Skip to main content
Glama
RECOVERY_PROCEDURES.md8.01 kB
# EuConquisto Composer MCP - Recovery Procedures **Version**: 1.0.0 **Date**: July 5, 2025 **Purpose**: Emergency recovery and rollback procedures ## Quick Reference ### Emergency Commands ```bash # Full system rollback to v4.0.3 ./scripts/rollback-to-v4.0.3.sh # JWT token recovery only ./scripts/recover-jwt-token.sh # System health check ./scripts/system-health-check.sh # Create backup before changes ./scripts/backup-system.sh ``` ### Emergency Contact - **Migration Manager**: Claude Code - **Documentation**: `/docs/euconquisto-migration/` - **Backup Location**: `/backups/migration-safety/` --- ## Recovery Scenarios ### Scenario 1: Modular Infrastructure Failure **Symptoms**: New infrastructure modules not working, errors in module loading **Recovery Time**: < 5 minutes #### Steps: 1. **Immediate Action**: Stop all processes to prevent corruption 2. **Execute Rollback**: ```bash ./scripts/rollback-to-v4.0.3.sh ``` 3. **Verify System**: Run health check after rollback 4. **Test Functionality**: Create test composition to verify #### Success Criteria: - ✅ v4.0.3 system operational - ✅ JWT server running on port 8080 - ✅ Authentication working - ✅ Composition creation successful --- ### Scenario 2: JWT Token Corruption **Symptoms**: Authentication failures, "JWT token not found" errors **Recovery Time**: < 2 minutes #### Steps: 1. **Diagnose Issue**: Check token file and length ```bash ls -la /Users/ricardokawasaki/Desktop/euconquisto-composer-mcp-poc/archive/authentication/ wc -c /Users/ricardokawasaki/Desktop/euconquisto-composer-mcp-poc/archive/authentication/correct-jwt-new.txt ``` 2. **Execute Recovery**: ```bash ./scripts/recover-jwt-token.sh ``` 3. **Verify Recovery**: Check token length is 3276 characters 4. **Test Authentication**: Verify JWT server responds #### Success Criteria: - ✅ JWT token file restored - ✅ Token length = 3276 characters - ✅ JWT server responds to requests - ✅ Authentication flow working --- ### Scenario 3: Configuration Corruption **Symptoms**: npm errors, dependency issues, build failures **Recovery Time**: < 10 minutes #### Steps: 1. **Check Configuration**: ```bash ./scripts/system-health-check.sh ``` 2. **Restore Configuration**: Use full rollback to restore package.json ```bash ./scripts/rollback-to-v4.0.3.sh ``` 3. **Rebuild Dependencies**: ```bash npm install ``` 4. **Verify Build**: Check TypeScript compilation if needed #### Success Criteria: - ✅ package.json restored - ✅ Dependencies installed successfully - ✅ No npm errors - ✅ System builds without errors --- ### Scenario 4: Complete System Failure **Symptoms**: Nothing works, multiple components broken **Recovery Time**: < 15 minutes #### Steps: 1. **Full System Restore**: ```bash ./scripts/rollback-to-v4.0.3.sh --clean-modular ``` 2. **Rebuild Environment**: ```bash npm install ``` 3. **Start Services**: ```bash # JWT server will be started automatically by rollback script # Verify with: lsof -i :8080 ``` 4. **Complete Health Check**: ```bash ./scripts/system-health-check.sh ``` 5. **End-to-End Test**: Create a test composition #### Success Criteria: - ✅ All system components restored - ✅ All health checks pass - ✅ End-to-end workflow functional - ✅ System ready for normal operation --- ## Backup Management ### Creating Backups ```bash # Create comprehensive backup ./scripts/backup-system.sh # Verify backup creation ls -la backups/migration-safety/ ``` ### Backup Contents - **Tier 1 Essential**: v4.0.3 system, JWT token, server files - **Tier 2 Configuration**: Documentation, config files - **Tier 3 Historical**: Previous versions, development tools ### Backup Verification Each backup includes checksums for integrity verification: ```bash # Check backup integrity cd backups/migration-safety/backup_TIMESTAMP/ cat checksums/*.sha256 ``` --- ## Service Management ### JWT Redirect Server ```bash # Check if running lsof -i :8080 # Start manually node tools/servers/jwt-redirect-server-v1.0.2.js & # Stop if needed kill $(lsof -ti :8080) ``` ### Process Management ```bash # Stop all MCP processes pkill -f "mcp.*composer" # Stop JWT server kill $(lsof -ti :8080) # Check running processes ps aux | grep -E "(mcp|jwt|composer)" ``` --- ## Validation Procedures ### System Health Check ```bash ./scripts/system-health-check.sh ``` #### Health Check Categories: - **Node.js Environment**: Version, npm availability - **Playwright**: Browser automation capability - **Project Files**: package.json, node_modules - **v4.0.3 System**: Main file, syntax validation - **JWT Token**: File existence, length validation - **JWT Server**: File, syntax, running status - **Modular Infrastructure**: New modules (if implemented) - **External Services**: EuConquisto platform, API connectivity - **Backup System**: Scripts, existing backups ### Manual Validation ```bash # Check JWT token wc -c archive/authentication/correct-jwt-new.txt # Should output: 3276 # Check main file syntax node -c dist/browser-automation-api-direct-save-v4.0.3.js # Test JWT server curl -s http://localhost:8080 # Check Node.js version node --version # Should be v18 or higher ``` --- ## Troubleshooting ### Common Issues #### "Permission denied" errors ```bash chmod +x scripts/*.sh ``` #### "Command not found" errors ```bash # Check PATH includes node and npm which node which npm # Install missing dependencies npm install ``` #### Port 8080 in use ```bash # Find what's using the port lsof -i :8080 # Kill the process if safe kill $(lsof -ti :8080) ``` #### JWT token issues ```bash # Check token exists and has correct length ls -la archive/authentication/correct-jwt-new.txt wc -c archive/authentication/correct-jwt-new.txt # Recover from backup if needed ./scripts/recover-jwt-token.sh ``` --- ## Emergency Response Plan ### Critical Failure Response (< 30 minutes) #### Phase 1: Immediate (0-5 minutes) 1. **STOP**: Halt all processes to prevent data corruption 2. **ASSESS**: Determine scope of failure 3. **COMMUNICATE**: Document the issue #### Phase 2: Recovery (5-20 minutes) 1. **CHOOSE SCENARIO**: Select appropriate recovery procedure 2. **EXECUTE**: Run recovery scripts 3. **VERIFY**: Validate recovery with health checks #### Phase 3: Validation (20-30 minutes) 1. **TEST**: Run end-to-end functionality tests 2. **MONITOR**: Watch for additional issues 3. **DOCUMENT**: Record incident and lessons learned ### Escalation Procedures 1. **Level 1**: Script-based recovery (automated) 2. **Level 2**: Manual recovery using documentation 3. **Level 3**: Full system rebuild from backups --- ## Backup Schedule ### Automated Backups - **Before Phase 2**: Mandatory backup before content generation migration - **Daily**: During active development - **Pre-deployment**: Before any production changes ### Manual Backups Create backup before: - Major code changes - Configuration updates - Dependency upgrades - Migration phase transitions --- ## Recovery Testing ### Monthly Recovery Drills 1. Create test backup 2. Simulate failure scenario 3. Execute recovery procedure 4. Validate system functionality 5. Document any issues ### Recovery Time Objectives (RTO) - **JWT Token Recovery**: < 2 minutes - **Configuration Recovery**: < 10 minutes - **Full System Recovery**: < 15 minutes - **Complete Rebuild**: < 30 minutes --- ## Contact Information ### Support Resources - **Migration Documentation**: `/docs/euconquisto-migration/` - **System Health Check**: `./scripts/system-health-check.sh` - **Backup Location**: `/backups/migration-safety/` ### Emergency Procedures 1. **First**: Run appropriate recovery script 2. **Second**: Check system health 3. **Third**: Test functionality 4. **Finally**: Document the incident --- **Remember**: When in doubt, run the health check first to understand the current system state before attempting any recovery procedures.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/rkm097git/euconquisto-composer-mcp-poc'

If you have feedback or need assistance with the MCP directory API, please join our Discord server