What caused Base blockchain’s 33-minute network outage?
Base, Coinbase’s Ethereum layer-2 blockchain, experienced a 33-minute network outage on Aug. 5 after a faulty sequencer transition halted block production.
- Base’s block production paused for 33 minutes on August 5 due to a misconfigured sequencer.
- The outage was resolved manually, with infrastructure upgrades now underway.
- Other recent outages include Hyperliquid and TON, showing scalability scaling risks.
A bug in Conductor, an OP ( OP ) Stack component that oversees Base’s high-availability sequencer cluster, was the root cause of the incident. According to a postmortem released by the Base team, the sequencer handoff mechanism chose a backup node that wasn’t fully provisioned, which caused a halt in transaction processing.
Sequencer handoff fails during on-chain congestion
The disruption began at 6:07 am UTC when the primary sequencer lagged due to a spike in on-chain activity. Conductor, designed to maintain uptime by rotating sequencers, elected a new node mid-provisioning.
Lacking full Conductor functionality, the replacement couldn’t initiate another handoff, leaving the chain unable to produce blocks. At 6:12 am, Base formally declared the incident and began mitigation steps.
By 6:40 am UTC, the team had manually transferred leadership to a healthy sequencer and resumed block production. No funds were lost during the event. Protocols like Aave ( AAVE ) and Moonwell avoided liquidation errors thanks to Chainlink’s ( LINK ) Sequencer Uptime Feed circuit breakers.
The team paused Conductor, manually coordinated a safe transition, and evaluated reorgarnization risks. The outage, which impacted Flashblocks, withdrawals, and deposits, lasted 33 minutes in total.
Infrastructure fixes in progress
Base plans to upgrade its infrastructure to ensure that all sequencers added to the cluster can execute handoffs in order to stop recurrence. Additional testing and monitoring enhancements will also be rolled out in the coming days.
The outage is Base’s first major downtime since September 2023, when the network was down for 43 minutes shortly after its public launch. The incident has renewed scrutiny of centralized sequencer architectures, especially as Base now secures over $4.1 billion in total value locked.
Base joins a growing list of recent L2 disruptions, including Hyperliquid’s ( HYPE ) July 29 API outage caused by a traffic spike, and TON’s 40-minute halt on June 1 during the DOGS token launch.
Disclaimer: The content of this article solely reflects the author's opinion and does not represent the platform in any capacity. This article is not intended to serve as a reference for making investment decisions.
You may also like
How much CPO demand will Nvidia Rubin Ultra generate? Bernstein: Optical value per Pod will increase fivefold
Bernstein believes that the co-packaged optics (CPO) and fiber demand driven by Nvidia's Rubin Ultra architecture will surge significantly, with the optical content per Pod rising from $1 million to $5.4 million. This will propel the size of Nvidia's optical ecosystem market to $46.8 billion by 2028 (a 91% annual increase). Although the industry faces supply bottlenecks with InP substrates, the outlook remains positive. The institution reiterates its “outperform” rating for Lumentum, Coherent, and Ciena.
UBS Q3 implements a two-way strategy with gold options, heavily invests in technology stocks, and builds positions in SK Hynix (SKHY.US)
UBS Group (UBS.US) has submitted its Q3 holdings report (13F) as of September 30, 2026.
SpaceX (SPCX.US) reportedly attempts “de-Nvidiaization” at the rack level: developing Kessel cabinets to bypass L6 compute boards, purchasing only chips
SemiAnalysis: SpaceX has developed its own Kessel cabinet, bypassing Nvidia's L6 cards to provide more options for manufacturing its own hardware.
$3.7 billion “data center power equipment prepayments” abandoned! What did "the most active buyer of gas turbines", SoftBank, see?
In the second quarter of fiscal year 2026, SoftBank reported a net loss of $7.1 billion, primarily due to a $3.7 billion impairment charge from a $9.5 billion advance payment for power generation equipment for US data centers—of which about $2.9 billion was non-refundable and effectively lost. The CFO admitted that SoftBank placed orders before the project plan was finalized, and is now conducting a comprehensive review of its procurement strategy. "The most aggressive purchaser of gas turbines has now written off the deposit in this way."
Crypto prices
More
