages re-implemented against
+the new architecture. Each entry pairs the dev commit with the refactor
+commit so a reviewer can audit equivalence._
+
+(none yet)
+
+## Deferred to cutover (features)
+
+_Non-critical changes that landed on `dev` during MODE 2. These will be
+re-implemented directly in the new architecture during MODE 3 (Phases
+10-12) or post-cutover. List the dev PR number / commit and the target
+location in the new layout._
+
+(none yet)