Time, Time Slots, and Event Ordering in Ethereum Attestation

On April 2, a malicious participant exploited a vulnerability in mev-boost-relay to steal approximately $20 million. In the following days, developers fixed this vulnerability with five patches, but combined with existing network latency and validator strategies, it led to a brief instability in the Ethereum network on April 6. Reorganization is detrimental to network health and can reduce block production rates and settlement guarantees.

This article explores the interaction between mev-boost and consensus, revealing the subtleties of Ethereum PoS and suggesting some possible directions for improvement.

mev-boost Introduction

mev-boost is a protocol designed to mitigate the negative impact of maximum extractable value (MEV) on Ethereum. It includes three roles:

• Relays: Trusted auctioneers connecting proposers and builders
• Builders: Complex entities that construct blocks to maximize MEV
• Proposers: Ethereum PoS validator

The general process of each block is:

1. Builders receive transactions from various sources to create blocks.
2. Builders submit the block to Relay
3. Relay verifies the validity of the block and calculates the fees paid to the proposer.
4. Relay sends "blinded" header and payment value to the current slot proposer.
5. The proposer evaluates all bids and signs the blinded head of the highest payment.
6. The proposer will send the signature header back to the Relay
7. Relay publishes blocks through local beacon nodes and returns them to the proposer.

mev-boost is a critical infrastructure because it allows all proposers to fairly access MEV without having to establish trust with builders or seekers, helping Ethereum's long-term decentralization.

Ethereum fork selection rules and mev-boost

In Ethereum PoS, time is divided into 12-second slots. Each slot randomly designates a validator as the proposer. The 12 seconds are divided into three 4-second phases.

The most critical moment in the time slot is the proof deadline at t=4. If the validator does not see the block before this time, they will vote for the previous chain head. The earlier the block is released, the longer the propagation time, and the more accumulated proof there is.

From the perspective of network health, the optimal block time is t=0. However, since the value of blocks increases over time, proposers have an incentive to delay publication to accumulate more MEV.

To promote the transition from rational behavior to honest behavior, the "honest restructuring" mechanism has been introduced.

Proposal Enhancements and Honest Restructuring

Introduced two new concepts:

• Proposer Enhancement: Grant the proposer a "boost" equivalent to 40% of the complete attestation weight for fork selection, lasting for only one time slot.
• Honest Reorganization: Allows honest proposers to use enhanced forced reorganization of blocks with proof weight below 20%.

Honest reorganization will be avoided in certain special circumstances, such as during epoch boundary blocks.

Fix for unbinding attack

After the attack on April 2, the relay and core development teams released several patches:

1. Relay check known malicious proposers
2. Check whether the complete block has been transmitted to the P2P network.
3. Introduce random delays before release
4. Beacon node verifies block validity
5. Check if there are equivalent blocks on the network

These changes have increased the relay publication delay, causing blocks to potentially miss the attestation deadline. Combined with honest restructuring, this has led to a sharp increase in the number of forks.

Future Direction

The research community should assess the "acceptable" number of reorganizations and consider the general risks of equivalent attacks. Active directions being explored include:

• Implement "headlock" protection for mev-boost
• Implement a bug bounty program
• Expand simulation software
• Optimize relay publishing path
• Incorporate mev-boost into the consensus client (ePBS)
• Add more tests
• Encourage diversity of relay clients
• Adjust equivalent penalty measures

In summary, split attacks and mitigation measures have allowed us to understand the key relationship between latency, mev-boost, and consensus mechanisms. We hope the protocol can continue to strengthen.