Cloud Provider Availability Zone (AZ) Failure

Root Cause & Remediation

Cloud provider hardware failure, power event, or network partition within one physical AZ. Services not architected for multi-AZ failover are fully impacted. Stateful databases using single-AZ deployments are the primary casualty.

Remediation steps

1. 1Verify the outage scope against the cloud provider's Health Dashboard and Status API.
2. 2Trigger your multi-AZ or multi-region failover runbook for affected services.
3. 3Promote read replicas in healthy AZs to primary and update DNS/load balancer targets.
4. 4Communicate an ETA to the business; begin DORA triage immediately — cross-border financial impact is likely.
5. 5Post-incident: enforce IaC policies that prevent single-AZ deployments for CIF-adjacent workloads.

DORA Risk Matrix

Likelihood

Low

Blast radius

Potentially catastrophic for non-HA architectures. Cross-border impact is likely for EU-wide services hosted in a single region.

CIF impact

AZ failures often breach the Geographic Spread criterion (Art. 9(4)) as a single EU region commonly serves users across multiple member states simultaneously.

Analyst notes

A cloud AZ failure is a high-profile event that almost always triggers the cross-border geographic spread criterion (≥2 EU member states). Even if your direct customer impact seems small, the regulatory expectation is to assess the total geographic footprint of affected services.