Base has released a post-mortem report detailing two block production outages on its mainnet. The first outage occurred on June 25 and lasted 116 minutes, followed by a second shorter outage of 20 minutes on June 26. Despite the interruptions, Base confirmed that all user funds remained safe and no assets were compromised. The incidents have drawn attention to the reliability of L2 sequencers, particularly those based on the OP Stack (Optimism's technology stack) used by Base.
Incident Details: Two Outages on Base Mainnet
The first outage on June 25 halted block production for nearly two hours. The second outage on June 26 was shorter but still disrupted transaction confirmations. Both events prevented the Base chain from producing new blocks, causing a temporary pause in L2 operations. Base emphasized that funds were never at risk, and no cross-chain or on-chain transfers were affected during the downtime. The root cause was identified as a sequencer bug in the block-building logic.
Root Cause Analysis: Sequencer Block-Building Logic Vulnerability
The sequencer is a critical component of any L2 chain, responsible for ordering, bundling transactions, and submitting them to the Ethereum mainnet. The report reveals that the bug originated from an incomplete cleanup mechanism. When a transaction fails execution, the sequencer's journal state is supposed to be reset to the previous valid state. However, in this case, the old journal state was not properly cleared after a failed transaction. When subsequent legitimate transactions were executed, the residual states caused incorrect gas calculations, leading to the generation of invalid state transition blocks. These invalid blocks were rejected by the Base network, causing the sequencer to stop producing blocks entirely. Base noted that this was a coding-level issue, not a fundamental protocol design flaw.
Fix and Future Prevention Plans
Base has already deployed an emergency patch to fix the sequencer bug. Looking ahead, the team plans to strengthen the protocol's fuzz testing and stress testing frameworks to proactively identify malicious transaction paths that could trigger similar vulnerabilities. Monitoring and operational workflows are also being optimized to reduce detection and response times. Additionally, Base is researching an automated recovery mechanism: once an abnormal block construction state is detected, the system can quickly switch to a safe mode and resume normal production with minimal manual intervention. This initiative aims to enhance the overall resilience of the Base network and reduce the duration of any future outages.