📢 Live Market News: Loading news...

Prysm Bug Knocks Ethereum Consensus Participation After Fusaka


Shortly after the Fusaka network upgrade, the Ethereum network saw a sharp drop in validator participation after a bug in the Prysm consensus client took a large portion of votes offline.

According to Prysm’s announcement on Thursday, v7.0.0 of the client unnecessarily created legacy instances while processing legacy certificates, a flaw that Prysm core developer Terence Tsao said prevents nodes from working properly. The developers recommended that users run the client with the “-disable-last-epoch-targets” flag as a temporary workaround.

Beaconcha.in network data shows that on epoch 411,448, the network achieved only 75% synchronization share (the percentage of 512 randomly selected nodes to sign the chain’s headers) and 74.7% vote share. Voting participation fell by 25%, less than 9% of the network losing the two-thirds majority needed to maintain finality and regular operation.

At the time of writing, the current Ethereum network era (411,712) is seeing nearly 99% vote participation and has reached 97% sync participation, indicating a recovery for the network. Before this issue arose, eras routinely saw upwards of 99% voter turnout.

The drop in voting participation roughly matches the share of validators using the Prysm consensus client, which was estimated at 22.71% on Wednesday, before falling to 18% after the incident. This suggests that attestation failures were likely concentrated among Prysm validators.

Customer diversity chart. source: megalabs

The Ethereum Foundation and Prysm development organization Offchain Labs did not respond to Cointelegraph’s request for comment by publication.

Related to: Exclusive data from EigenPhi reveals that sandwich attacks on Ethereum have diminished

Brushing with loss of finish

If voting participation drops below two-thirds of the total ether (ETH), the Ethereum network loses its finality. Under Ethereum’s design, blocks can still be produced in this scenario, but the chain is no longer considered final.

As a likely result of such an outage, Layer 2 bridges will freeze, pooling operations will pause for withdrawals, and exchanges will increase their block confirmation requirements amid increasing risks of chain reorganization.

A similar event that could lead to Ethereum losing finality is not purely theoretical. In early May 2023, the Ethereum mainnet lost its effectiveness – an incident that occurred twice within 24 hours due to errors in the handling of legacy target certificates in the Prysm and Teku consensus clients.

The incident could have had much worse consequences, as Prysm’s developers estimated it was running on more than two-thirds of agreed-upon nodes in September 2021. Data shared in January 2022 by Michael Sproul, a developer working on the current majority-compliant client, Lighthouse, showed that Prysm was running on 68.1% of nodes.

Customer diversity chart. source: Michael Sproul

Related to: Fusaka launches as Ethereum approaches an “instant sensation” user experience.

Customer diversity is still insufficient

While Ethereum’s client diversity has made some progress since 2022, it is still far from achieving a client count of less than 33%, a threshold that ensures that a single client error is not enough to halt the network’s finality. Current MigaLabs data indicates that Lighthouse alone accounts for 52.55% of agreed nodes, with Prysm taking second place at 18%.

Customer diversity chart. source: megalabs

This represents a deterioration from before the accident, when Lighthouse’s performance was below 48.5% and Prysm’s performance was about 22.71%, according to MigaLabs.

“If Lighthouse had crashed instead, the network would have lost the termination process,” Ethereum guru Anthony Sassano noted in an X post.

magazine: Ethereum L2s Will Be Interoperable ‘Within Months’: Complete Guide