Reporting window: November 2, 2026, 00:00:00 to November 9, 2026, 00:00:00 WITA (Asia/Makassar), end exclusive.

Ethereum made compatibility work an earlier, explicit responsibility for proposal champions this week. Fredrik’s expanded impact-analysis and outreach guidance merged on November 4. The guidance asks champions to assess protocol and application effects at CFI and contact affected parties before testnets. Meanwhile, Nethermind researchers measured a tradeoff between frame-validation capacity and honest transaction admission under attack. Together, the developments sharpen the practical question for upgrade readiness: which workloads and users does the evidence actually cover?

Champions must account for application behavior before testnets

The merged guide tells champions to assume deployed contracts may depend on existing behavior, then check. Gas costs, opcodes, precompiles, transaction types and account semantics generally warrant application-layer analysis. Champions should record their reasoning and findings in the discussion thread and Forkcast entry, so client teams can use the evidence when signaling support.

The guidance also calls for outreach early enough to receive and address feedback. Editorially, the useful next signal is a proposal whose compatibility findings change its design or rollout plan. A documentation merge establishes an expectation; it does not establish that any particular upgrade has met that expectation.

Frame validation protects blocks at a cost to admission

For EIP-8141: Frame Transaction, Nethermind’s Stefano De Angelis, Daniil Ankushin and Manuel Arto tested unpaid validation floods on three machines, restricting the node to one CPU core. Their November 7 Bali-time report finds that pausing validation during block processing makes a 300,000-gas cap viable under their tested delay budgets at 100 failing transactions per second. The authors derive mainnet-block delays from measured contention and block-processing times, rather than demonstrating network-wide performance.

The admission result deserves equal attention. With a one-second validation CPU budget per head, the tested nodes admitted only 12–27% of honest frame transactions during the 300,000-gas, 100-transaction-per-second flood. The editorial implication is to evaluate honest-user access alongside block latency before raising the cap. Replication across clients and peer configurations would be more persuasive than treating one client’s mitigation as a settled network parameter.

Toni Wahrstätter separately opened a draft clarification excluding calldata charges from MAX_VERIFY_GAS. His rationale distinguishes computation from bandwidth and block-space charges, which could exhaust the limit through large post-quantum signatures alone. The draft remains an accounting clarification under review, not an adopted increase in the validation budget.

Payload-gossip protection moves to a parent-derived bound

Lodestar contributor nflaig replaced the bid-based transaction-count proposal with a parent-payload-based bound. The proposed rule uses the gas limit permitted above the closest full ancestor’s payload, falling back to the bid when no full ancestor exists. The change revises the input to the bound, rather than merely adding another check to the earlier proposal.

The corresponding open Lodestar implementation reads transaction counts from SSZ offsets before deserialization and checks again before signature verification. Its author explicitly conditions merging on release of the specification change. That provides a concrete implementation to review, but neither pull request establishes deployed protection. The next signal is agreement on the rule, followed by client adoption and adversarial envelope tests.

Execution tests stop treating valid rejection differences as failures

Execution-specs merged a correction to oversized legacy chain-ID exception expectations on November 3. The affected cases now accept either INVALID_SIGNATURE_VRS or INVALID_CHAINID, reflecting different rejection paths rather than different transaction validity.

Reviewer danceratopz reported that the correction cleared five failures on Erigon v3.8.1 and twelve on Geth v1.17.10. The reviewer requested a patch fixture release to remove those false positives. For readiness reviews, the distinction matters: fewer red tests here mean more accurate expectations, not seventeen repaired consensus bugs. Watch the fixture release and downstream reruns before interpreting changed pass counts as client progress.

Sources and further reading