Backup & Recovery
This document describes oklido's backup strategy and disaster recovery capabilities.
Backup Strategy
Overview
| Metric | Target |
|---|
| RPO (Recovery Point Objective) | 1 hour |
| RTO (Recovery Time Objective) | 4 hours |
| Backup frequency | Continuous + Daily |
| Backup retention | 30 days |
What We Back Up
| Data | Method | Frequency |
|---|
| Database | Automated snapshots | Daily + point-in-time |
| Documents | S3 replication | Real-time |
| Configuration | Version control | On change |
| Secrets | AWS Secrets Manager | Encrypted backup |
Database Backups
Automated Snapshots
- Frequency: Daily at 03:00 UTC
- Retention: 30 days
- Type: Full snapshot
- Encryption: AES-256
Point-in-Time Recovery
- Granularity: Up to 5 minutes
- Retention: 7 days
- Use case: Precise recovery to specific time
Testing
- Monthly restore tests
- Verified data integrity
- Documented recovery times
Document Backups
S3 Storage
Documents are stored with multiple protections:
- Versioning: All versions retained
- Replication: Cross-region (within UK)
- Durability: 99.999999999% (11 nines)
Recovery
| Scenario | Method | Time |
|---|
| Accidental deletion | Restore from version | Minutes |
| Corruption | Restore from replica | Minutes |
| Region failure | Failover to backup region | Hours |
Disaster Recovery
DR Tiers
| Scenario | Impact | Recovery |
|---|
| Single service failure | Partial | Automatic failover |
| Availability zone failure | Partial | Automatic failover |
| Region failure | Full | Manual failover to DR |
DR Procedures
- Monitoring detects failure
- Automated response for tier 1-2
- Manual activation for tier 3
- Validation of recovery
- Communication to customers
DR Testing
- Quarterly failover tests
- Annual full DR exercise
- Documented results and improvements
Business Continuity
Service Dependencies
| Service | Redundancy |
|---|
| Web application | Multi-AZ |
| API servers | Multi-AZ |
| Database | Multi-AZ |
| Storage | Multi-region |
| DNS | Global anycast |
Failover Capabilities
- Automatic health checks
- Load balancer failover
- Database replica promotion
- DNS failover
Recovery Procedures
Database Recovery
- Identify recovery point
- Create new instance from backup
- Verify data integrity
- Update connection strings
- Resume service
Document Recovery
- Identify affected documents
- Restore from S3 versioning
- Verify file integrity
- Update metadata if needed
Full Service Recovery
- Deploy infrastructure from code
- Restore database from backup
- Verify document access
- Run integration tests
- Resume service
- Monitor for issues
Customer Data Recovery
What You Can Recover
| Data | Self-Service | Support |
|---|
| Deleted documents (30 days) | Yes | Yes |
| Older deleted documents | No | No |
| Previous versions | No | Contact support |
| Account data | Yes | Yes |
How to Request Recovery
- Deleted documents: Check your Deleted folder
- Older data: Contact support@oklido.com
- Enterprise: Dedicated support channel
Security
Backup Security
- Encrypted at rest (AES-256)
- Encrypted in transit (TLS 1.3)
- Access controlled (IAM)
- Audit logged
- Separate from production
Recovery Security
- Verified before restore
- Integrity checks
- Access logs maintained
- No public access
Compliance
Our backup and recovery meets:
- Cyber Essentials Plus requirements
- GDPR data protection requirements
- Business continuity best practices
Questions
For backup and recovery questions:
