ChainThink report, December 14: The Prysm team released an incident post-mortem analysis stating that on December 4, during the Ethereum mainnet Fusaka phase, nearly all Prysm beacon nodes experienced resource exhaustion while processing specific attestations, leading to delayed responses to validators and resulting in a significant number of missing blocks and attestations.

The incident affected 42 epochs from epoch 411439 to 411480, with 248 blocks missing across 1,344 slots, yielding a missing rate of approximately 18.5%; network participation dropped to as low as 75%, causing validators to lose around 382 ETH in attestation rewards. The root cause was Prysm receiving attestations from potentially unsynchronized nodes on the mainnet, which referenced block roots from the previous epoch. To validate their legitimacy, Prysm repeatedly replayed old epoch states and executed high-cost epoch transitions, triggering resource exhaustion under high concurrency.

The Prysm team provided a temporary mitigation by enabling the --disable-last-epoch-target parameter in version v7.0.0; subsequent releases v7.0.1 and v7.1.0 include a long-term fix that leverages the head state for attestation validation, thereby preventing redundant replay of historical states.