The 1980 ARPANET collapse showed how a freak hardware fault could generate control messages that drove privileged network software into resource exhaustion across the network.
AWS’s December 2021 US-EAST-1 outage showed how congestion on an internal network could impair monitoring, control planes, support tooling, and customer services all at once.
Google Cloud’s June 2019 outage began when maintenance automation descheduled network-control software across multiple locations, turning independent safeguards into correlated failure.
Amazon S3’s 2017 US-EAST-1 outage began with an authorized maintenance command whose incorrect input removed far more capacity than intended and exposed hidden regional dependencies.