0x7a1f949e7a1f…7a1f94a1
Ethereum Delays Glamsterdam Testing Over Risk to 15,000 Contracts
Ethereum Foundation developer Marius van der Wijden says gas-rule changes in Glamsterdam could disrupt up to 15,000 contracts, delaying Hoodi testnet activation indefinitely.

Outputs
Up to 15,000 of Ethereum's ~90 million contracts could be disrupted by Glamsterdam's proposed gas changes, per developer Marius van der Wijden.
EIP-8037 would raise contract deployment cost from 32,000 to 183,600 gas (~5.7x) and new storage slots from 20,000 to 97,920 gas (~4.9x).
Hoodi testnet activation, canceled from October 26, remains on hold pending bug fixes; developers revisit the proposals Thursday.
ChainSecurity and Dedaub audits produced roughly 200 combined findings, including 70 medium/high-risk issues and 22 the Foundation missed.
Foundation outreach identified 3 high-severity and at least 25 medium-severity flaws, with most already being patched.
Ethereum has delayed public testing of its Glamsterdam upgrade after a Foundation developer estimated that proposed gas-accounting changes could disrupt up to 15,000 smart contracts. Marius van der Wijden gave colleagues the figure on a Monday developer call, according to CoinDesk's November 17 account, putting the potentially affected contracts within a deployed population of roughly 90 million and the value at risk at a few hundred thousand dollars.
The warning forced a pause in the Hoodi testnet rollout, which the Ethereum Foundation says will stay on hold until the bugs are resolved. Core developers ended Monday's discussion without agreement and plan to revisit the proposals on Thursday.
Which proposals are causing the problem?
The dispute centers on EIP-8037 and EIP-8038, two proposals that change how Ethereum charges for storing and retrieving data. EIP-8037 would introduce a separate charge for creating permanent state — deploying a contract or adding a storage entry — on top of ordinary execution gas.
The numbers are significant. Contract deployment would cost 183,600 gas under the proposal, up from 32,000 today, roughly a 5.7x increase. Creating a new storage slot would run 97,920 gas, up from 20,000, about 4.9x the current charge.
The reach goes beyond new deployments. Existing applications can call other contracts with a predetermined gas allowance, which makes them sensitive to any change in the charges imposed during those calls. An operation that fit comfortably within an app's execution budget under current pricing could require more gas than the software allows once the new accounting takes effect.
What did the audits find?
The Foundation hired outside security firms to stress test the proposals against live applications while running its own research and outreach to software teams.
- ChainSecurity presented 152 findings; about half overlapped with the Foundation's internal scan, putting combined findings at roughly 200.
- Dedaub reviewed about 25,000 contracts with unusual gas usage and flagged 70 medium- and high-risk issues, including 22 the Foundation had missed.
- Foundation outreach to application teams identified three high-severity flaws and at least 25 medium-severity flaws, with most already being patched.
Tanguy Rocher of ChainSecurity presented his firm's results on the call. Foundation researcher Ansgar Dietrichs, who helped write the proposals, argued for proceeding with the upgrade while acknowledging that core engineers needed to communicate more clearly with outside developers whose applications would encounter the new rules.
Why did Hoodi slip in the first place?
Developers canceled the Hoodi activation scheduled for October 26. That testnet milestone also included a planned increase in Hoodi's gas limit to 200 million, per Cryptonews coverage republished by Yahoo Finance. The postponement gave engineers room to investigate how the new gas accounting would interact with contracts already deployed on Ethereum mainnet.
The October concerns included how applications handle failed operations. Researchers identified scenarios where a contract could retain a different amount of gas than its developers expected after a failure, potentially changing what the error-handling logic does next.
On November 6, the Foundation directed developers' attention to eight core technical assumptions. Most of the compatibility problems described this week stem from just two of them, narrowing the investigation to particular execution patterns.
Why does immutable code face protocol risk?
The episode exposes a structural tension in Ethereum's promise of immutable software. A deployed contract's code can remain unchanged while the network rules governing its execution shift beneath it.
For application teams, the question is whether existing logic still behaves as intended under revised rules. For Ethereum's engineers, the challenge is improving the network while accounting for software that its authors may have designed to run permanently, untouched.
Van der Wijden's upper estimate covers about 0.017% of Ethereum's roughly 90 million contracts. Even at that small share, the absolute number makes compatibility across deployed software a substantial testing task — and his few-hundred-thousand-dollar at-risk estimate gives the concern a concrete financial frame.
Why push higher state charges at all?
The fee changes aim to keep Ethereum's underlying database from expanding too quickly as future upgrades allow more activity per block. Charging separately for permanent state creation is part of managing the consequences of increased network capacity.
Some participants on the call argued that dropping EIP-8037 could obstruct developers' ability to deploy larger contracts and weaken Glamsterdam's broader security goals. Those arguments anchored the case for keeping the proposal alive despite the audit findings. Glamsterdam's wider roadmap also includes enshrined proposer-builder separation and block-level access lists, so the gas-accounting fight is unfolding inside a larger upgrade program.
What happens Thursday?
Core developers return to the proposals after Monday's stalemate, weighing the proposed gas rules against their interaction with existing applications. The Foundation has tied any resumption of Hoodi testing to resolution of the bugs, leaving the public testnet stage dependent on the compatibility work now underway among core engineers and application teams.
Glamsterdam has no confirmed release date. Thursday's discussion is the next decision point, and a resolution of the testing problems is the immediate milestone for anyone tracking the upgrade's path to mainnet.
via bitcoinfoundation.org (Source)
More from Daniel Okafor
Show full bio
Correspondent covering people, launches and day-to-day practice in crypto and web3 industry at Mempool Brief.
1,514 articles