Automated Backup Systems: Disaster Recovery Planning

Data is Irreplaceable

Data loss can be catastrophic. Whether it's customer data, application databases, or configuration files, losing data means losing business. Automated backup systems aren't optional—they're essential infrastructure.

Backup Strategy

3-2-1 Rule

The industry standard:

• 3 copies: Original + 2 backups
• 2 different media: Different storage types
• 1 off-site: Geographic separation

Backup Types

Different backup strategies:

Full Backups

• Complete copy of all data
• Slow, but complete recovery
• Use for weekly/monthly backups

Incremental Backups

• Only changed data since last backup
• Fast, but requires full + all incrementals
• Use for daily backups

Differential Backups

• Changed data since last full backup
• Faster restore than incremental
• Balance between speed and storage

Database Backups

PostgreSQL

# Full backup
pg_dump -h localhost -U user -d database -F c -f backup.dump

# Automated daily backup script
#!/bin/bash
BACKUP_DIR="/backups/postgres"
DATE=$(date +%Y%m%d_%H%M%S)
pg_dump -h localhost -U user -d mydb -F c -f "$BACKUP_DIR/mydb_$DATE.dump"

# Keep only last 30 days
find $BACKUP_DIR -name "*.dump" -mtime +30 -delete

MySQL

# Full backup
mysqldump -u user -p database > backup.sql

# Automated backup
mysqldump -u user -p --all-databases | gzip > backup_$(date +%Y%m%d).sql.gz

MongoDB

# Full backup
mongodump --host localhost --db mydb --out /backups/mongodb

# With compression
mongodump --host localhost --db mydb --archive=/backups/mydb.archive --gzip

File System Backups

rsync for Incremental Backups

# Incremental backup script
rsync -avz --delete \
  --exclude 'node_modules' \
  --exclude '.git' \
  /source/directory/ \
  /backup/destination/

Tar Archives

# Create compressed archive
tar -czf backup_$(date +%Y%m%d).tar.gz /path/to/backup

# Automated weekly full backup
tar -czf /backups/weekly_$(date +%Y%m%d).tar.gz \
  --exclude=/backups \
  /home /var/www /etc

Off-Site Backups

Cloud Storage

Popular options:

AWS S3

• Highly durable (99.999999999%)
• Lifecycle policies for cost optimization
• Glacier for long-term storage
• Cross-region replication

Backblaze B2

• Cost-effective
• S3-compatible API
• Good performance
• Simpler pricing

Google Cloud Storage

• Integrated with GCP services
• Multiple storage classes
• Strong consistency

Backup to Cloud

# Upload to S3
aws s3 sync /local/backups s3://my-backup-bucket/backups/

# With lifecycle policy (move to Glacier after 30 days)
# Configured in S3 bucket settings

Encryption

Encrypt backups before uploading:

# Encrypt before upload
tar -czf - /data | openssl enc -aes-256-cbc -salt \
  -out backup_encrypted.tar.gz.enc \
  -pass file:/path/to/password

# Decrypt
openssl enc -d -aes-256-cbc -in backup_encrypted.tar.gz.enc \
  -out backup.tar.gz \
  -pass file:/path/to/password

Backup Automation

Cron Jobs

Schedule regular backups:

# Daily database backup at 2 AM
0 2 * * * /usr/local/bin/backup-database.sh

# Weekly full backup on Sunday at 1 AM
0 1 * * 0 /usr/local/bin/full-backup.sh

# Monthly archive backup
0 3 1 * * /usr/local/bin/monthly-archive.sh

Backup Scripts

Comprehensive backup script:

#!/bin/bash
set -e

BACKUP_DIR="/backups"
DATE=$(date +%Y%m%d_%H%M%S)
LOG_FILE="/var/log/backups.log"

log() {
  echo "[$(date)] $1" >> $LOG_FILE
}

# Database backup
log "Starting database backup..."
pg_dump -h localhost -U user -d mydb \
  -F c -f "$BACKUP_DIR/db_$DATE.dump"
log "Database backup completed"

# File backup
log "Starting file backup..."
tar -czf "$BACKUP_DIR/files_$DATE.tar.gz" /var/www
log "File backup completed"

# Upload to S3
log "Uploading to S3..."
aws s3 cp "$BACKUP_DIR/db_$DATE.dump" \
  s3://my-backup-bucket/database/
aws s3 cp "$BACKUP_DIR/files_$DATE.tar.gz" \
  s3://my-backup-bucket/files/
log "Upload completed"

# Cleanup local backups older than 7 days
find $BACKUP_DIR -name "*.dump" -mtime +7 -delete
find $BACKUP_DIR -name "*.tar.gz" -mtime +7 -delete

log "Backup process completed successfully"

Backup Verification

Verify Backups

Never assume backups work—verify them:

1. Test restores: Regularly restore from backups
2. Checksum verification: Verify file integrity
3. Automated testing: Script restore tests
4. Documentation: Document restore procedures

Verification Script

#!/bin/bash
# Verify backup integrity

BACKUP_FILE="$1"

# Check if file exists and is readable
if [ ! -r "$BACKUP_FILE" ]; then
  echo "Error: Cannot read backup file"
  exit 1
fi

# Verify PostgreSQL dump
if [[ "$BACKUP_FILE" == *.dump ]]; then
  pg_restore --list "$BACKUP_FILE" > /dev/null
  if [ $? -eq 0 ]; then
    echo "PostgreSQL backup is valid"
  else
    echo "Error: PostgreSQL backup is corrupted"
    exit 1
  fi
fi

# Verify tar archive
if [[ "$BACKUP_FILE" == *.tar.gz ]]; then
  tar -tzf "$BACKUP_FILE" > /dev/null
  if [ $? -eq 0 ]; then
    echo "Tar archive is valid"
  else
    echo "Error: Tar archive is corrupted"
    exit 1
  fi
fi

Disaster Recovery

Recovery Procedures

Document recovery steps:

1. Assess damage: What data is lost?
2. Choose backup: Select appropriate backup point
3. Prepare environment: Set up recovery environment
4. Restore data: Execute restore procedures
5. Verify restoration: Confirm data integrity
6. Resume operations: Bring systems back online
7. Post-mortem: Document what happened

Recovery Time Objectives (RTO)

Define acceptable downtime:

• Critical systems: < 1 hour
• Important systems: < 4 hours
• Non-critical: < 24 hours

Recovery Point Objectives (RPO)

Define acceptable data loss:

• Critical data: < 15 minutes
• Important data: < 1 hour
• Non-critical: < 24 hours

Monitoring & Alerts

Backup Monitoring

Monitor backup health:

• Backup success/failure: Alert on failures
• Backup size: Detect anomalies
• Backup duration: Performance monitoring
• Storage usage: Prevent disk full
• Restore tests: Regular verification

Alerting

Set up alerts for:

• Backup failures: Immediate notification
• Storage issues: Disk space warnings
• Restore test failures: Backup corruption
• Cloud upload failures: Off-site backup issues

Real-World Implementation

For a client's infrastructure, I set up:

• Daily incremental backups: Databases and files
• Weekly full backups: Complete system snapshots
• Monthly archives: Long-term retention
• S3 off-site storage: Geographic redundancy
• Automated verification: Daily restore tests
• Monitoring and alerts: Backup health tracking
• Documented procedures: Recovery runbooks

Results:

• 100% backup success rate (monitored and verified)
• < 1 hour RTO for critical systems
• < 15 minute RPO for critical data
• Zero data loss incidents
• Automated 95% of backup process

Best Practices

1. Automate everything: Manual backups are unreliable
2. Test restores regularly: Backups are useless if you can't restore
3. Encrypt sensitive data: Protect backups from unauthorized access
4. Off-site storage: Geographic separation is essential
5. Monitor backup health: Know immediately if backups fail
6. Document procedures: Recovery needs clear documentation
7. Regular reviews: Update backup strategy as needs change

Backup systems are insurance policies you hope to never use, but when disaster strikes, they're the difference between business continuity and business failure. Invest in robust, automated backup systems and verify they work regularly.