Business Continuity

Client workloads run on managed cloud services, with AWS as a primary infrastructure provider. WebCastle inherits the resilience of the underlying platform rather than operating its own hardware.

• Compute, storage and database services are provisioned through the cloud provider managed offerings.
• Provider maintenance, hardware replacement and facility resilience are handled by the provider.
• Where a client requires or already uses a different provider, the same principles apply on that platform.

Managed databases and storage in WebCastle-operated environments use cloud provider native backup capabilities, configured per environment.

• Backup configuration is set during environment build and reviewed with the client where the contract specifies requirements.
• WebCastle does not publish a company-wide backup schedule, because configuration is set per engagement.

Backups inherit the cloud provider managed encryption applied to the environment they belong to, and the same restricted access as production.

• Backup restoration is available only to the limited technical personnel with production access for that engagement.

Because client environments are logically separate, a failure or destructive change in one engagement does not propagate to another.

• Recovery for one client can proceed without touching or coordinating with any other client environment.

WebCastle is documenting how the business continues to operate through a major disruption, including delivery, communication and decision-making. The plan is drafted but not yet approved.

• Scope covers loss of an office, loss of key personnel availability and prolonged provider disruption.

Recovery steps are understood by the engineers who build each environment. Writing them into per-environment procedures that another engineer could follow is under way.

• Procedures are held internally and are not published, since recovery detail is operationally sensitive.
• WebCastle does not claim a documented and tested disaster recovery plan today.

Customers are contacted directly during a significant disruption. A consistent status channel and a standard update format are being established.

• The security status page is the intended public channel for service-affecting events.

Restores are performed as part of normal engineering work, such as environment refreshes and migrations, which exercises the mechanism. A scheduled and recorded restoration test programme is not yet established.

• Introducing periodic tests with recorded outcomes is a priority item on the continuity roadmap.

WebCastle does not publish recovery time or recovery point objectives. Realistic objectives are being assessed per architecture pattern before any figure is committed to.

• Where a customer requires specific objectives, they are discussed at design time and set in the contract for that engagement.
• Publishing a figure that architecture does not support would be misleading, so none is published.

Redundancy varies by engagement and by what each client has commissioned. Standard resilience patterns to offer as a baseline are being evaluated.

• Managed services provide provider-level redundancy within their region by default.
• Multi-region deployment is a commercial and architectural decision made with the client, not a standard inclusion.

A structured analysis of which services and processes matter most, and how long each can tolerate being unavailable, is on the roadmap. It will inform the recovery objectives above.

Scenario exercises covering provider outage, data loss and loss of key personnel will follow approval of the continuity plan. None have been run to date.