Key facts
- A sequencer bug in block-building logic caused two outages on the Base layer-2 network last week.
- The outages halted block production for a total of 136 minutes.
- A "race condition" after a system reset contributed to the second outage.
- The Base engineering team has applied a patch to fix the issue.
- Future improvements will focus on fuzz testing and graceful recovery mechanisms.
The Base layer-2 network experienced two block production outages last week, totaling 136 minutes, due to a sequencer bug. According to a post-mortem released by the Base engineering team, a flaw in the sequencer's block-building logic allowed stale journal state to persist after a transaction validation failure. This issue, combined with a "race condition" that occurred after a system reset, prevented the sequencers from catching up, leading to the second outage. The team has since applied a patch to ensure the journal state is properly updated. Base, which runs a single sequencer, has experienced similar outages previously. The engineering team plans to enhance protocol "fuzz testing" and implement "graceful recovery" features to prevent future incidents.