Sui publishes post-mortem after mainnet stall halted transactions for six hours

ambcryptoОпубликовано 2026-01-16Обновлено 2026-01-16

Введение

Sui Foundation published a post-mortem detailing a six-hour mainnet stall on January 15, caused by an edge-case bug in consensus commit logic. This led to validators processing conflicting transactions differently under specific garbage-collection conditions, resulting in divergent checkpoint digests. The network automatically halted to prevent inconsistent state finalization, prioritizing safety over availability. No forks, rollbacks, or fund losses occurred, and read operations remained functional. Sui is implementing improvements for faster recovery and enhanced testing to prevent similar issues, confirming the safety mechanisms worked as intended.

Sui Foundation has published a detailed post-mortem explaining the cause of the mainnet stall that disrupted transaction processing on 15 January. It confirmed that the network halted as a safety measure to prevent inconsistent state finalization.

According to the Foundation, the disruption lasted for approximately six hours. An internal divergence in validator consensus processing caused it.

During the incident, validators were unable to certify new checkpoints, leading to transaction submissions timing out while the network prioritized safety.

Sui consensus divergence triggered safety halt

The Foundation said the incident stemmed from an edge-case bug in the consensus commit logic that affected how conflicting transactions were handled under certain garbage-collection conditions.

As a result, different validators derived different consensus outputs and attempted to execute incompatible candidate checkpoints.

When validators detected that more than one-third of stake was signing a different checkpoint digest, checkpoint certification became impossible. Validators then halted progress to avoid finalizing an inconsistent state.

“This is the intended failure mode for this class of issue,” the Foundation said, noting that the network is designed to stop safely rather than risk forks or irreversible inconsistencies.

No forks, rollbacks, or fund losses

Sui stressed that the stall was not caused by network congestion, transaction volume, or external threats. Throughout the incident:

No certified state forks occurred
No certified transactions were rolled back
User funds were never at risk
Network safety and consistency guarantees were preserved

While transaction execution halted during the incident window, read operations continued to serve the last certified state. This ensured data consistency for users and applications.

Improvements planned after incident

The Sui Foundation said it is implementing several changes to reduce recovery time in the event of similar issues in the future.

Planned improvements include faster detection of checkpoint inconsistencies and more automated operator tooling to clean up divergent internal state. Also, expanded consensus-specific testing to reproduce and validate fixes before deployment.

The Foundation added that while the interruption was disruptive, it confirmed that Sui’s safety-focused architecture behaved as designed.

Final Thoughts

Sui’s explanation confirms the mainnet stall was the result of a consensus edge case, with safety mechanisms halting the network to avoid inconsistent finalized state.
While disruptive, the incident highlights the trade-off between availability and safety as high-throughput networks push performance limits.

Связанные с этим вопросы

QWhat was the primary cause of the mainnet stall on the Sui network on January 15th?

AThe stall was caused by an edge-case bug in the consensus commit logic that affected how conflicting transactions were handled under certain garbage-collection conditions, leading to an internal divergence in validator consensus processing.

QHow long did the network disruption last, and what was the core function that was halted?

AThe disruption lasted for approximately six hours, during which validators were unable to certify new checkpoints, halting transaction processing.

QAccording to the post-mortem, what key negative outcomes were avoided during the incident?

ANo certified state forks occurred, no certified transactions were rolled back, user funds were never at risk, and the network's safety and consistency guarantees were preserved.

QWhat safety mechanism was triggered that caused the network to halt progress?

AWhen validators detected that more than one-third of the stake was signing a different checkpoint digest, checkpoint certification became impossible, and they halted progress to avoid finalizing an inconsistent state.

QWhat improvements is the Sui Foundation implementing to prevent similar issues in the future?

APlanned improvements include faster detection of checkpoint inconsistencies, more automated operator tooling to clean up divergent internal state, and expanded consensus-specific testing to reproduce and validate fixes before deployment.

Похожее

The Higher Bitcoin Goes, The Less Institutions Want It? Coinbase Executive Weighs In

Coinbase Head of Institutional Strategy John D'Agostino states that institutions which previously bought Bitcoin at higher prices are showing increased interest as its price drops to around $60,000. He notes no signs of panic selling or major liquidation risks among large investors; instead, many, including family offices and sovereign wealth funds, see the correction as a buying opportunity. D'Agostino emphasizes that institutional interest often grows when prices fall, reflecting a long-term holding strategy. Despite recent market volatility and geopolitical tensions, exposure via spot Bitcoin ETFs remains high, exceeding $750 billion. He attributes price swings to normal commodity-like behavior and believes current market infrastructure is stronger than in past downturns. Bitcoin was trading near $63,800 at the time of reporting.

bitcoinist22 мин. назад

The Higher Bitcoin Goes, The Less Institutions Want It? Coinbase Executive Weighs In

bitcoinist22 мин. назад

Bitmine Continues Aggressive Ethereum Accumulation as Others Scale Back

Bitmine, an Ethereum treasury, accumulated 126,971 ETH last week, reinforcing its aggressive accumulation strategy amid a broader market downturn. Its holdings now total $9.6 billion in assets, including 5.54 million ETH—representing 4.59% of the total supply—203 Bitcoin, stakes in Beast Industries and Eightco Holdings, and cash. The company has staked approximately 4.72 million ETH, generating significant yields and potential annual income of around $230 million. Chairman Thomas 'Tom' Lee views the market decline as the beginning of a "crypto spring," driven by anticipated rising demand for decentralized networks like Ethereum from advancing AI systems. Unlike many other treasury firms that have halted or reversed accumulation, Bitmine continues to acquire. It recently announced plans to raise capital via a public stock offering, with proceeds potentially used for further ETH purchases, ecosystem investments, and infrastructure expansion. Bitmine now holds the world's second-largest corporate crypto treasury and the largest dedicated Ethereum treasury.

TheNewsCrypto1 ч. назад

Bitmine Continues Aggressive Ethereum Accumulation as Others Scale Back

TheNewsCrypto1 ч. назад

Apple's Desired On-Device AI Sees a Dark Horse Emerge: The First Cognitive Model is Born, 4B Matches GPT-5.4

A Chinese company, Tomorrow's Journey (Nextie), has introduced what it is calling the industry's first "cognitive model" for edge devices. Named New Journey Alpha, this 4-billion-parameter model reportedly matches the performance of trillion-parameter giants like GPT-5.4 in group intelligence tasks such as debate and collective decision-making. The development follows Andrej Karpathy's vision of stripping vast factual knowledge from large language models to retain only a smaller "cognitive core" capable of reasoning, planning, and knowing its own limits. This approach directly addresses the soaring computational costs and token expenses hindering AI's widespread deployment, as highlighted by incidents like Amazon shutting down an internal AI tool due to prohibitive costs. Trained via reinforcement learning on a corpus of academic papers from 1800-2020 to enhance generalization, the model enables three key advancements: 1) Improved decision quality in multi-agent systems, 2) Drastically reduced compute costs, allowing for cost-effective cloud or on-device (e.g., MacBook) deployment, and 3) The feasibility of "proactive" AI agents that act autonomously without user prompts, unlocking new commercial possibilities beyond today's reactive models. Built by the former Microsoft Xiaoice team—known for creating a 3.6B model that outperformed a 65B Llama model—the company is now focusing on the multi-agent systems sector, a field gaining significant investor interest. The model's economic impact is profound; by achieving high-level performance with minimal parameters, it fundamentally alters the cost structure of AI services, challenging the prevailing model of ever-larger parameter counts.

marsbit2 ч. назад

Apple's Desired On-Device AI Sees a Dark Horse Emerge: The First Cognitive Model is Born, 4B Matches GPT-5.4

marsbit2 ч. назад

Hoskinson Claims Cardano Can Surpass Bitcoin By Solving Crypto’s Trust Problem

In a June 8 livestream, Cardano founder Charles Hoskinson presented an ambitious long-term vision for the network, framing its mission as solving the global "trust problem" rather than merely competing for crypto market share. He argued that the cryptocurrency industry's core purpose is to reduce the massive annual costs associated with trusted third parties in finance and commerce through "verifiable reflexivity"—where transactions carry their own proof of correctness. Hoskinson outlined four key requirements for this goal: a decentralized engine (via Ouroboros), the right accounting model (extended UTXO), modular expansion (exemplified by partner chains like Midnight), and robust decentralized governance. He emphasized that Cardano's design aims to avoid dependency on intermediaries for reconciliation and scale for specialized applications through technologies like Hydra. While governance structures are still developing, Hoskinson claimed that if Cardano succeeds in building this trust infrastructure, ADA could become the "currency of global trust" and has the potential to eventually surpass Bitcoin. He stressed that for Cardano to reach its next level, it must prove it is a self-healing system capable of surviving even a loss of confidence in its founder.

bitcoinist2 ч. назад

Hoskinson Claims Cardano Can Surpass Bitcoin By Solving Crypto’s Trust Problem

bitcoinist2 ч. назад

OpenAI's 'Blueprint for the Future': Making AI Beneficial for Every Person on the Planet

A new transformative technology emerges every few generations. OpenAI draws a parallel with the advent of electricity in the 1920s, which initially brought convenience but ultimately enabled unprecedented progress in medicine, engineering, and living standards by empowering people to create new possibilities. AI is poised to recreate this phenomenon. Its true significance lies not in the technology itself, but in what people can achieve with it—from understanding a medical bill or starting a business to aiding scientific discovery. OpenAI believes AI should be universally accessible, allowing everyone to use it according to their own needs. This future, however, is not guaranteed. While transformative tech can centralize power, OpenAI's philosophy is that AI must serve humanity, augmenting human capabilities and broadly distributing its benefits. The company's first commitment is to build AI for human service, aiming to empower the many rather than concentrate power in a few. Safety, alignment with human intent, and oversight are paramount. OpenAI is optimistic about AI's potential to expand human welfare but remains clear-eyed about risks. The goal is to help people achieve more, not to replace them. Full automation is not the desired future; human judgment, values, and direction will become even more critical. OpenAI outlines three core goals: 1. Build automated AI researchers to accelerate and increasingly automate the research process itself, maintaining close human collaboration. The internal projection is that by March 2028, a significant portion of their research will be conducted by AI systems working alongside human researchers. 2. Accelerate economic development by advancing science, boosting productivity, and fostering growth, while ensuring the fruits are widely shared. 3. Provide a personal AGI for everyone on Earth, allowing individuals to benefit from this transformative technology in their own way. The company is entering its third phase, moving from foundational AGI research (Phase 1) to product deployment and learning from real-world use (Phase 2). The current challenge is making advanced AI abundant, affordable, safe, practical, and usable for all individuals and organizations. OpenAI concludes that a widely distributed power structure leads to a more resilient, adaptable, and free society. A positive AI future should not be controlled by a handful of entities but built, benefited from, and owned by many. If realized correctly, AI can become a cornerstone for enhancing global productivity, creativity, scientific advancement, and economic opportunity, fulfilling the mission to ensure AGI benefits all of humanity.

marsbit3 ч. назад

OpenAI's 'Blueprint for the Future': Making AI Beneficial for Every Person on the Planet

marsbit3 ч. назад

Торговля

Спот

Фьючерсы

Обсуждения

Добро пожаловать в Сообщество HTX. Здесь вы сможете быть в курсе последних новостей о развитии платформы и получить доступ к профессиональной аналитической информации о рынке. Мнения пользователей о цене на SUI (SUI) представлены ниже.