Ethereum News
Ethereum Fusaka upgrade validator slashing changes: myths vs facts

PeerDAS scaling and the L2 fee drop
The Fusaka upgrade arrived on the Ethereum mainnet on December 3, 2025. This deployment immediately reduced Layer 2 fees on Arbitrum and Optimism by 40% to 60%. PeerDAS (EIP-7594) manages how consensus nodes handle data. Instead of downloading entire blocks, nodes sample random fragments of data. This mechanism allows for an eight-fold increase in throughput for rollups. It reduces bandwidth and storage requirements for validators. L2 networks benefit because they post more data at lower costs. The PeerDAS design splits data into 128 columns spread across the network. Each node stores only a fraction of the total. This design separates blob throughput from validator hardware costs. This shift in data handling supports the rollup-centric roadmap. EIP-7918 sets a minimum blob base fee to prevent fees from falling to 1 wei. This ensures rollups contribute proportionally to the load they impose on nodes. As L2s post more data, more ETH burns, which increases deflationary pressure on the supply. You know that Ethereum developers prioritize network stability during major transitions.
The Prysm malfunction and consensus risks
The deployment caused an immediate crisis in validator participation. A bug in the Prysm consensus client version v7.0.0 generated outdated blockchain states when processing old attestations. This error disrupted node functioning. The malfunction in the Prysm consensus client version v7.0.0 caused consensus participation to drop to 74.7% at epoch 411,448, which fell dangerously close to the two-thirds threshold required to maintain network finality. This drop fell below the two-thirds limit needed to prevent transaction history reorganization. The network survived because the Prysm share stood at 22.71% at the time. This malfunction revealed how dangerous single-client dominance remains for Ethereum. At the time, only 75% of nodes signed current block headers. If a bug had occurred in the Lighthouse client, which holds 52.55% of the share, the network would have lost finalization. This risk of losing finalization could freeze fund withdrawals via L2 bridges and rollups. This incident mirrors data processing errors that previously impacted Prysm and Teku clients in May 2023.
Gas limits and the hardware trade-off
The upgrade increased the block gas limit to 60 million. This expansion provides more room for complex smart contracts. EIP-7825 limits any single transaction to 16.7 million gas units to prevent denial-of-service attacks. This change helps keep blocks balanced during high traffic. The upgrade also added secp256r1 support (EIP-7951) for passkey-style logins. This allows wallets to use device-level biometrics instead of seed phrases. The CLZ opcode (EIP-7939) makes certain bitwise math operations cheaper. EIP-7934 increases the 10 MB Recursive Length Prefix (RLP) block size limit to reduce the risk of DoS attacks. EIP-7823 and EIP-7883 reprice the ModExp precompile to prevent a single heavy crypto call from stalling an entire block. Verkle Trees replace older Merkle structures to allow light users and mobile apps to operate more efficiently. These changes save hundreds of gigabytes of space per node.
| Feature | Specification |
|---|---|
| Block Gas Limit | 60 million (EIP-7935) |
| Max Transaction Gas | 16.7 million (EIP-7825) |
| Max Block Size | 10 MB (EIP-7934) |
| Max Blob Target (BPO2) | 14 blobs |
| Max Blob Maximum (BPO2) | 21 blobs |
BPO forks and the path to 100,000 TPS
The network handles blob capacity through Blob-Parameter-Only (BPO) forks. These forks change blob settings without requiring a full hard fork. BPO1 activated on December 9, 2025, and moved the blob target to 10. BPO2 followed on January 7, 2026, and raised the target to 14 while the maximum reached 21. These steps expanded capacity for L2 networks. The increase in gas limits and PeerDAS provides the foundation for reaching 100,000 transactions per second. The upgrade also included deterministic proposer lookahead (EIP-7917) to aid preconfirmations. This provides validators with advance visibility into who proposes upcoming blocks.
The market responded to the upgrade with significant volatility. ETH traded near $3,200 on the launch day. However, the price fell below $1,800 by February 2026 due to recession fears and ETH sales from Vitalik Buterin. By June 2026, the price moved near $1,570. This disconnect shows that improved network quality does not automatically lift price. The next upgrade, Glamsterdam, targets the second half of 2026 and will focus on proposer-builder separation and block-level access lists. How will the network handle even larger validator client concentrations in the future?