Technology•

Ethereum's Glamsterdam Fee Overhaul Could Disrupt 15,000 Smart Contracts, Foundation Developer Tells Colleagues

An Ethereum Foundation developer estimates Glamsterdam's storage fee changes could disrupt up to 15,000 contracts. Auditors logged hundreds of issues, and core developers revisit the proposals on Thursday.

Written By :MINRK
Ethereum's Glamsterdam Fee Overhaul Could Disrupt 15,000 Smart Contracts, Foundation Developer Tells Colleagues

Photo by Gabriel Heinzer on Unsplash

Mentioned in This Article

ETH
ETH
Ethereum
$1,808.86
0.86%

1. Ethereum Upgrade Estimate Puts 15,000 Contracts in Question

Ethereum Foundation developer Marius van der Wijden told fellow engineers on a developer call on Monday, November 16, that the planned Glamsterdam Ethereum upgrade could disrupt as many as 15,000 smart contracts, with a few hundred thousand dollars potentially at risk. The call closed without agreement, and core developers are due to reconsider the fee changes behind the problem on Thursday, November 19.

The figure adds a broad, network-wide estimate to a debate that had so far been driven by individual bug reports. Compatibility concerns have already held up testing of Glamsterdam since October, and the new number is considerably larger than what earlier reviews had turned up. It arrives alongside separate audit work that has catalogued hundreds of specific places where existing applications could misbehave once the new rules take effect.

Van der Wijden framed the exposure as limited in proportion to the network as a whole. Ethereum (ETH) hosts roughly 90 million smart contracts, the self-executing programs that power financial applications and asset transfers on the chain, so 15,000 represents a very small share of the total. He likewise described the sum of money at stake as minor when set against the size of the Ethereum ecosystem. The count of affected programs nonetheless far exceeds anything previous assessments had identified.

2. What Is the Glamsterdam Upgrade Changing?

Glamsterdam would rewrite how Ethereum charges for storing and retrieving data, through two proposals known as EIP-8037 and EIP-8038. Those two Ethereum Improvement Proposals sit at the center of the dispute, because they alter costs that many deployed applications treated as fixed.

To see why that matters, it helps to recall how the network prices activity. Every operation a smart contract performs consumes gas, a unit that measures the computing work the network has to carry out. Users pay for that gas, and different operations carry different gas prices set by the protocol itself. Developers write their code with those protocol-level prices in mind, and some hard-code assumptions about them directly into their contracts.

EIP-8037 introduces a separate charge for creating new permanent state, meaning data that the network must keep indefinitely. Deploying a contract is one example, and adding a new storage entry is another. Under the proposal, the cost of a deployment would climb from about 32,000 gas to about 183,600 gas. Writing a new storage slot would rise from about 20,000 gas to about 97,920 gas. In both cases the price increases several times over.

Engineers designed the steeper charges to keep the database underlying the network from expanding too quickly. That concern becomes more pressing because future updates are intended to fit more activity into each block. More activity per block means more opportunities to add permanent data, and the higher fees are meant to act as a brake on that growth.

Permanent state is a particular burden for a blockchain because it never goes away on its own. A transfer is processed once and then becomes history, but a deployed contract or a filled storage slot has to stay available to every machine that verifies the chain. The larger that body of data becomes, the more demanding it is to run the software that keeps the network operating. Charging more for the operations that enlarge it is a way of making those who add to the long-term load bear more of its cost.

The difficulty is that a price signal aimed at future behavior also lands on code written years ago. New projects can design around the higher figures. Programs already on the chain were built to the old ones, and many of them cannot be edited.

3. Why Would Existing Contracts Break?

Existing contracts can fail because a large number of them were built on the assumption that storage and deployment costs would remain predictable. When those costs jump, logic that once had comfortable headroom can run out of gas partway through an operation and revert.

A second group of applications faces a related problem. These contracts hand a fixed gas allowance to other contracts when they call them. The allowance was sized for the old fee schedule. If the called contract now needs to create storage at the new, higher price, the allowance may no longer cover the work, and the call fails even though nothing about either contract's code has changed.

To gauge how widespread such failures might be, the Ethereum Foundation commissioned outside security firms to stress test the network under the proposed fee rules. The aim was to learn which major applications would stop working properly, rather than to rely solely on internal analysis or on developers volunteering reports about their own code.

4. What the Outside Auditors Found

Hundreds of potential trouble spots have emerged from the commissioned reviews, separate from the 15,000-contract figure that came out of broad scans. The detailed numbers were set out in a summary of Monday's call posted on the Ethereum Magicians forum, the venue where protocol changes are discussed in public.

Tanguy Rocher of the security firm ChainSecurity reported 152 findings. About half of them overlapped with issues the Foundation's own researchers had already identified, which brings the combined total across the two efforts to roughly 200.

A second firm, Dedaub, examined about 25,000 contracts that showed unusual patterns of gas usage. It flagged 70 issues rated medium or high risk. Of those, 22 were problems the Foundation had not caught in its own work, which indicates that the internal review alone would have left a meaningful number of cases undiscovered.

The two sets of results measure different things, and they should be read that way. The 15,000 figure is an upper estimate of how many deployed programs might be touched by the change, produced by scanning the chain at scale. The audit numbers count individual issues that reviewers examined and classified by severity. A broad scan shows how far the effects could reach, while the audits show how serious particular cases are and whether they can be fixed.

Most of the problems trace back to a narrow set of causes. On November 6 the Foundation published a warning to developers listing eight core technical assumptions that the upgrade would affect. The bulk of the audit findings stem from just two of those eight. The Foundation's own outreach to application teams surfaced three flaws rated high severity and at least 25 rated medium severity. Most of them are being patched.

5. The Immutability Question

Beneath the technical findings lies a disagreement about what Ethereum's promise of immutable software really covers. A contract's code cannot be altered after it has been deployed, and that permanence is one of the properties users and developers rely on. Yet the same unchanged code can behave differently when the rules of the network beneath it are revised. A contract that worked on one fee schedule may stall on another without a single line having been edited.

Ansgar Dietrichs, a Foundation researcher who helped write the proposals, argued on the call that the upgrade should go ahead. He also conceded that core engineers need to communicate more clearly with developers outside the core group, an acknowledgment that application teams have not always had enough warning about changes that affect them.

Other participants cautioned against the alternative of dropping the change. In their view, scrapping EIP-8037 could block larger contract deployments and weaken the security goals the upgrade is meant to achieve. That leaves developers weighing two costs against each other: the disruption to a small fraction of existing programs if the proposals stay, and the loss of planned capabilities and protections if they are removed.

6. Testing Delays on the Hoodi Network

Glamsterdam's schedule has already slipped because of these compatibility problems. They have held up testing since October and forced developers to postpone the rollout on Hoodi, a public test network. Test networks let client teams and application developers run a new version of the protocol under realistic conditions before it reaches the main chain, so a pause at this stage delays every step that follows.

The Foundation has said Hoodi will remain on hold until the bugs are fixed. With the test deployment paused and the fee proposals still under dispute, the main Glamsterdam upgrade has no confirmed release date.

For teams running applications on Ethereum, the practical consequence is a period of uncertainty. Projects whose contracts depend on fixed gas allowances or on stable storage costs have reason to review their code against the Foundation's November 6 list of assumptions, particularly the two that account for most of the reported issues. Contracts that cannot be modified may need new versions deployed or workarounds built around them, and that work is easier to plan once the final fee figures are settled.

7. What Happens Next

Core developers will take up the proposals again on Thursday, November 19, after Monday's meeting ended with no consensus. The options discussed so far range from proceeding with the fee changes as written, as Dietrichs advocates, to removing EIP-8037, which other participants warned would carry its own costs.

Several points remain open after the first round of discussion. The 15,000 figure is an estimate drawn from broad scans, while the audit totals of roughly 200 combined findings and 70 medium- and high-risk issues come from closer inspection of specific contracts. Most of the high- and medium-severity flaws found through the Foundation's outreach are being patched, but the test network stays paused until the bugs are resolved.

Until developers agree on the fate of EIP-8037 and EIP-8038 and testing on Hoodi resumes, there is no timeline for Glamsterdam to reach the main network. Thursday's session is the next scheduled opportunity for that agreement to take shape.

Related Articles

NEWSLETTERS

Don't miss another story.

Subscribe to the MINRK Newsletter today.

Tick the newsletters you want. No selection? You'll get MINRK Headlines.

By signing up, you will receive emails about MINRK products and you agree to our terms of use and privacy policy.

MINRK Daybook

Market analysis for crypto traders and investors.

EVERY WEEKDAY

MINRK for Advisors

Defining crypto, digital assets and the future of finance for financial advisors.

EVERY THURSDAY

MINRK Protocol

Exploring the tech behind crypto one block at a time.

WEEKLY

MINRK Long & Short

A must read for institutions. Insights, news and analysis delivered weekly.

EVERY WEDNESDAY

MINRK Headlines

The biggest crypto news and ideas of the day.

EVERY WEEKDAY

MINRK Policy

Examining the intersection of cryptocurrency and government.

WEEKLY

MINRK Research

Join thousands of readers who rely on MINRK for data-driven insights on the latest digital asset trends.

MONTHLY