Okta disaster recovery

Okta is designed with a highly resilient architecture that ensures service continuity during regional infrastructure outages. Okta provides all customers with Standard Disaster Recovery across two regions. Each region contains an active-active-active deployment across three availability zones.

When a primary region and availability zones within that region fail due to an infrastructure outage, Okta initiates a failover to a secondary disaster recovery region. This process usually takes one hour to complete after Okta identifies the outage. When an org is failed over, admins have read-only access to the Admin Console. Users can continue to access their apps, but they can't reset their passwords. After the org's primary region is available again, Okta initiates a failback to that region. See Understanding Okta's "Read-only Mode".

Disaster recovery regions

These are the primary and secondary disaster recovery regions for each Production cell:

Cell	Primary region	Disaster recovery region
EU1	Frankfurt	Ireland
OK1	North Virginia	Oregon
OK2	North Virginia	Ohio
OK3	North Virginia	Ohio
OK4	North Virginia	Oregon
OK6	Ohio	North Virginia
OK7	Oregon	Ohio
OK8	Sydney	Melbourne
OK9	Ireland	Frankfurt
OK11	Ohio	Oregon
OK12	Oregon	North Virginia
OK14	Oregon	North Virginia
OK16	Tokyo	Osaka
OK18	Montreal	Calgary
OK19	Mumbai	Hyderabad

Guidelines and limitations

• Okta disaster recovery remediates AWS infrastructure, storage, or networking issues that impact core Okta products. When these issues occur, your org may experience elevated authentication failure rates, degraded latency, or HTTP error codes.

• Okta disaster recovery doesn't protect orgs against the following scenarios:

  • Denial of service or distributed denial of service attacks (sometimes called "request floods")

  • Issues that are related to third-party vendors and app connection

  • Bad actors who delete or modify data

  • Configuration errors

• If your server or network policies restrict traffic to certain IPs, Okta recommends that you allow access to the Okta cell IPs. See Allow access to Okta IP addresses.

Enhanced Disaster Recovery

For customers with more rigorous uptime requirements, Enhanced Disaster Recovery significantly reduces the Recovery Time Objective from one hour to five minutes. In the event of a regional failure, Okta initiates a failover for the affected Production orgs within five minutes.

Enhanced Disaster Recovery doesn't support the following Okta products:

• Okta Privileged Access (OPA)

• Advanced Server Access

• Identity Security Posture Management (ISPM)

• Okta Workflows

• Okta Identity Governance (OIG)

• DynamicScale

• Identity Threat Protection (ITP)

• Auth0 Customer Identity Cloud

• Preview orgs

Self-service Enhanced Disaster Recovery

Early Access release. See Enable self-service features.

Okta provides a self-service app and APIs that admins can use to initiate failover and failback on their Production orgs. When this feature is enabled, the Okta Disaster Recovery Admin app appears on the Okta End-User Dashboard for authorized admins. Okta recommends that admins bookmark the Okta Disaster Recovery Admin app in case their Production org isn't available during an outage.

Admins can access the app using any of these authenticators:

• Okta Verify

• Okta FastPass

• Google Authenticator

• FIDO2 authenticators

• Email

• SMS

• Call factors

• Security question

• YubiKey

Related topics

Initiate failover and failback for your org

Understanding Okta's "Read-only Mode"