# Business Continuity & Disaster Recovery Plan

**Touch2Sign Ltd**  
**Version:** 1.0  
**Effective date:** 11 July 2026  
**Owner:** Robert — Engineering Lead  
**Approved by:** Robert — CEO / Managing Director  
**Next review:** 11 July 2027

---

## 1. Purpose

Ensure Touch2Sign can maintain or rapidly restore eSignature and eWitness services following disruption to AWS infrastructure, application failure, or regional outage. Supports ISO 27001 A.5.29 and SOC 2 Availability criteria.

## 2. Scope

Covers:

- Touch2Sign SaaS platform (AWS eu-west-1, Ireland)  
- Customer documents, audit trails, and signing workflows  
- Supporting services: Cognito, RDS, S3, App Runner, CloudFront, SES  
- Internal operations required to communicate with customers during outage  

Excludes: customer-owned integrations outside Touch2Sign control (e.g. customer SMTP relays).

## 3. Recovery objectives

| Metric | Target | Notes |
|--------|--------|-------|
| **RTO** (Recovery Time Objective) | 4 hours | Restore core signing and document access |
| **RPO** (Recovery Point Objective) | 1 hour | Maximum acceptable data loss |
| **Communication SLA** | 1 hour | Customer status update after confirmed major outage |

Tier 2 functions (reporting, non-critical admin) may recover within **24 hours**.

## 4. Critical systems

| System | Function | Priority |
|--------|----------|----------|
| App Runner / application tier | Signing, witnessing, API | P1 |
| Amazon RDS (PostgreSQL) | Documents metadata, audit_log, accounts | P1 |
| Amazon S3 | Document PDFs, signed artifacts | P1 |
| Amazon Cognito | Authentication | P1 |
| CloudFront | CDN, TLS termination | P1 |
| AWS SES | Transactional email (signing links, OTP) | P2 |
| Stripe | Billing (degraded mode acceptable short-term) | P3 |

## 5. AWS region strategy

| Item | Current state |
|------|---------------|
| Primary region | **eu-west-1** (Ireland) |
| Data residency | Customer data stored in eu-west-1 unless sub-processor DPA specifies otherwise |
| Multi-region active-active | Not currently deployed — single-region with backup restore |
| Failover region | **eu-west-2** (London) or **eu-central-1** (Frankfurt) — target for future DR drill |

Cross-region failover requires CEO and Engineering Lead approval and customer communication if data location changes.

## 6. Backup strategy

### 6.1 Amazon RDS

| Control | Configuration |
|---------|---------------|
| Automated backups | Enabled; retention **35 days** minimum |
| Backup window | Off-peak UTC window |
| Point-in-time recovery | Enabled |
| Manual snapshots | Before major migrations or schema changes |
| Encryption | KMS-encrypted backups |

### 6.2 Amazon S3

| Control | Configuration |
|---------|---------------|
| Versioning | Enabled on document buckets where supported |
| Cross-region replication | Evaluate for critical buckets — target Q4 |
| Lifecycle | Per Data Retention & Disposal Policy |

### 6.3 Configuration and code

- Infrastructure and application code in GitHub — tagged releases match production  
- Secrets in Secrets Manager — backed up via AWS resilience (not exported)  
- Runbooks in `docs/compliance/` and engineering wiki  

## 7. Recovery procedures

### 7.1 Application failure (single service)

1. Confirm scope via CloudWatch alarms and error rates  
2. Roll back to last known good deployment — see Change Management Policy  
3. Scale App Runner instances if capacity-related  
4. Verify signing flow end-to-end (`docs/QA_SIGN_FLOW.md` smoke tests)  
5. Post-incident note if customer-visible > 15 minutes  

### 7.2 Database failure or corruption

1. Stop writes if corruption suspected  
2. Assess latest restorable point (PITR or snapshot)  
3. Restore to new RDS instance in eu-west-1  
4. Update application connection string via Secrets Manager  
5. Validate audit_log integrity — sample document lifecycle checks  
6. Notify affected customers if data loss exceeds RPO  

### 7.3 Regional AWS outage

1. Activate incident response — P1  
2. Assess AWS Health Dashboard and Enterprise Support case  
3. If extended outage (> 4 hours): evaluate restore to alternate EU region from snapshots  
4. Customer communication via status page and email  
5. Document decision and timeline for regulators if personal data impact  

### 7.4 Total platform loss

1. Provision infrastructure from IaC / runbooks in alternate EU region  
2. Restore RDS from latest snapshot; S3 from replication or backup  
3. Redeploy application from tagged release  
4. DNS cutover to recovery environment  
5. Full QA sign flow before declaring recovery  

## 8. Restore testing

| Test | Frequency | Owner | Evidence |
|------|-----------|-------|----------|
| RDS point-in-time restore to isolated instance | **Semi-annual** | Engineering Lead | Test report + ticket |
| S3 object restore (versioning) | Annual | Engineering | Checklist |
| Full DR tabletop | Annual | Security + Engineering | Meeting notes |
| Signing smoke test post-restore | Every restore test | QA / Engineering | QA_SIGN_FLOW.md results |

First restore test target: **Month 6** of compliance programme.

## 9. Roles during disruption

| Role | Responsibilities |
|------|------------------|
| **Incident Commander** | Declares DR; coordinates recovery; customer comms |
| **Engineering Lead** | Executes restore; validates data integrity |
| **DPO** | Assesses personal data impact; breach notification if applicable |
| **CEO / Comms** | External statements; enterprise customer calls |

## 10. Communication plan

- **Internal:** Slack incident channel; on-call phone tree  
- **Customers:** Status page update within 1 hour of confirmed major outage  
- **Enterprise customers:** Direct CSM contact for P1 outages > 2 hours  
- **Regulators:** If personal data breach — see Breach Notification Procedure  

## 11. Related documents

- Incident Response Plan  
- Change Management Policy  
- Data Retention & Disposal Policy  
- SUB_PROCESSORS.md  

## 12. Review

Reviewed **annually** and after every DR test or major AWS architecture change.

---

**Approval**

| Name | Role | Signature | Date |
|------|------|-----------|------|
| Robert | CEO / Engineering Lead | Electronic | 11 July 2026 |