Incident Timeline and Root Cause Analysis
On June 27, Base engineering team released a detailed post-mortem of two block production outages that occurred on June 25 and 26. The first incident started at 11:47 AM ET on June 25 and lasted 116 minutes; the second occurred at 11:28 AM ET on June 26 and lasted 20 minutes. Both outages shared the same underlying cause.
The root cause was identified as a bug in the sequencer's block-building logic: when the block builder received an invalid transaction that failed execution as expected, the system failed to clear expired log state (which included records of accessed accounts and storage slots). Subsequently, before the log state was properly cleared, a valid transaction was executed, leading to a Gas accounting mismatch. This produced a sequenced block containing an invalid state transition, which other nodes could not accept. As a result, the chain came to a complete halt: no new L2 blocks could be produced, sequencer and validator nodes could not advance past the invalid block, and new transactions could not be included on-chain.
Base emphasized that chain integrity was unaffected and all user funds remained safe. Block production resumed safely after the team applied mitigation measures.
Fix and Future Improvements
To resolve the outage, Base deployed a patch to the sequencer that ensures log state is correctly updated during execution, preventing Gas mismatches of this kind. The team stated they will enhance protocol fuzz testing and load testing tools to better identify adversarial transaction patterns that could expose similar vulnerabilities, while also continuing to improve operational and monitoring capabilities to strengthen overall network reliability.