Restore from Backup

# Verify backup file integrity
jq empty backup.json && echo "✅ Valid JSON" || echo "❌ Invalid JSON"

# Check environment match
BACKUP_ENV=$(jq -r '.environment' backup.json)
CURRENT_ENV={ENVIRONMENT:-development}

if [ "$BACKUP_ENV" = "$CURRENT_ENV" ]; then
    echo "✅ Environment matches: $BACKUP_ENV"
else
    echo "❌ Environment mismatch: $BACKUP_ENV != $CURRENT_ENV"
fi

# Check backup contents
echo "📊 Backup contains $(jq '.flags | length' backup.json) flags"

# Create pre-restore backup
TIMESTAMP=$(date +%Y%m%d-%H%M%S)
curl -o "pre-restore-backup-$TIMESTAMP.json"      "http://localhost:3000/migration/export"

echo "📦 Pre-restore backup saved: pre-restore-backup-$TIMESTAMP.json"

Example restore steps for production environment:

1. DELETE current_only_flag (exists in current, not in backup)
2. CREATE_DEFAULTVALUE backup_flag (exists in backup, not current)
3. SET_VALUE backup_flag = true (restore backed up value)
4. UPDATE_SCHEMA_DEFAULTVALUE changed_flag (schema changed)
5. SET_VALUE changed_flag = "restored_value" (restore backed up value)

✅ All steps will be selected by default for complete restoration

# Scenario: Complete data loss in production
# Situation: etcd cluster failure, all flag data lost

1. Identify most recent backup file
   - flagflow_migration_production_20240810-020000.json (daily backup)
   
2. Verify backup integrity and environment match
   - Environment: production ✓
   - 47 flags in backup ✓
   - JSON structure valid ✓

3. Restore steps generated:
   - CREATE_DEFAULTVALUE for all 47 flags
   - SET_VALUE for all flags with non-default values
   
4. Execute complete restore to rebuild entire configuration

# Scenario: Recent changes caused issues, need to rollback
# Situation: Bad deployment 2 hours ago, rollback to pre-deploy state

1. Use pre-deployment backup
   - flagflow_migration_production_20240810-140000.json
   
2. Restore will generate steps to:
   - DELETE new flags created during deployment
   - UPDATE_SCHEMA_DEFAULTVALUE for modified flags
   - SET_VALUE to restore previous values
   
3. Execute to return to known good state

# Scenario: Reset development environment to baseline
# Situation: Development environment cluttered with test flags

1. Use baseline configuration backup
   - flagflow_migration_development_baseline.json
   
2. Restore will clean up:
   - DELETE all experimental flags
   - Reset all values to baseline state
   - Restore original flag schemas
   
3. Development environment returns to clean state

#!/bin/bash
# emergency-restore.sh - Quick restore for disaster recovery

BACKUP_FILE=$1
FLAGFLOW_URL="http://localhost:3000"

if [ -z "$BACKUP_FILE" ]; then
    echo "Usage: $0 <backup-file.json>"
    exit 1
fi

echo "🚨 Emergency Restore Starting..."
echo "📁 Backup file: $BACKUP_FILE"

# Verify backup file
if [ ! -f "$BACKUP_FILE" ]; then
    echo "❌ Backup file not found: $BACKUP_FILE"
    exit 1
fi

# Validate JSON
if ! jq empty "$BACKUP_FILE" 2>/dev/null; then
    echo "❌ Invalid JSON in backup file"
    exit 1
fi

# Check environment match
BACKUP_ENV=$(jq -r '.environment' "$BACKUP_FILE")
CURRENT_ENV={ENVIRONMENT:-development}

if [ "$BACKUP_ENV" != "$CURRENT_ENV" ]; then
    echo "❌ Environment mismatch: $BACKUP_ENV != $CURRENT_ENV"
    echo "   Use migration instead of restore for cross-environment operations"
    exit 1
fi

echo "✅ Environment verified: $CURRENT_ENV"
echo "📊 Backup contains $(jq '.flags | length' "$BACKUP_FILE") flags"

# Create pre-restore backup
echo "📦 Creating pre-restore backup..."
PRE_RESTORE_BACKUP="pre-restore-backup-$(date +%Y%m%d-%H%M%S).json"
curl -s "$FLAGFLOW_URL/migration/export" -o "$PRE_RESTORE_BACKUP"

if [ $? -eq 0 ]; then
    echo "✅ Pre-restore backup created: $PRE_RESTORE_BACKUP"
else
    echo "❌ Failed to create pre-restore backup"
    exit 1
fi

echo "🎯 Ready for restore. Upload $BACKUP_FILE to FlagFlow migration interface."
echo "🔄 Pre-restore backup available at: $PRE_RESTORE_BACKUP"
echo ""
echo "Manual steps:"
echo "1. Open $FLAGFLOW_URL/ui/migration"
echo "2. Upload $BACKUP_FILE"
echo "3. Click 'Analyze Migration'"
echo "4. Review restore steps and execute"

#!/bin/bash
# validate-restore-readiness.sh - Test restore capability regularly

BACKUP_DIR="/path/to/backups"
FLAGFLOW_URL="http://localhost:3000"
TEST_BACKUP=""

# Find most recent backup
TEST_BACKUP=$(ls -t "$BACKUP_DIR"/flagflow_migration_*.json | head -1)

if [ -z "$TEST_BACKUP" ]; then
    echo "❌ No backup files found in $BACKUP_DIR"
    exit 1
fi

echo "🧪 Testing restore readiness with: $(basename "$TEST_BACKUP")"

# Verify backup can be loaded by migration system
# Note: This would require API endpoint to validate without executing
curl -s -X POST "$FLAGFLOW_URL/migration/validate"      -F "file=@$TEST_BACKUP" | jq .

echo "✅ Restore validation completed"

# Scenario: Specific flags corrupted, others intact
# Strategy: Selective restore to recover only affected flags

1. Identify affected flags through application errors or monitoring
2. Load backup file and analyze which flags need restoration
3. Use selective restore:
   - Uncheck CREATE/DELETE steps for unaffected flags
   - Include only UPDATE and SET_VALUE steps for corrupted flags
4. Execute selective restore to fix specific issues

# Scenario: Flag schemas corrupted but values intact
# Strategy: Restore schemas while preserving current values

1. Load backup with correct schemas
2. Use migration analysis to identify schema differences
3. Execute only UPDATE_SCHEMA_DEFAULTVALUE steps
4. Skip SET_VALUE steps to preserve current values
5. Manually verify and adjust values if needed