🔧 Boot Test Plan - Phase 1D Verification

Date: December 28, 2025
Goal: Verify vitruvyan-core boots without crashes after node neutralization

🎯 Minimal Stack for Testing

Core Services (Required):

postgres (omni_postgres) - Port 9432
redis (omni_redis) - Port 9379
qdrant (omni_qdrant) - Ports 9333, 9334
vitruvyan_api_graph - Port 9004 (LangGraph orchestration)
vitruvyan_api_neural - Port 9003 (Neural Engine)

Optional (can skip for minimal test):

embedding (omni_api_embedding) - Port 9010
babel_gardens - Port 9009
semantic - Port 9002
monitoring (prometheus, grafana)

📋 Test Sequence

Step 1: Check Dependencies

# Check if Docker is installed
docker --version
docker-compose --version
 
# Check if Python packages exist
ls -la infrastructure/docker/requirements/

Step 2: Start Infrastructure Only

cd /home/caravaggio/projects/vitruvyan-core/infrastructure/docker
docker-compose up -d postgres redis qdrant
docker ps
docker logs omni_postgres
docker logs omni_redis
docker logs omni_qdrant

Step 3: Start Neural Engine

docker-compose up -d vitruvyan_api_neural
docker logs -f omni_api_neural
# Wait for: "Application startup complete"

Step 4: Start Graph API (LangGraph)

docker-compose up -d vitruvyan_api_graph
docker logs -f omni_api_graph
# Wait for: "Application startup complete"
# Watch for: DOMAIN_NEUTRAL logs from neutralized nodes

Step 5: Health Checks

# Check all containers running
docker ps --format "table {{.Names}}\t{{.Status}}\t{{.Ports}}"
 
# Test endpoints
curl -s http://localhost:9004/health || echo "Graph API not ready"
curl -s http://localhost:9003/health || echo "Neural Engine not ready"

Step 6: Conversational Test

# Minimal test payload to graph API
curl -X POST http://localhost:9004/api/v1/chat \
  -H "Content-Type: application/json" \
  -d '{
    "user_id": "test_user",
    "input_text": "Analyze entity X under uncertainty",
    "language": "en"
  }' | jq .
 
# Expected response:
# - NO crashes
# - domain_neutral flags present
# - DOMAIN_NEUTRAL logs in docker logs

✅ Success Criteria

Infrastructure boots: postgres, redis, qdrant healthy
Neural Engine starts: No import errors, listens on 9003
Graph API starts: LangGraph initializes, listens on 9004
Nodes execute: entity_resolver, screener, collection, advisor log DOMAIN_NEUTRAL
No crashes: All containers stay running for 2+ minutes
Logs clear: No Python exceptions, no missing module errors

❌ Expected Failures (Acceptable)

Empty/neutral responses (expected - domain plugin not implemented)
"Not implemented" messages
Zero analysis results
NO_ACTION recommendations

🚨 Red Flags (Stop and Fix)

ImportError: Module not found
Syntax errors in neutralized nodes
Container crashes after 30 seconds
LangGraph fails to build state graph
Database connection failures

📝 Logging Strategy

# Terminal 1: Follow graph API logs
docker logs -f omni_api_graph 2>&1 | grep -E "(DOMAIN_NEUTRAL|ERROR|entity_resolver|entity_screener|collection_analyzer|decision_advisor)"
 
# Terminal 2: Follow neural engine logs
docker logs -f omni_api_neural 2>&1 | grep -E "(ERROR|startup|health)"
 
# Terminal 3: Check all errors
docker-compose logs --tail=50 | grep ERROR

🔄 Rollback Plan

If boot test fails catastrophically:

# Stop all containers
docker-compose down
 
# Restore original nodes from backups
cd /home/caravaggio/projects/vitruvyan-core/vitruvyan_core/core/orchestration/langgraph/node
mv entity_resolver_node.py.backup entity_resolver_node.py
mv screener_node.py.backup screener_node.py
mv portfolio_node.py.backup portfolio_node.py
mv advisor_node.py.backup advisor_node.py
 
# Retry boot

📊 Test Results Log

Start Time: [TBD]
Infrastructure Status: [TBD]
Neural Engine Status: [TBD]
Graph API Status: [TBD]
Node Execution Status: [TBD]
Conversational Test Result: [TBD]

Vitruvyan Docs