Solana Network Stays Online Despite Infrastructure Routing Failure

The disruption originated at a Teraswitch facility in Miami, where a faulty default route propagated through a reflector in Amsterdam, impacting connectivity across twelve sites in Europe and Asia. Within ten minutes, engineers isolated the Miami node and restored traffic by 04:16:15 UTC. Although 597 of 699 staked validators maintained voting status throughout the event, the collective downtime of the remaining nodes meant that 28.83% of all staked SOL became delinquent. This figure brought the network within striking distance of the 33.34% threshold required to halt transaction finality entirely.

Jacob Creech, a technology executive at the Solana Foundation, characterized the event as a demonstration of the network's resilience, noting that no mainnet restart was required. This outcome stands in contrast to the February 2024 outage, which saw the network freeze for nearly five hours. However, data from Marinade Finance suggests the margin for error was slim, with the delinquent stake reaching roughly 86% of the level necessary to stop the chain. The incident specifically highlighted the risks of infrastructure clustering, as a single autonomous system accounted for over 25% of total staked SOL, with nearly all of that stake affected by the routing error. Moving forward, validator operators face increased pressure to diversify hosting providers and improve failover redundancy to prevent a single point of failure from threatening the network's consensus mechanism.