# Microsoft outage shows why redundancy can fail, Part II

URL: https://decisioninsights.ai/microsoft-outage-shows-why-redundancy-can-fail-part-ii/

This second installment follows an earlier examination of the November 13 FAA telecommunications outage and a brief look at the November 12 Microsoft global cloud disruption. The two events involved different systems and have not been linked to a shared cause, yet both raise the same question about why existing infrastructure and recovery arrangements did not keep essential services running.Redundancy alone does not guarantee continuity. Network resilience depends on whether backup facilities are truly independent, whether failures can be detected and isolated, whether restoration happens automatically, and whether recovery has been tested under adverse conditions. Because centralized cloud services support organizations across geographic regions, a single disruption can affect many organizations at once.Apparent redundancy often hides common-mode failures. Two fiber routes may share a conduit, bridge crossing, central office or power source, and two data centers may rely on the same control plane, identity service, software release or configuration system. True route diversity calls for physically separate paths and diverse points of presence, while cloud resilience also requires attention to critical dependencies and management systems rather than server distribution alone.A July 2024 IEEE Techblog article documented a prolonged Comcast outage in Santa Clara, California, linked to a physical fiber cut and reported vandalism, during which service was unavailable for about 17 hours. Established approaches such as 1+1 protection, self-healing rings and mesh-network restoration can move traffic to alternate paths, but they work only when those paths are diverse, have sufficient capacity and switch as designed.Operators should eliminate common-mode failures, provide independent alternate paths, test automatic failover and recovery under load, limit failure domains, and rehearse disaster-recovery procedures. For critical infrastructure, operators should also publish post-incident reports covering scope, the role of backup systems and corrective actions, and regulators should judge resilience by demonstrated performance. The FAA outage is significant because communications failures can affect the safe operation of the national airspace system.