Sanitised case study

Resilient private networking

A multi-site private network designed to remain understandable, recoverable, and supportable when connectivity changes or the primary path is unavailable.

Problem

Long-distance private connectivity can become difficult to repair when an update, reboot, routing change, or provider issue disrupts the primary path. The design needed to support routine use without turning recovery into an on-site task.

Approach

• Separated normal routed traffic from administrative recovery access.
• Documented route ownership, restart order, and verification steps.
• Used health checks to distinguish tunnel failure from upstream connectivity problems.
• Kept configuration changes repeatable and reversible.
• Tested recovery after reboots and deliberately interrupted connectivity.

Evidence and outcomes

• Private services remain reachable without exposing application ports publicly.
• Recovery does not depend on the same path used for normal traffic.
• Post-reboot checks make routing and tunnel state visible before users notice a problem.
• The documentation describes symptoms, checks, and rollback actions instead of relying on memory.

Public-safety boundary

This case study intentionally omits real addresses, hostnames, physical locations, device models, authentication details, firewall rules, and recovery topology. The public version demonstrates the operational design without providing a map of the private environment.