Reporting window: October 26, 2026, 00:00:00 to November 2, 2026, 00:00:00 WITA (Asia/Makassar), end exclusive.

Ethereum’s upgrade work now faces two distinct tests: contract compatibility for Glamsterdam and integration readiness for Hegotá. At ACDT, Fredrik reported commissioning a broader ChainSecurity and Dedaub assessment, with an estimated two-week turnaround. ACDC meanwhile advanced legacy deposit-field removal to SFI and moved custom validator sweeps and faster payload broadcast to CFI. The week’s practical distinction is between choosing scope and demonstrating safe behavior. A merged builder-deposit correction and FOCIL block-building failures reported under load show why the next useful evidence must come from implementation and testing.

Glamsterdam’s next decision depends on affected contracts

Toni Wahrstätter said Hoodi was not ready for rescheduling while teams assessed gas-compatibility findings and contacted affected applications. Maria Silva favored application fixes first, reserving targeted specification changes for significant problems that applications could not resolve. Łukasz Rozmej argued for exploring specification changes in parallel, so the protocol would have a prepared alternative if outreach exposed an unavoidable incompatibility. These were positions on how to proceed, not an agreed redesign.

The immediate signal to watch is the assessment’s treatment of unupgradeable contracts. Fredrik raised mitigation and user exit as questions for the review; the call did not establish that every affected application has a workable remedy. A new activation date would be less informative than evidence that the proposed remedies preserve the relevant contract invariants.

The delay also reopened a concrete consensus change. Lido’s Dima Gusakov warned that fork-time builder onboarding could redirect a validator deposit to a builder sharing the public key. The consensus-specs repository merged withdrawal-credential-based routing on October 27, Bali time. Validator deposits stay in the validator queue, while valid builder deposits register or fund builders even when the public key appears in both registries. This closes a specification-level separation problem. ACDC’s request for explicit testing of the consensus-breaking change remains the relevant follow-through, rather than assuming the merge proves client interoperability.

Hegotá advances cleanup, sweeps and payload propagation

Alex Stokes confirmed SFI for EIP-8015: Remove deposit and eth1data fields, after Barnabas Busa said the change was already under devnet testing. The proposal removes obsolete deposit and Eth1 data fields from beacon structures. The call’s decision advances the upgrade commitment without establishing mainnet activation.

Dima Gusakov’s case for EIP-8148: Custom sweep threshold for validators focused on letting compounding validators receive automatic rewards below the default 2,048 ETH threshold. After Lighthouse indicated it could revise its position in response to other teams’ support, Stokes confirmed CFI. For staking operators, the next useful signal is a tested implementation of configurable sweeps, including the cross-layer request path.

Kamil Salakhiev also secured CFI for EIP-8411: Fast Execution Payload Broadcast. The design splits payload envelopes into independently verifiable chunks, allowing forwarding before a node receives the whole payload. Salakhiev presented simulation results as the performance case; those results do not establish production-client gains. Watch propagation measurements with actual clients and constrained peers before treating the proposal as additional gas-limit or shorter-slot capacity.

FOCIL exposes an integration problem as Bogota fixtures arrive

For EIP-7805: Fork-choice enforced Inclusion Lists (FOCIL), Busa reported execution clients discarding an existing payload when inclusion-list attributes arrived, then rebuilding too late under full blocks. His proposed direction was to append missing inclusion-list transactions instead. ACDC referred the question for execution-layer discussion after participants challenged whether the change was straightforward. This is an unresolved construction-policy problem, not evidence that the protocol has adopted a universal append-only rule.

Separately, execution-specs published the first Bogota devnet 2 fixture release on October 28, Bali time. The fixtures bring EIP-8141: Frame Transaction testing onto the aggregate max_fee budget, including its GASPRICE treatment and signature-length commitments. The release tracker still records the devnet launch date as TBD. The next milestone is a named multi-client run against that release, followed by combined testing of frame transactions and inclusion lists.

Sources and further reading