# Deployment Fix Guide - Requirement Processor Migration Issue

## Problem Summary
The deployment failed due to a database migration constraint issue in the requirement processor service. The error was:
```
❌ Migration failed: 001_business_context_tables.sql - null value in column "service" of relation "schema_migrations" violates not-null constraint
```

## Root Cause
The requirement processor's migration system was using an outdated schema for the `schema_migrations` table that didn't include the required `service` field, while the main database migration system expected this field to be present and non-null.

## Fix Applied

### 1. Updated Migration Script (`migrate.py`)
- ✅ Updated `schema_migrations` table schema to include `service` field
- ✅ Modified `is_applied()` function to check by both version and service
- ✅ Updated `mark_applied()` function to include service and description
- ✅ Fixed `run_migration()` function to use service parameter

### 2. Fixed Migration Files
- ✅ Removed foreign key constraint from initial migration to avoid dependency issues
- ✅ The second migration already handles the constraint properly

### 3. Created Fix Script
- ✅ Created `scripts/fix-requirement-processor-migration.sh` to clean up and restart the service

## Deployment Steps

### Option 1: Use the Fix Script (Recommended)
```bash
cd /home/ubuntu/codenuk-backend-live
./scripts/fix-requirement-processor-migration.sh
```

### Option 2: Manual Fix
```bash
# 1. Stop the requirement processor
docker compose stop requirement-processor

# 2. Clean up failed migration records
PGPASSWORD="password" psql -h localhost -p 5432 -U postgres -d dev_pipeline << 'EOF'
DELETE FROM schema_migrations WHERE service = 'requirement-processor' OR version LIKE '%.sql';
EOF

# 3. Restart the service
docker compose up -d requirement-processor

# 4. Check status
docker compose ps requirement-processor
```

### Option 3: Full Redeploy
```bash
# Stop all services
docker compose down

# Clean up database (if needed)
PGPASSWORD="password" psql -h localhost -p 5432 -U postgres -d dev_pipeline << 'EOF'
DELETE FROM schema_migrations WHERE service = 'requirement-processor';
EOF

# Start all services
docker compose up -d
```

## Verification Steps

1. **Check Service Status**
   ```bash
   docker compose ps requirement-processor
   ```

2. **Check Migration Records**
   ```bash
   PGPASSWORD="password" psql -h localhost -p 5432 -U postgres -d dev_pipeline << 'EOF'
   SELECT service, version, applied_at, description 
   FROM schema_migrations 
   WHERE service = 'requirement-processor' 
   ORDER BY applied_at;
   EOF
   ```

3. **Check Service Logs**
   ```bash
   docker compose logs requirement-processor
   ```

4. **Test Health Endpoint**
   ```bash
   curl http://localhost:8001/health
   ```

## Expected Results

After the fix:
- ✅ Requirement processor service should start successfully
- ✅ Migration records should show proper service field
- ✅ Health endpoint should return 200 OK
- ✅ All other services should continue running normally

## Prevention

To prevent this issue in the future:
1. Always ensure migration scripts use the correct `schema_migrations` table schema
2. Include service field in all migration tracking
3. Test migrations in development before deploying to production
4. Use the shared migration system consistently across all services

## Troubleshooting

If the issue persists:
1. Check database connectivity
2. Verify PostgreSQL is running
3. Check disk space and memory
4. Review all service logs
5. Consider a full database reset if necessary

## Files Modified
- `services/requirement-processor/migrations/migrate.py` - Updated migration system
- `services/requirement-processor/migrations/001_business_context_tables.sql` - Removed FK constraint
- `scripts/fix-requirement-processor-migration.sh` - Created fix script