Coinbase’s Ethereum Layer-2 network, Base, suffered two outages on its mainnet within two days, causing block production to stall. On June 27, the Base development team announced on X that initial fixes had been completed, and the planned Beryl hard fork and B20 Registry activation would be urgently postponed. A new timeline will be announced after testing and verification.
Technical Causes of the Two Outages: Sequencer Bug and Unsafe Head Stall
According to Base’s incident reports, the causes of the two outages are as follows:
First outage: A sequencer software bug prevented the system from producing new blocks, halting mainnet block production for approximately two hours.
Second outage: An "Unsafe Head Stall" occurred, where nodes could not keep synchronizing with the latest block height, again impacting block production.
Base stated that the causes of the two events were different. Initial fixes have been completed, and root cause analysis is ongoing to reduce the likelihood of similar incidents.
Original Scope of the Beryl Hard Fork and B20 Registry, and Postponement Details
Beryl is one of Base’s key upgrades this year, originally planned to include improvements in network performance, transaction processing efficiency, and developer tools, alongside the activation of the B20 Registry to establish a new standardized registration mechanism. Since the two updates are interdependent, the team decided to adjust the timeline synchronously.
Base stated that the postponement was due to the recent mainnet anomalies and the need to reschedule the B20 Registry deployment. The upgrade will be rescheduled after testing is completed.
Base’s Explanation of User Asset Security During the Outages
In the incident reports, Base stated that no user assets were affected during either outage, and no data loss or transaction tampering occurred. The main impact during the outages was on block production and transaction confirmation speed. Base noted that root cause analysis and system improvements are ongoing.
Frequently Asked Questions
What specifically are the technical differences between the two Base outages?
According to Base’s incident reports, the first outage was caused by a sequencer software bug that prevented the system from creating new blocks, halting mainnet block production for about two hours. The second outage was due to an "Unsafe Head Stall," where nodes could not keep synchronizing with the latest block height. Base explicitly stated that the causes of the two events were different. Initial fixes have been completed, and root cause analysis continues.
When will the new upgrade timeline for the postponed Beryl hard fork be announced?
According to Base’s public statements, the Beryl hard fork and B20 Registry activation will be rescheduled after testing and verification. As of June 29, 2026, Base has not provided a specific new upgrade date or time range.
What is a Sequencer, and why does its failure cause the entire Layer-2 network to halt?
A sequencer is a core component in the Optimistic Rollup architecture responsible for ordering and submitting transactions. Currently, most Optimistic Rollups use a centralized sequencer design. If the sequencer malfunctions, the entire Layer-2 block production and transaction confirmation process is affected, creating a single point of failure risk. Both of Base’s recent outages were directly related to sequencer or node synchronization issues.
