Free test ether allows fake builders to outbid competitors and hold back transaction payloads, while client teams are given only half their normal review time before Sepolia.

Ethereum developers have set Oct. 6 as the testing date for the highly anticipated “Glamsterdam” upgrade. However, the announcement has also raised concerns that bad actors using “fake” identities could disrupt the upgrade’s dress rehearsal.

Developers warned that the Ethereum network’s auctions for processing the next block could be repeatedly won by anyone, including a teenager using free test Ether and a few fake accounts. The malicious actor could then withhold the required data and effectively disappear from the network, leaving blockchain traffic stalled.

This is significant because if network traffic were frozen, developers could face difficulties testing how the upgrade performs under real-world conditions. The Oct. 6 event is being treated as a dress rehearsal, or public trial, for the final Glamsterdam upgrade, which will be rolled out once developers are confident that everything is functioning safely.

Ethereum consensus developer Potuz said during Thursday’s core developer call that a person could easily create thousands of builders, rotate them, place unusually high bids and then withhold the required payloads. The approach, he warned, could be carried out by almost anyone with access to free test Ether.

Such an attack would not put mainnet funds at risk. Any disruption would be limited to Sepolia, where test Ether carries little to no real-world value, but blocks could be left without transaction payloads, disrupting the infrastructure testing required before Glamsterdam is deployed on Ethereum’s mainnet.

What Is Ethereum’s Glamsterdam Upgrade?#

Glamsterdam is set to be Ethereum’s next major upgrade, aimed at fitting more activity into each block without placing excessive strain on the computers that verify transactions. Alongside adjustments to gas pricing, the upgrade is expected to support a block gas limit of around 200 million, allowing more payments and trades to be processed before higher fees are triggered by user demand.

The upgrade brings the interaction between validators and specialized block builders directly into Ethereum’s protocol. Transaction blocks are assembled by builders, who then compete to have their proposals selected. After a validator accepts the winning bid, the underlying transactions are expected to be revealed by the builder.

Stay in the loop

Get crypto news before the market moves

Join thousands of investors who read our daily briefing.

No spam. Unsubscribe anytime.

That process can be easily exploited on a free test network. A malicious operator could place bids far above those offered by legitimate builders, repeatedly secure the winning position, and then withhold the promised payload.

Developers said current safeguards are generally designed to switch to locally built blocks only after several payloads have already been missed.

Potuz added that clients must also be able to detect and block individual builders, preventing an attacker from reappearing under a different identity and continuing to win bids.

The warning was issued a day after a large private rehearsal successfully completed the Glamsterdam transition, with the block gas limit being raised toward 200 million while finality was maintained.

Client teams have been given until Sept. 29 to deploy software that is ready for Sepolia. That provides just seven days ahead of the fork, leaving only half of the 14-day period normally set aside by Ethereum for security reviews and bug-bounty testing.

Developers accepted the tighter timeline because Sepolia is relatively centralized and can be recovered more easily if something goes wrong. Production builder software run by Titan and Ultrasound has also yet to complete the Glamsterdam fork transition, leaving another gap before the upgrade can be considered ready for mainnet deployment.

The next public test on Hoodi is tentatively scheduled for Oct. 27. Developers will review Sepolia’s results before deciding whether that timeline should be maintained, while no date has yet been set for the mainnet activation.