Ethereum Sets Glamsterdam Test for October 6 and Warns of Fake Builders

Ethereum developers have set October 6 for the public test of the Glamsterdam upgrade, CoinDesk reports. The date came with a warning on the core developers call. Fake builders, meaning specialized block assemblers, could jam the test network. User funds are not at risk, but the upgrade schedule could slip.

What exactly was scheduled for October 6?

Glamsterdam is the next major Ethereum upgrade. It is meant to fit more activity into each block without overloading the computers that verify those blocks. Together with changes to gas pricing (the fee for computation), it should raise the block gas limit to roughly 200 million. That leaves room for more payments and trades before users start bidding fees higher. The gas limit is a ceiling on computation in a single block, so the higher it is, the more transactions the network handles at once.

October 6 is a dress rehearsal, not a launch. It will run on Sepolia, a test network where coins cost nothing. Developers use it to try changes before they reach the main network. The next stage, on the Hoodi network, is tentatively set for October 27. There is no activation date for the main network yet. In news like this, "fork" means the moment a network switches to new rules. It happens on test networks first, and only then on the main one.

How does the block auction work, and where is the weak spot?

To understand the risk, it helps to recall who builds blocks on Ethereum. A validator, the participant that confirms blocks, rarely assembles them itself. It takes an offer from a builder that bids the highest fee. That is because arranging transactions profitably is hard, so validators hand the job to specialists. Glamsterdam moves this arrangement into the protocol, so the rules of the auction become part of the network itself.

Here is how it looks step by step.

• The bid: builders submit offers stating how much they will pay for the right to build the next block.
• The validator picks the highest bid and accepts it.
• The builder must then reveal the transactions inside the block.
• If the builder vanishes without the data, the block stays empty and the network loses a beat.

On a test network the weak spot is obvious. Test ether is free, so an attacker can spin up hundreds of builders, offer the highest prices, win every auction and hand over nothing. There is no penalty for failing, because there is nothing to lose. The network would not stop for good, but every empty block slows the test. Existing safeguards switch to locally built blocks only after several promised transaction sets in a row fail to arrive.

"I can just spin up a thousand builders, rotate them, offer very high bids, and not produce payloads. Any teenager can do this."

Potuz, Ethereum consensus developer, from the core developers call on September 17, 2026

Why was the review window cut to seven days?

Clients, the programs that run network nodes, must ship a release ready for Sepolia by September 29. That leaves seven days before the fork. Ethereum's normal process reserves 14 days for security review and bug bounty testing, so the window was halved.

Developers accepted this knowingly. Sepolia is relatively centralized, and if something breaks it is easier to recover. Builder software used by the Titan and Ultrasound teams has not yet completed the Glamsterdam transition. That is one more gap before the main network can be called ready. The trade-off is clear: a faster schedule lowers the risk of delay but leaves less time to find bugs.

A day earlier, a large private rehearsal completed the Glamsterdam transition and lifted the gas limit to roughly 200 million. The network kept finality, the irreversible confirmation of blocks, throughout. Next comes checking how a public network survives such a change, and that is what Sepolia will show.

Does this threaten your ETH?

No. The attack would only work on Sepolia, where test ether has no price. The main network and the money in wallets stay out of danger. Only the test can suffer. If blocks keep coming out empty, developers cannot see how the upgrade copes with real load.

That is why client teams need to learn to spot and reject individual builders. According to Potuz, otherwise an attacker simply returns under a new identity and wins the auction again.

An ETH holder has nothing to do at this stage, because the test runs on a network without real money. Anyone planning to sell Ethereum for hryvnia can keep their plans as they are. But more room in blocks should keep fees from spiking as often at peak hours. We at Kurslog watch this because fees decide how much reaches you after an exchange.

What comes next?

If Sepolia runs without failures, teams will move on to the Hoodi test. The October 27 date may still change after the October 6 results, and the main network launch date depends on both tests. Sepolia will show whether clients hold up under load without failures and reject malicious builders correctly.

At the same time, the community is talking about faster blocks. According to Cointelegraph, Ethlabs is pushing proposal EIP-8198 (Quick Slots), which would first cut block time from 12 to 10 seconds. The team published a piece quoting 20 decentralized finance founders who back the idea, and the nonprofit Ethereum Institutional tied it to growing institutional activity on the network. Backers want it in the Hegotá upgrade, the stage after Glamsterdam, which developers could begin implementing in late 2026.

Other networks are moving the same way. In August Solana cut its slot time from 400 to 350 milliseconds, and in June its foundation spoke of aiming for 200. Zcash holders voted to cut the target block time from 75 to 25 seconds. The race for speed goes on.