Fusaka:
Glamsterdam:
Hogota:
Gas limit:
State bloat:
Meeting Time: Monday, February 16, 2026 at 14:00 UTC (60 minutes)
The team reviewed progress updates on various Ethereum clients for the BlobT0 testnet, with P2P functionality nearly complete across multiple clients and nodes successfully running with specified latency parameters. Toni presented a proposal for implementing a hard cap on BlockLab access lists to prevent DOS attacks, which the team agreed to target for devnet 3 implementation. The team discussed the timeline and scope for devnet-3, including the implementation of EIPs 8037 and 7954 along with Ethereum 70 and 71, while also addressing concerns about client readiness and testing requirements, ultimately deciding to push the launch date to March 4th.
The team discussed updates on ConsensusLayer client progress for the BlobT0 testnet. Dustin reported that P2P for Teku is nearly ready, while Barnabas noted that LightHouse and Prysm are close to being ready. The current testnet has about 20 nodes, half being super nodes and half full nodes, with 80ms latency between nodes and a MEV node limited to 2 peers to mimic poor network performance. The team also addressed block access list updates, mentioning some bugs over the weekend that have been reverted, with current participation at 100%.
Toni presented an update on a proposed hard cap for the BlockLab access list to prevent DOS attacks from malicious builders. The current PR incorporates max withdrawals and max consolidations into the formula, but Toni suggested simplifying it by removing these constants and using only the gas limit as the constraint. The team agreed to target implementing this check in devnet 3, with Barnabas noting that clients should invalidate blocks with unrealistic access list sizes before execution. Toni also provided an update on a related PR for JSON RPC methods, which has been approved and updated to return JSON-encoded block-lab access lists instead of RLP bytes.
The team discussed progress on devnet implementation, with Andrew reporting that draft implementations are in progress but some test failures remain to be fixed. Marius provided an update on NFT DevNet 10, which was launched with ETH 70 activation, though some clients like Nethermind and Geth are experiencing syncing issues due to large receipts. The team confirmed that ETH 70 is now a prerequisite for ETH 71, and Toni clarified that only optimizations are expected for BlockNove Access List 3 beyond the current changes. Barnabas proposed targeting devnet 3 for launch around February 25th, with Fabio suggesting the possibility of merging ETH 70 and ETH 71 into a single P2P protocol version.
The team discussed the scope and timeline for devnet-3, focusing on implementing EIPs 8037 and 7954 along with ETH 70 and 71. Concerns were raised about the lack of tests for EIP 8037, with Mario and Spencer highlighting the need for additional time to address potential issues. The group agreed to aim for a new target date of March 4th instead of February 25th to allow for proper testing and implementation. Barnabas also mentioned the importance of adding eth_simulateV1 RPC support for benchmarking purposes.
The team discussed including Ethereum 71 in devnet 3 scope, with Andrew and Karim expressing concerns about implementation challenges, particularly for Erigon and Besu clients. After clarifying that ETH 71 is optional and only used for sync improvements, the group decided to include it as optional in devnet 3, allowing clients to implement it at their discretion. Stefan provided an update on Teku’s progress, noting that while they implemented the Beacon API for execution payloads, they still don’t support finalized retrieval of execution payloads, which should be sufficient for devnet zero synchronization.
The team discussed the status of various Ethereum clients for the upcoming testnet launch. Stefan indicated that Teku would not be ready for the February 18th mid-February deadline, needing at least another week for a complete version. Gaia reported that Lighthouse had a devnet branch with block production working but facing bugs in Gossip envelope processing, and Dustin mentioned that Nimbus was working on envelope syncing and Gossip functionality, also not ready for the February deadline. NFLaig noted that Lodestar’s ForkChoice implementation was nearly complete but would need 1-2 weeks to finalize. Barnabas also shared that Consensus passed all tests for epoch transitions but was not production-ready. The conversation ended with a brief discussion about the performance-2 network issues, where Marius reported that all clients except Nethermind had failed, leading to a discussion about potentially relaunching the CL chain.
+1f1Tg5S)+1f1Tg5S)+1f1Tg5S)YouTube Stream Links: