Key Takeaways
- • Ethereum's Glamsterdam upgrade has been delayed: a second round of testing set for October 26 on Hoodi was postponed on October 22 while developers study how a gas fee change affects apps already running on Ethereum.
- • The fee change raises costs for actions that permanently add new data to the network, such as creating accounts or deploying apps, and is the component now under review.
- • Glamsterdam targets the main Ethereum network's own throughput, allowing transactions that don't overlap to be processed in parallel, versus the Layer 2 focus of recent upgrades.
- • ChainSecurity flagged that the new costs could cause legitimate actions to be recorded as permanent failures, and Nixo, an Ethereum developer, said the delay "more than likely" pushes mainnet into 2027.
- • Curve Finance argues the main gain is more predictable blockspace access during volatility, but Coinbase Institutional researchers caution that a busier network does not automatically lift ETH's price.
Summary
Ethereum's next major protocol upgrade, Glamsterdam, is taking longer than planned. Testing began on October 6 on Sepolia, one of the network's main test networks, as scheduled by the Ethereum Foundation. A second round, set for October 26 on Hoodi, was postponed on October 22 while developers examine how a fee change could interact with applications already live on the network, according to the Investing News Network.
The upgrade follows Dencun (March 2024), Pectra (May 2025) and Fusaka (December 2025). Unlike recent upgrades that mainly benefited layer 2 networks such as Arbitrum, Optimism and Base, Glamsterdam is designed to increase the main network's own capacity. Its three main components are parallel transaction processing, enabled by a block-level access list; protocol-level block building that removes reliance on third-party intermediaries; and higher gas costs for actions that permanently add data, such as creating accounts or deploying apps. Glamsterdam does not by itself raise the block gas limit, which validators set; on Sepolia validators can raise it to roughly three times the current mainnet level, but no mainnet decision has been made.
The postponement stems from concerns raised by ChainSecurity, a blockchain code auditor, which found the new costs could alter how some existing apps behave without any code change. In certain cases, an actor could send too little gas, causing a legitimate action to fail permanently rather than being reverted for retry. This is the second schedule slip; July developer plans had Sepolia upgrading in late August. The Ethereum Foundation has not set a mainnet date, and Nixo said the delay "more than likely" moves the launch into 2027.
Curve Finance, a long-running DeFi protocol, said that "for DeFi on the whole, the main gain from this upgrade is more predictable access to blockspace when activity picks up," potentially tightening prices and improving routing. It cautioned that the real test will come during market turmoil: "A higher gas limit on paper means little if transactions still stall during volatility when people need them."
For ETH holders, the relevance lies in where activity settles. Mainnet fees are partly burned, reducing supply, while layer 2 fees are posted to mainnet in batches and shared across users. Coinbase Institutional researchers David Duong and Colin Basco argued in March that activity and stablecoin balances had "tilted back toward Ethereum," and that Glamsterdam "could make that shift durable." According to DefiLlama data as of October 28, far more stablecoin value sits on Ethereum's mainnet than on Base, Arbitrum and OP Mainnet combined. The researchers also warned that "the market is increasingly separating ecosystem activity from tokenholder accrual," and that cheaper transactions could reduce ETH burned unless total activity grows enough to offset it.
The upgrade follows Dencun (March 2024), Pectra (May 2025) and Fusaka (December 2025). Unlike recent upgrades that mainly benefited layer 2 networks such as Arbitrum, Optimism and Base, Glamsterdam is designed to increase the main network's own capacity. Its three main components are parallel transaction processing, enabled by a block-level access list; protocol-level block building that removes reliance on third-party intermediaries; and higher gas costs for actions that permanently add data, such as creating accounts or deploying apps. Glamsterdam does not by itself raise the block gas limit, which validators set; on Sepolia validators can raise it to roughly three times the current mainnet level, but no mainnet decision has been made.
The postponement stems from concerns raised by ChainSecurity, a blockchain code auditor, which found the new costs could alter how some existing apps behave without any code change. In certain cases, an actor could send too little gas, causing a legitimate action to fail permanently rather than being reverted for retry. This is the second schedule slip; July developer plans had Sepolia upgrading in late August. The Ethereum Foundation has not set a mainnet date, and Nixo said the delay "more than likely" moves the launch into 2027.
Curve Finance, a long-running DeFi protocol, said that "for DeFi on the whole, the main gain from this upgrade is more predictable access to blockspace when activity picks up," potentially tightening prices and improving routing. It cautioned that the real test will come during market turmoil: "A higher gas limit on paper means little if transactions still stall during volatility when people need them."
For ETH holders, the relevance lies in where activity settles. Mainnet fees are partly burned, reducing supply, while layer 2 fees are posted to mainnet in batches and shared across users. Coinbase Institutional researchers David Duong and Colin Basco argued in March that activity and stablecoin balances had "tilted back toward Ethereum," and that Glamsterdam "could make that shift durable." According to DefiLlama data as of October 28, far more stablecoin value sits on Ethereum's mainnet than on Base, Arbitrum and OP Mainnet combined. The researchers also warned that "the market is increasingly separating ecosystem activity from tokenholder accrual," and that cheaper transactions could reduce ETH burned unless total activity grows enough to offset it.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.