Sui official releases post-mortem on mainnet outage: Three major failures caused by upgrades, no rollback of any confirmed transactions
Odaily Planet Daily News: Sui Chinese official released a post-mortem of the mainnet outage, stating that on May 28 (Thursday) and May 29 (Friday), 2026, Pacific Time (UTC-7), the Sui mainnet experienced three network failures. The first two failures originated from a crash bug caused by the interaction between the gas billing logic and the recently released version 1.72 (which introduced the Address Balances feature). The fix for Thursday's incident was a temporary workaround aimed at restoring network operation as quickly as possible while the Sui core team developed a long-term solution. The team was aware that this temporary fix had an extremely low probability of causing network failure, but accepted this risk to restore mainnet operation promptly. On Friday morning, another variant of this known issue was triggered, leading to another failure.
The third failure occurred during the routine Epoch transition on Friday afternoon. When validator nodes restarted to deploy the Friday morning fix, a long-dormant defect in saving the Randomness State was triggered, causing the network to fail again.
Failure timeline: First failure: started Thursday around 7:00 PT, recovered at 13:30 PT; Second failure: started Friday around 5:00 PT, recovered at 8:30 PT; Third failure: started Friday around 13:30 PT, recovered at 19:20 PT.
User funds remained safe throughout the entire incident. No confirmed transactions were rolled back after network recovery.
Currently, validators have fully patched the original Gas Charging and Randomness State vulnerabilities, and network activity has returned to normal.
