백업, 복원 및 재해 복구

In the world of applications, your data is one of your most valuable assets. Losing it due to an accident, corruption, or malicious activity can be catastrophic.

This lesson will guide you through Supabase's backup mechanisms, how to perform manual backups, and how to create a solid Disaster Recovery Plan (DRP).

Supabase provides robust automatic backup solutions to protect your database. These backups are taken daily and stored securely.

• Daily Snapshots: Your entire database is backed up automatically every 24 hours.
• Retention: These backups are typically retained for 7 days on the free plan and up to 30 days on paid plans.
• Managed Service: Supabase handles the complexity of these backups for you, ensuring data durability.

When discussing database backups, it's helpful to understand the difference between logical and physical backups:

• Logical Backups: These export your database schema and data as a series of SQL commands. They are flexible and portable, great for migrating data. pg_dump creates logical backups.
• Physical Backups: These are direct copies of the database files on disk. They are faster for large databases but less portable. Supabase's automatic backups are typically physical snapshots.

While Supabase handles automatic backups, performing manual dumps gives you direct control over specific data points or for local development. You can use the standard PostgreSQL tool pg_dump via your terminal.

Remember to replace <YOUR_DB_HOST> with your actual database host (e.g., db.<PROJECT_REF>.supabase.co) and set the PGPASSWORD environment variable before running.

To restore your database from a .sql backup file, you can use the psql command-line client. This is useful for migrating data or recovering to a specific state.

Important: Restoring usually involves dropping existing tables, so be extremely careful and ensure you have a fresh backup of your current state before proceeding!

For business-critical applications, Supabase offers Point-in-Time Recovery (PITR) on higher-tier plans. This advanced feature allows you to restore your database to *any* specific second within a given retention period.

PITR is crucial for minimizing data loss in scenarios where you need to undo a specific accidental change or recover from a precise moment before an incident occurred.

Beyond your database, Supabase Storage holds your application's files (images, videos, documents). Unlike the database, Storage buckets are not automatically backed up by Supabase in the same way.

You should implement your own strategy for backing up critical files from your Storage buckets. This might involve:

• Regularly downloading files using the Supabase CLI.
• Using a custom script that interacts with the Storage API.
• Integrating with cloud storage backup services.

A Disaster Recovery Plan (DRP) is a documented process for an organization to recover from an unplanned event that disrupts normal operations. It's more comprehensive than just backups.

A DRP outlines steps to ensure your application can quickly resume critical functions after a major incident, such as a regional outage, data center failure, or severe data corruption.

A robust Disaster Recovery Plan (DRP) should include:

• Recovery Time Objective (RTO): The maximum acceptable downtime.
• Recovery Point Objective (RPO): The maximum acceptable amount of data loss.
• Roles & Responsibilities: Who does what during a disaster.
• Communication Plan: How to inform users and stakeholders.
• Testing Schedule: Regularly testing the DRP to ensure it works.
• Runbooks: Step-by-step guides for recovery procedures.

Consider the core components of a Disaster Recovery Plan (DRP).

Congratulations! You've learned about the critical aspects of data safety with Supabase.

• Supabase provides automatic daily database backups.
• You can perform manual logical backups and restores using pg_dump and psql.
• Point-in-Time Recovery offers advanced restoration capabilities.
• Remember to plan for backing up your Supabase Storage files.
• A comprehensive Disaster Recovery Plan (DRP) is essential for business continuity, defining RTO, RPO, roles, and testing.

Always prioritize your data's safety!