The Ethereum Foundation has warned that wallets, indexers and gas estimation tools may stop working when the Glamsterdam upgrade introduces a new gas model to the network.
In a notice published Monday, the EF's Protocol DevOps team said any tool that relies on a hardcoded maximum gas limit "will break" and must be updated. The foundation urged developers to test their systems on Plataberget, a public testnet launched on August 13 that is intended to operate for several months ahead of the fork's wider rollout.
A Second Gas Dimension
The centerpiece of the change is EIP-8037, which introduces a separate state-gas dimension for operations that create new state. Under the new model, a plain ETH transfer to an existing account would still cost the familiar 21,000 gas — but sending ETH to a brand-new account would incur an additional state-gas charge on top.
The implication is that software assuming 21,000 gas covers every ETH transfer, or that uses only a single gas dimension when estimating transaction costs, will produce wrong estimates or fail outright once the rules activate. Wallets, block explorers, indexers and infrastructure providers all fall inside that blast radius.
Glamsterdam is much more than a gas tweak. The upgrade also includes enshrined proposer-builder separation, block-level access lists and increased limits for contract and initialization-code sizes — one of the most consequential hard forks in Ethereum's recent roadmap.
Tight Testing Timeline
According to upgrade tracker Forkcast, the Glamsterdam fork is scheduled to activate on one network this Thursday, before deploying to the Sepolia and Hoodi testnets and eventually mainnet. That gives teams a compressed window to verify their tooling against the new gas rules on Plataberget.
The warning revives memories of past upgrade friction, where insufficiently tested tooling caused issues at the application layer even when the protocol itself activated cleanly. The EF's early, explicit alert — naming the exact class of software at risk — is designed to avoid a repeat.
For developers, the message is direct: if your code assumes one gas number, test it now. For users, the practical risk is broken wallets and inaccurate fee estimates in the weeks after mainnet activation if tooling lags behind the fork.