Sui’s Blockchain Blues: When Fixes Go Fizz!

it caused another halt. Shocking, I know. Who would’ve thought ignoring a problem would make it come back with a vengeance?

The Sui Foundation, in a moment of sheer transparency (or was it desperation?), published a post-mortem that reads like a comedy of errors. Turns out, they knowingly deployed a fix with a “recognized probability” of causing another halt. And lo and behold, the network hit that exact failure the very next morning. Because why fix it right when you can fix it twice?

This disclosure adds a layer of transparency, sure, but it also raises questions like, “Who’s in charge here? The Keystone Cops?” And let’s not forget, this is one of the highest-profile chains built by alumni of Meta’s shelved Diem crypto project. So much for second chances, huh?

Two Bugs, Three Halts, 18+ Hours of Downtime: The Sui Trilogy No One Asked For

According to Sui’s timeline, all three outages trace back to the v1.72 software release, which introduced a feature called “address balances.” Because what blockchain doesn’t need more ways to store funds and pay for gas? It was part of a broader push to reduce friction, alongside the rollout of gasless stablecoin transfers. Because nothing says “smooth sailing” like introducing new features that break everything!

The first halt began around 7 a.m. PT on Thursday and lasted until roughly 1:30 p.m. PT. It was caused by a bug in how the network handles “gas smashing,” which, again, sounds like something you’d do at a bad party. The problem surfaced when a transaction was canceled for insufficient funds, but the gas smashing step still attempted to spend those same funds. That produced a negative balance delta that crashed the validator settlement process. Because why have a safety net when you can just fall flat on your face?

Following last week’s outages related to the 1.72 release, the Sui Core Team has completed an investigation and incident review, detailing what happened and the steps taken by validators to restart the network. Spoiler: it involved a lot of head-scratching and probably some coffee.

– Sui (@SuiNetwork) May 31, 2026

The Sui Foundation said this edge case could only be triggered when two transactions competed to spend from the same address balance simultaneously, and one had to be canceled to prevent an overdraft. The cancellation worked as intended, but the subsequent gas charge did not respect the cancellation. Because, you know, rules are just suggestions in the blockchain world.

The Interim Fix That Broke Things Again: A Classic Sui Move

Here’s where the post-mortem gets hilariously candid. The core team proposed a fix on Thursday at around noon PT: simply skip gas smashing when a transaction is canceled with an InsufficientFundsForWithdraw error. Validators adopted it, and the network came back by 1:30 p.m. PT. Problem solved, right? Wrong. They accepted the risk that this fix could cause another halt. And guess what? It did. Because why learn from history when you can repeat it?

The known shortcoming was that a transaction can be canceled for multiple reasons at once, and one error code can mask another. If the InsufficientFundsForWithdraw error was overridden by a different cancellation reason, the fix would be bypassed entirely, and the same underflow crash would happen again. And it did. On Friday morning. Because why not?

The second outage started around 5 a.m. PT and was resolved by roughly 8:30 a.m. PT after the team deployed a more robust fix it had been building in parallel. Because nothing says “we’ve got this” like needing a backup fix for your fix.

A Third Halt Nobody Saw Coming? Sure, Why Not?

The network ran normally for about five hours after the Friday morning recovery. Then, at approximately 1:30 p.m. PT, the scheduled epoch change failed to complete, freezing the chain for a third time. Because why stop at two when you can go for the hat-trick?

This halt had nothing to do with gas charging. It was triggered by a previously unknown bug in how validators preserve the state of their distributed key generation (DKG) protocol across restarts. Because what’s a blockchain without a few hidden surprises?

At the start of each epoch, Sui validators run DKG to set up the random beacon that certain applications depend on. DKG requires higher participation than regular consensus. When validators restarted to install Friday morning’s patch, not enough were ready for DKG, so it disabled itself. That was expected. What wasn’t expected was that the failure status never got written to disk. As validators continued restarting, each one came back up unaware that DKG had already failed. Transactions requiring randomness could neither execute nor be canceled, and the queue backing up those transactions prevented the epoch from closing. Because why have a plan when you can wing it?

The fix had two parts: persist the DKG failure status across restarts, and add a mechanism for validators to force-close a stuck epoch at a coordinated point. Sui confirmed it used the force-close tool once during this recovery, and the network entered its next epoch normally with randomness restored. The third outage lasted from about 1:30 p.m. PT to 7:20 p.m. PT. Because why rush when you’re already behind schedule?

Sui emphasized that across all three halts, no user funds were at risk and no committed transactions were rolled back when the network resumed. The chain simply stopped processing new transactions during each window of downtime. Because nothing says “we’re reliable” like stopping all operations!

Sui Acknowledges Systemic Weaknesses: Shocking, I Know

The post-mortem closes with several forward-looking admissions that are worth noting. The Foundation acknowledged that gas charging logic has grown complex enough that edge cases like these are hard to rule out by inspection alone, and said the system deserves the same rigor as its Move virtual machine or Mysticeti consensus engine. It also flagged end-of-epoch resilience as too narrow, noting that the existing “safe mode” fallback did not cover the DKG failure scenario. Because why prepare for everything when you can just wing it?

On the operational side, Sui disclosed that AI agents with access to production state helped speed up diagnosis across all three incidents by querying validator logs and assembling metrics on demand. Because when in doubt, let the robots handle it.

The Foundation said it plans to invest in failure containment strategies so that a future bug of this class would drop the problematic transaction rather than take down the entire network. Because why fix the whole system when you can just ignore the problem?

SUI Price Continues to Slide: Surprise, Surprise

SUI was trading around $0.87 at the time of writing, down roughly 13% from $1.04 recorded one week ago, according to CoinMarketCap data. The token’s market cap has slipped to approximately $3.48 billion. SUI hit an all-time high of $5.35 on January 6, 2025, meaning the token is now trading roughly 84% below that peak. Because nothing says “confidence” like a nosedive in value!

As The Crypto Times reported during the outages, $1.88 million in SUI positions were liquidated across the three halts, with long traders bearing the brunt of the losses. Because why have a safety net when you can just let everyone fall?

A Pattern That’s Getting Harder to Ignore: The Sui Saga Continues

This was not an isolated event. Sui experienced a six-hour consensus divergence stall in January 2026, and before that, a validator crash bug knocked the network offline in November 2024. Three major outage windows in 2026 alone puts Sui in uncomfortable company, drawing comparisons to Solana’s early-era reliability struggles. Because why be original when you can just copy the mistakes of others?

Sui, a Layer 1 blockchain built by Mysten Labs, has positioned itself as a scalable platform for DeFi, gaming, and institutional adoption. The very features meant to make the network more accessible, address balances and fee-free stablecoin transfers, created the new code paths where Thursday’s bug first appeared. Because why make things simple when you can overcomplicate them?

Whether the Foundation’s promised investments in failure containment and gas logic refactoring can break the cycle of upgrade-driven downtime will likely determine how much confidence developers and institutional users place in Sui going forward. Because why trust them when they’ve given us every reason not to?

2026-06-01 08:05