# Backup and Restore

Backup and Restore is the practice of preserving recoverable copies of data and using them to return systems or information to a usable state after loss, corruption, or disruption.

Cloud concept. Updated 2026-10-29. Page: https://blowstack.com/tools/cloud-concepts-hub/backup-and-restore

> Backups preserve earlier states of data so engineers can recover from accidental deletion, application errors, ransomware, or infrastructure failures. Effective protection requires more than creating copies: it also requires suitable retention, protected access, and tested restoration procedures that meet business recovery objectives.

Practice exams covering Backup and Restore: [CLF-C02 AWS Certified Cloud Practitioner](https://blowstack.com/practice-exams/aws-certified-cloud-practitioner), [SOA-C03 AWS Certified CloudOps Engineer - Associate](https://blowstack.com/practice-exams/aws-certified-cloudops-engineer-associate), [DVA-C02 AWS Certified Developer - Associate](https://blowstack.com/practice-exams/aws-certified-developer-associate), [DOP-C02 AWS Certified DevOps Engineer - Professional](https://blowstack.com/practice-exams/aws-certified-devops-engineer-professional), [SAA-C03 AWS Certified Solutions Architect - Associate](https://blowstack.com/practice-exams/aws-certified-solutions-architect-associate).

## 6 key aspects of Backup and Restore

Reliable backup and restore combines measurable recovery objectives, automated protection, retained and isolated recovery points, and tested restoration procedures.

1. **Recovery objectives**: Define acceptable data loss through RPO and acceptable restoration time through RTO. Use these targets to choose backup frequency, recovery mechanisms, and restore procedures.
2. **Policy-based protection**: Automate backup schedules, resource selection, and retention rather than relying on manual copies. Review coverage as resources and tags change so new production data is not accidentally excluded.
3. **Retained recovery points**: Keep historical states long enough to recover from problems discovered after they occur. A recent backup may already contain corruption, making older recovery points essential.
4. **Isolation and deletion protection**: Restrict access to backups and use separate accounts, geographic copies, or immutable retention where appropriate and supported. These controls reduce the risk that one incident destroys both production data and its recovery copies.
5. **Consistency and recovery scope**: Choose protection that captures a usable state of the application, including coordination across related resources when necessary. Determine whether restoration recovers individual objects, volumes, databases, or a broader workload.
6. **Tested restoration**: Regularly restore into an isolated environment and validate data, dependencies, and application behavior. Measure elapsed recovery time rather than treating backup completion as proof of recoverability.

## Cloud services for Backup and Restore

AWS Backup provides a centralized policy and monitoring layer for supported resources, while services such as Amazon EBS, Amazon RDS, and Amazon S3 provide resource-specific storage and recovery capabilities. EBS snapshots support volume recovery, RDS backups support database recovery, and S3 can hold backup files or protected object data. Engineers should choose recovery mechanisms according to each workload's consistency requirements and recovery objectives, then coordinate retention, access controls, and isolation. Restoring the data is only part of recovery: application configuration, connectivity, and validation must also be covered by the recovery procedure.

- **[AWS Backup](https://blowstack.com/tools/cloud-services-hub/aws-backup)**: AWS Backup centrally manages backup policies, schedules, retention, and monitoring for supported resources. It is a common choice for consistent protection across multiple AWS services, with capabilities such as cross-account copies, cross-Region copies, and vault-based retention protection subject to resource support and configuration.
- **[Amazon RDS](https://blowstack.com/tools/cloud-services-hub/amazon-rds)**: Amazon RDS provides automated backups and manual DB snapshots, with point-in-time recovery available for supported configurations within the applicable retention window. Restoration creates a new database resource, so recovery procedures must also address validation and application connection changes.
- **[Amazon S3](https://blowstack.com/tools/cloud-services-hub/amazon-s3)**: Amazon S3 can store backup files and provides features such as Versioning and Object Lock to support data recovery and retention protection. AWS Backup also supports protection of S3 data, subject to its prerequisites and supported capabilities; an S3 bucket alone does not automatically back up an application.

## Backup and Restore exam questions

Real practice exam questions on Backup and Restore, with the correct answer and its explanation.

### Question 1 (CLF-C02)

**Question:**

A company backs up its application storage volumes once each night and retains the backups for 30 days. After an accidental deletion, the company discovers that restoring the latest backup would lose nearly a full day of changes. The company now requires recovery with no more than four hours of lost changes. Which change directly addresses this requirement?

**Correct answer:** Use AWS Backup to create recovery points every four hours instead of once each night.

**Explanation:** More frequent backups reduce the amount of data changed since the latest recovery point. A four-hour backup schedule addresses the requirement to lose no more than four hours of changes.

### Question 2 (SOA-C03)

**Question:**

A company uses AWS Backup to protect Amazon EC2 instances. During a disaster recovery exercise, a CloudOps engineer selects a recovery point and an existing IAM restore role. The request to start the restore fails with AccessDenied before a restore job is created.

The engineer's IAM role allows backup:StartRestoreJob and the read actions needed to select the recovery point and retrieve restore metadata. The selected restore role trusts backup.amazonaws.com and has all permissions required to restore the instance. An administrator can successfully initiate the same restore with the same restore role and metadata. No applicable policy contains an explicit deny.

Which change should the engineer request to enable the restore with the least privilege?

**Correct answer:** Grant the engineer's IAM role iam:PassRole on the selected restore role, restricted by iam:PassedToService to backup.amazonaws.com.

**Explanation:** The caller must be allowed to pass the selected restore role to AWS Backup. Permission to call StartRestoreJob does not provide iam:PassRole. Restricting the permission to the specific restore role and AWS Backup grants only the delegation capability needed.

### Question 3 (DOP-C02)

**Question:**

A company uses AWS Backup to protect the EBS volumes of a business application running on Amazon EC2. Backups run every 12 hours, and monitoring confirms that the newest completed recovery point is never more than 13 hours old. The company requires an RPO of 24 hours and an RTO of 4 hours.

A security review adds a requirement that every completed recovery point remain recoverable for at least 35 days, even if an attacker obtains administrator permissions in the production account and attempts to delete backups or shorten their retention. The encryption keys and the recovery automation are independently protected against modification or deletion by those administrators.

Recovery drills demonstrate that an automated workflow can provision replacement infrastructure, restore from warm-tier recovery points, validate the application, and redirect traffic in 2 hours. Restoring archived snapshots takes at least 24 hours. The company does not want to maintain running recovery application servers.

Which solution meets all the requirements?

**Correct answer:** Retain recovery points in warm storage for 35 days. Apply AWS Backup Vault Lock in compliance mode with a minimum retention of 35 days, and allow the changeable period to expire before relying on the protection. Use the tested recovery workflow to restore replacement resources from a suitable recovery point.

**Explanation:** After the changeable period expires, compliance-mode Vault Lock prevents administrators, including the account root user, from deleting protected recovery points or reducing their retention below the enforced requirements. The existing backup frequency satisfies the 24-hour RPO, and the tested warm-tier recovery workflow meets the 4-hour RTO without running standby application servers.

### Question 4 (DVA-C02)

**Question:**

An application stores customer documents in an Amazon S3 bucket that has had versioning enabled since it was created. All documents use the S3 Standard storage class. A developer accidentally deletes a document by calling DeleteObject without specifying a version ID. No subsequent writes or deletes occur for that key. The application now receives a 404 response when it retrieves the document without a version ID.

The developer must make the document available again through its original key while preserving every stored data version and without downloading or uploading the document's contents. Which solution meets these requirements?

**Correct answer:** Use ListObjectVersions to identify the current delete marker, and call DeleteObject with the key and that delete marker's version ID.

**Explanation:** In a versioning-enabled bucket, deleting an object without specifying a version ID creates a delete marker as the current version. Permanently deleting that specific marker makes the preceding data version current again. All stored data versions remain intact, and no document contents need to be transferred.

### Question 5 (SAA-C03)

**Question:**

A company runs an order-processing application on Amazon RDS for PostgreSQL. Automated backups are enabled with a 7-day retention period. An administrator accidentally dropped an independent lookup table 30 minutes ago. The table was updated throughout the morning, including shortly before it was dropped. Other tables have continued receiving valid transactions.

The company must recover the lookup table's contents as they existed immediately before the deletion without losing subsequent changes to other tables. It does not want to maintain additional database instances permanently.

Which recovery procedure meets these requirements MOST cost-effectively?

**Correct answer:** Perform a point-in-time restore to a temporary DB instance using a time immediately before the deletion. Export the lookup table, import it into the production database, validate the recovery, and delete the temporary instance.

**Explanation:** RDS point-in-time recovery uses automated backups and transaction logs to restore a new DB instance to a specified time within the retention window. Recovering the table through a temporary instance preserves valid changes in the production database and incurs additional instance charges only during recovery.

## Practice Backup and Restore in a full exam

Questions on Backup and Restore in each practice exam, with every answer option and an explanation for each:

- [SAA-C03](https://blowstack.com/practice-exams/aws-certified-solutions-architect-associate): 20 questions
- [SOA-C03](https://blowstack.com/practice-exams/aws-certified-cloudops-engineer-associate): 19 questions
- [DOP-C02](https://blowstack.com/practice-exams/aws-certified-devops-engineer-professional): 6 questions
- [CLF-C02](https://blowstack.com/practice-exams/aws-certified-cloud-practitioner): 3 questions
- [DVA-C02](https://blowstack.com/practice-exams/aws-certified-developer-associate): 2 questions

## What is Backup and Restore

Backup and restore separates the preservation of data from its recovery. A backup captures a recoverable state of a database, file system, storage volume, or other supported resource. A restore uses that state to recover data, often into a newly created resource rather than directly replacing the original. The goal is to recover usable information and application functionality, not simply to demonstrate that a backup job completed.

Recovery requirements should drive the design. The recovery point objective, or RPO, defines the maximum acceptable amount of data loss measured in time. The recovery time objective, or RTO, defines the maximum acceptable time to restore service after disruption. Nightly backups may leave nearly a day of changes unprotected, while more frequent backups or continuous protection can reduce that exposure. Actual recovery time also includes provisioning resources, restoring data, validating applications, and reconnecting clients.

Cloud environments commonly provide volume snapshots, database backups, object protection, and centralized backup management. AWS Backup is relevant when an organization needs policy-based protection across supported AWS resources, centralized scheduling and retention, and monitoring of backup activity. Engineers define backup plans and assign resources explicitly or through tags. Backup vaults organize recovery points and provide access controls. Cross-account and cross-Region copies can improve isolation and geographic resilience where supported by the resource type and configuration.

For example, a team might protect production storage volumes daily and retain recovery points for several weeks. If a deployment corrupts files, engineers can restore a volume from a recovery point taken before the deployment and attach it to a recovery instance for validation. A managed database with point-in-time recovery can provide finer recovery granularity by combining backups with transaction logs. These mechanisms have different capabilities, so a single schedule should not be assumed suitable for every workload.

Backups are not the same as replication or high availability. Replication helps maintain another current copy, but it can also propagate unwanted changes. High availability reduces downtime from infrastructure failures but does not necessarily recover deleted records. Likewise, durable cloud storage does not remove the customer's responsibility to configure appropriate data protection. Backups preserve selected earlier states, provided that retention policies keep those states long enough for problems to be discovered.

Protection has costs and operational trade-offs. More frequent backups, longer retention, isolated copies, and cross-Region transfers can increase charges. Incremental backups reduce repeated storage of unchanged data, but recovery performance still depends on the service and workload. Archive tiers can lower storage costs while introducing retrieval delays or minimum storage durations. Encryption, restricted deletion permissions, and immutable retention controls help protect backups, but encryption keys must remain available for restoration.

Restore testing closes the gap between planned protection and actual recoverability. Teams should verify data integrity, application consistency, permissions, networking, and dependencies in an isolated environment. A successful data restore does not automatically recreate infrastructure, application configuration, or client connections. Documented recovery procedures and measured test results establish whether the complete workload can meet its RPO and RTO.

## Learn Backup and Restore in lessons

11 short lessons on Backup and Restore in 4 certification courses.

- CLF-C02: Backup and Restore: recovering earlier data states (Premium), https://app.blowstack.com/lessons/backup-and-restore-recovering-earlier-data-states-clf-c02
- SOA-C03: Backup and Restore: restoring workloads from recovery points (Premium), https://app.blowstack.com/lessons/backup-and-restore-restoring-workloads-from-recovery-points-soa-c03
- DOP-C02: Backup and Restore: automated restore workflows (Premium), https://app.blowstack.com/lessons/backup-and-restore-automated-restore-workflows-dop-c02
- SAA-C03: Backup and Restore: recovery-point policies (Premium), https://app.blowstack.com/lessons/backup-and-restore-recovery-point-policies-saa-c03
- CLF-C02: Backup and Restore vs. Disaster Recovery (Premium), https://app.blowstack.com/lessons/backup-and-restore-vs-disaster-recovery-clf-c02
- SOA-C03: Backup and Restore vs Disaster Recovery (Premium), https://app.blowstack.com/lessons/backup-and-restore-vs-disaster-recovery-soa-c03
