1. Introduction
The purpose of this Plan is to guarantee the operability of the Agora software in situations of contingency or infrastructure failure.
2. Outreach
This plan applies to the following components:
-
SoftwareÁgora POS (both cloud and hybrid versions)
-
Infrastructure: Servers, databases, communications, POS devices
-
Points of sale: Owned, franchised or mixed model shops
-
Services: Backend, Frontend, API, integrations (Sihot, Delivery, payments, etc.)
3. Critical risk analysis
| Risk | Impact on operation | Level of criticality | Mitigation strategy |
|---|---|---|---|
| Failure in the main cloud | Total standstill | High | HA, replication and automatic migration |
| Local POS failure | Partial shutdown | Medium | Offline mode enabled, remote support |
| Cyber-attack (ransomware, DDoS) | High | High | Firewall, immutable copies, segmentation |
| Internet outage in premises | Medium | High | Offline operation, subsequent synchronisation |
| Database corruption | Very high | Under | Backups + automatic validation |
| Human failure at headquarters | Medium | Medium | Standard Operating Procedures |
4. Backup strategy
-
Frequency:
-
Full daily + hourly incremental
-
-
Retention:
-
30 days for premium SLA customers
-
-
Location:
-
Local + remote ( S3)
-
-
Immutable backups:
-
Activated with minimum retention of 7 days
-
-
Verification:
-
Daily automated recovery test
-
Quarterly complete restorations
-
5. High availability infrastructure (HA)
-
CloudRedundant architecture in different availability zones
-
DatabaseCluster with automatic failover
-
Load balancers
-
Proxmox + Ceph: node, network or storage fault tolerance
-
UPS and electrical switching in physical facilities
-
Proactive monitoring 24/7 (Prometheus)
6. Recovery targets
| Type of incident | RTO (Recovery Time) | RPO (Maximum Data Loss) |
|---|---|---|
| Total cloud service failure | ≤ 1 hour | ≤ 15 minutes |
| Local impact on POS | ≤ 30 minutes | ≤ 1 hour |
| Restore from backup | ≤ 4 hours | Last copy available |
7. Recovery procedures
-
Partial failure (local)Remote intervention in <15 min. Activation of offline mode.
-
Total failure (cloud)automatic switchover to the contingency environment.
-
Restore from backup: snapshot upload + synchronisation of subsequent changes.
-
Crisis communication:
-
Email, telephone and direct channel to our technical team.
-
8. Validation and revision of the plan
-
Scheduled tests:
-
Full restoration: Quarterly
-
Infrastructure crash drill: Annual
-
-
Security audits and backupsMonthly
-
Review of the planAnnual or in case of critical changes
9. Additional information for the customer
-
Access to the fos.consulting technical portal (FAQs, updates, service status)
-
Possibility of specific training for the client's team
