Failover between cloud providers setup

Setting Up Failover Between Cloud Providers

Failover between cloud providers — extreme protection against vendor outage. Historically, major cloud providers have failed: AWS us-east-1 (multiple times), GCP (2019, 2020), Cloudflare (2019). If SLA requires 99.99%+, one provider is insufficient.

Prerequisites for cross-cloud failover

Without these conditions, cross-cloud failover makes no sense:

1. Cloud-agnostic architecture — application doesn't use provider-specific APIs
2. Containerization — Kubernetes provides uniform execution environment
3. Data synchronization — mechanism for data replication between providers
4. Infrastructure as Code — infrastructure of both providers described in Terraform

DNS as switching point

Cloudflare — optimal choice for managing failover between providers: works with both, belongs to neither.

import CloudFlare

cf = CloudFlare.CloudFlare(token=CF_TOKEN)

def switch_to_provider(zone_id: str, record_name: str, new_ip: str):
    records = cf.zones.dns_records.get(zone_id, params={'name': record_name})
    record_id = records[0]['id']

    cf.zones.dns_records.put(
        zone_id,
        record_id,
        data={
            'type': 'A',
            'name': record_name,
            'content': new_ip,
            'ttl': 60,
            'proxied': True
        }
    )

Cloudflare Load Balancing with health checks automates switching without manual intervention.

Timeline

Preliminary audit — 2-3 days Terraform for second provider — 5-10 days Data replication setup — 5-10 days Failover automation — 3-5 days Testing — 3-5 days