Roughly 28.83% of staked SOL dropped offline on Solana following a routing failure, bringing the network uncomfortably close to the 33.34% mark where transaction finality grinds to a halt. That’s a margin of about 4.5 percentage points between normal operations and a network that can no longer confirm transactions are permanent.
How close was too close
Solana’s consensus mechanism, Tower Byzantine Fault Tolerance (BFT), requires roughly two-thirds of all staked SOL, about 66.67%, to actively participate in order to finalize transactions. Flip that around, and it means if more than 33.34% of stake goes dark, the network loses the supermajority it needs. Blocks might still be produced, but nothing gets stamped as irreversible.
At 28.83% offline, Solana was roughly 4.5 percentage points from that cliff. In practical terms, just a few additional large validators going delinquent could have tipped the balance.
Validators that go offline on Solana don’t face slashing penalties, the punitive mechanism some other proof-of-stake chains use to discourage downtime. Instead, they simply stop earning rewards.
Solana’s uptime streak and its limits
Before this incident, Solana had been on an impressive run. The network’s last recorded full outage dates back to February 2024, and the official status page had shown all systems operational for over 30 months straight.
A network can keep producing blocks while still being unable to finalize them if enough stake goes delinquent. The 30-month streak refers to full network halts, where block production itself stops. The routing failure exposed a scenario where the chain could remain technically “up” while losing its ability to confirm that transactions are permanent.
Reports from 2026 have shown up to 32 validator delinquencies within a 30-day window on Solana. Most of these stem from mundane causes: hardware failures, misconfigured software, or connectivity problems. What made this incident different was the scale. Having nearly 29% of stake affected simultaneously points to a systemic issue rather than scattered individual failures.
The Alpenglow factor
Solana has been working on a major protocol upgrade called Alpenglow, which aims to compress transaction finality down to approximately 100-150 milliseconds.
One notable design philosophy behind Alpenglow is that it prioritizes safety over liveness. The upgrade is built so the network would rather pause block production entirely than risk confirming transactions that might later prove inconsistent.
Alpenglow also introduces a fault-tolerance model that distinguishes between validators that are actively malicious and those that are simply offline due to passive failures like the routing issue that caused this incident.
What this means for Solana’s competitive position
The lack of slashing penalties is likely to draw renewed debate. Proponents argue it keeps the validator set accessible and avoids punishing operators for honest mistakes. Critics counter that without meaningful financial consequences for downtime, there’s insufficient incentive for validators to invest in the kind of redundancy that prevents large-scale simultaneous failures.
For SOL holders who stake their tokens, validators that were offline missed out on staking rewards during the downtime, which flows through to their delegators as reduced returns.
Solana’s consensus model is designed to tolerate up to a third of stake going offline, and this incident tested that boundary more aggressively than anything since the February 2024 outage.