Ethereum News Today: Starknet's Grinta Upgrade Unravels in 4-Hour Sequencer Meltdown

Starknet, a layer-2 (L2) scaling solution for the Ethereum blockchain, experienced a significant network outage on September 2, caused by a failure of its sequencer, which is responsible for ordering and validating transactions before they are added to a block. The outage lasted approximately four hours, during which block production slowed and transactions were not processed. According to the status.starknet.io dashboard, the sequencer failed to recognize the “Cairo0 code” during the incident, leading to a freeze in network activity. This marks the second major outage for Starknet within two months, raising concerns about the platform's reliability as a key Ethereum L2 network with $548 million in total value locked (TVL) [1].

The disruption was attributed to the recent Grinta upgrade, also known as Starknet v0.14.0, which introduced major architectural changes, including modifications to the decentralized sequencer, fee market, and mempool. Developers had previously warned users of a brief downtime but estimated the disruption to be no more than 15 minutes. Instead, the network faced a series of cascading issues, including slow block creation and idle gateway alerts, which persisted for several hours. The team reported partial recovery after 20 minutes but did not achieve full functionality until hours later [2].

To restore service, transactions submitted between 2:23 am and 4:36 am UTC were not processed, necessitating their resubmission by users. A blockchain reorganization was committed from block 1960612, representing one hour of activity, and the team promised a detailed timeline outlining the root cause and measures to prevent future outages. Starknet confirmed that most RPC providers were back online, while others were expected to upgrade shortly. The incident highlights ongoing challenges in maintaining network stability as the platform advances its roadmap toward decentralization [1].

The Grinta upgrade is a critical step in Starknet’s long-term strategy to decentralize its core infrastructure, including block validation and production. While the platform has taken steps toward decentralization, sequencer operations remain under the control of StarkWare, the company behind Starknet. According to its published roadmap, the goal is to distribute these responsibilities across multiple independent nodes. The Grinta update, however, also revealed potential vulnerabilities in the system, as the sequencer’s failure led to widespread transaction delays [2].

In the wake of the outage, Starknet’s native token, STRK, fell more than 3% in value over a 24-hour period, trading at $0.1232. The incident occurred amid growing interest in Ethereum and its L2 ecosystems, as institutional inflows into Ethereum exceeded those of Bitcoin in August. The outage, however, may dampen investor confidence in Starknet’s reliability as a scalable solution for Ethereum’s network congestion. The team has yet to provide a comprehensive explanation of the root cause but emphasized the importance of service stability moving forward [2].

Source:

[1] Ethereum L2 Starknet suffers 2nd mainnet outage in 2 (https://cointelegraph.com/news/starknet-outage-ethereum-l2-reliability-concerns)

[2] Ethereum layer 2 StarkNet hit by 4-hour outage after upgrade (https://www.mitrade.com/insights/news/live-news/article-3-1088165-20250902)