How to Automate VPS Backups to an External Server or Amazon S3 Bucket

---

Introduction

1. Back up your VPS to an external server using SSH for secure file transfers.
2. Back up your VPS to an Amazon S3 bucket using the AWS Command Line Interface (CLI).
3. Enhance your backup process with encryption, retention policies, and verification.

---

Prerequisites

• A VPS running a Linux distribution (e.g., Ubuntu, CentOS).
• An external server with SSH access or an AWS account with an S3 bucket created.
• Basic knowledge of Linux commands and shell scripting.
• Root or sufficient permissions to install software and configure cron jobs on your VPS.
• Optional: A database (e.g., MySQL) if you need to back up database content.

---

Backing Up to an External Server

Step 1: Setting Up SSH Access

1. Generate an SSH key pair on your VPS (skip if already done):
   bash

   ssh-keygen -t rsa -b 4096

   Press Enter to accept the default file location (/root/.ssh/id_rsa) and optionally set a passphrase.
2. Copy the public key to the external server:
   bash

   ssh-copy-id user@external_server

   Replace user@external_server with your username and the external server’s IP address or hostname.
3. Test the SSH connection:
   bash

   ssh user@external_server

   You should log in without a password prompt. If successful, proceed.

Step 2: Creating the Backup Script

1. Create a temporary backup directory:
   bash

   mkdir /backup

2. Create the backup script: Open a file (e.g., /usr/local/bin/backup_to_external.sh) in a text editor like nano or vim:
   bash

   nano /usr/local/bin/backup_to_external.sh

   Add the following content:
   bash

   # Backup script for VPS to external server 
   
   # Variables BACKUP_DIR="/backup" DATE=$(date +%Y%m%d) ARCHIVE_NAME="vps_backup_$DATE.tar.gz" EXTERNAL_SERVER="user@external_server:/path/to/backup/dir" 
   
   # Dump databases (if applicable) mysqldump -u db_user -p'db_password' database_name > $BACKUP_DIR/db_dump.sql 
   
   # Create archive tar -czf $BACKUP_DIR/$ARCHIVE_NAME /var/www /etc $BACKUP_DIR/db_dump.sql 
   
   # Transfer to external server scp $BACKUP_DIR/$ARCHIVE_NAME $EXTERNAL_SERVER 
   
   # Clean up rm $BACKUP_DIR/$ARCHIVE_NAME $BACKUP_DIR/db_dump.sql

   Explanation:
   • BACKUP_DIR: Temporary directory for backup files.
   • DATE: Timestamp for unique backup filenames (e.g., vps_backup_20231015.tar.gz).
   • ARCHIVE_NAME: Name of the compressed backup file.
   • EXTERNAL_SERVER: Destination in the format user@host:/path.
   • mysqldump: Exports a MySQL database (skip or modify for other databases like PostgreSQL with pg_dump).
   • tar -czf: Archives and compresses the specified directories (/var/www, /etc) and database dump.
   • scp: Securely copies the archive to the external server.
   • rm: Removes temporary files after transfer.
3. Customize the script:
   • Update EXTERNAL_SERVER with your server’s details.
   • Replace db_user, db_password, and database_name with your database credentials (skip if no database).
   • Adjust the directories to back up (e.g., /var/www for websites, /etc for configurations).
4. Make the script executable:
   bash

   chmod +x /usr/local/bin/backup_to_external.sh

5. Test the script:
   bash

   /usr/local/bin/backup_to_external.sh

   Verify that the backup file appears on the external server.

Step 3: Scheduling the Script

1. Open the crontab editor:
   bash

   crontab -e

2. Add a schedule: For daily backups at 2 AM, add:
   bash

   0 2 * * * /usr/local/bin/backup_to_external.sh

   Save and exit. Cron syntax: minute hour day_of_month month day_of_week command.

---

Backing Up to an S3 Bucket

Step 1: Setting Up AWS CLI

1. Install AWS CLI on your VPS:
   • For Ubuntu:
     bash

     sudo apt-get update
     sudo apt-get install awscli

   • For CentOS:
     bash

     sudo yum install awscli

2. Configure AWS CLI: Run:
   bash

   aws configure

   Enter:
   • AWS Access Key ID and Secret Access Key (from your AWS IAM user).
   • Default region name (e.g., us-east-1).
   • Default output format (e.g., json).
3. Verify the setup: List your S3 buckets to confirm:
   bash

   aws s3 ls

Step 2: Creating the Backup Script

1. Ensure the backup directory exists:
   bash

   mkdir /backup

2. Create the backup script: Open /usr/local/bin/backup_to_s3.sh:
   bash

   nano /usr/local/bin/backup_to_s3.sh

   Add:
   bash

   #!/bin/bash 
   
   # Backup script for VPS to S3 bucket 
   
   # Variables BACKUP_DIR="/backup" DATE=$(date +%Y%m%d) ARCHIVE_NAME="vps_backup_$DATE.tar.gz" S3_BUCKET="s3://your-bucket-name" 
   
   # Dump databases (if applicable) mysqldump -u db_user -p'db_password' database_name > $BACKUP_DIR/db_dump.sql 
   
   # Create archive tar -czf $BACKUP_DIR/$ARCHIVE_NAME /var/www /etc $BACKUP_DIR/db_dump.sql 
   
   # Upload to S3 aws s3 cp $BACKUP_DIR/$ARCHIVE_NAME $S3_BUCKET 
   
   # Clean up rm $BACKUP_DIR/$ARCHIVE_NAME $BACKUP_DIR/db_dump.sql

   Explanation:
   • S3_BUCKET: Your S3 bucket URL (e.g., s3://my-backup-bucket).
   • aws s3 cp: Uploads the archive to S3.
   • Other parts mirror the external server script.
3. Customize the script:
   • Replace your-bucket-name with your S3 bucket name.
   • Update database credentials and directories as needed.
4. Make it executable:
   bash

   chmod +x /usr/local/bin/backup_to_s3.sh

5. Test the script:
   bash

   /usr/local/bin/backup_to_s3.sh

   Check your S3 bucket to confirm the file uploaded.

Step 3: Scheduling the Script

1. Edit the crontab:
   bash

   crontab -e

2. Schedule daily backups: Add:
   bash

   0 2 * * * /usr/local/bin/backup_to_s3.sh

---

Additional Considerations

Encrypting Backups

1. Generate a GPG key:
   bash

   gpg --gen-key

   Follow prompts to create a key tied to your email.
2. Modify the script: Replace the tar line with:
   bash

   tar -cz /var/www /etc $BACKUP_DIR/db_dump.sql | gpg --encrypt --recipient [email protected] > $BACKUP_DIR/$ARCHIVE_NAME.gpg

   Update the transfer/upload command to use $ARCHIVE_NAME.gpg.

Managing Backup Retention

• External Server: Add to the script:
  bash

  ssh user@external_server "find /path/to/backup/dir -type f -mtime +7 -delete"

  This removes files older than 7 days.
• S3 Bucket: Use an S3 lifecycle rule via the AWS Console to delete objects older than 7 days, or script it with:
  bash

  aws s3 ls $S3_BUCKET | awk '{print $4}' | while read file; do [[ $(date -d "$(echo $file | cut -d_ -f3 | cut -d. -f1)" +%s) -lt $(date -d "7 days ago" +%s) ]] && aws s3 rm $S3_BUCKET/$file; done

Verifying Backups

• Download a backup file.
• Decrypt it (if encrypted) with gpg –decrypt.
• Extract it with tar -xzf and restore to a test environment.
• Ensure all data (files, databases) is intact.

---

Conclusion