# peerdas-devnet-4 specs :::info :mega: Peerdas-devnet-4 is live as of Feb 7! https://peerdas-devnet-4.ethpandaops.io/ Dora: https://dora.peerdas-devnet-4.ethpandaops.io/ ::: :::info :mega: Note: PeerDAS is activated at the `FULU_FORK_EPOCH` ::: ## EIP List for peerdas-devnet-4 The list below links the specific commit versions of the EIPs included in devnet-3. - [EIP-7594: PeerDAS - Peer Data Availability Sampling](https://github.com/ethereum/EIPs/blob/master/EIPS/eip-7594.md) ### Spec changes for peerdas-devnet-4: Data column RPC must work prior to launching devnet **Consensus Specs** [v1.5.0-alpha.10](https://github.com/ethereum/consensus-specs/releases/tag/v1.5.0-alpha.10) :heavy_check_mark: [EIP-7594: Decouple network subnets from das-core](https://github.com/ethereum/consensus-specs/pull/3832) - Merged :heavy_check_mark: [Rebase eip7594 onto Electra](https://github.com/ethereum/consensus-specs/pull/3893) Merged :heavy_check_mark: [P2P clarifications when introducing engine_getBlobsV1](https://github.com/ethereum/consensus-specs/pull/3864) :heavy_check_mark: Merged [Fulu: Remove V3 of blob sidecar by root/range RPC](https://github.com/ethereum/consensus-specs/pull/4073) :heavy_check_mark: Merged **Beacon Metrics** [PR-13](https://github.com/ethereum/beacon-metrics/pull/13) - Peerdas metrics :exclamation: Open **Engine API** [PR-559](https://github.com/ethereum/execution-apis/pull/559) - Define engine_getBlobsV1 Merged :heavy_check_mark: **Execution APIs** [Define engine_getBlobsV1](https://github.com/ethereum/execution-apis/pull/559) Merged :heavy_check_mark: **Beacon API** :::info :mega: [Validator custody ](https://github.com/ethereum/consensus-specs/pull/3871) - This is up to clients to decide if they would like to implement it. It isn't a requirement for devnets. ::: ## Kurtosis Interop Conifg (Pre-devnet testing) ```yaml participants: # Super nodes # ----------- # Lighthouse - cl_type: lighthouse cl_image: ethpandaops/lighthouse:peerdas-devnet-4 cl_max_mem: 4096 cl_extra_params: - --subscribe-all-data-column-subnets # Prysm - cl_type: prysm cl_image: ethpandaops/prysm-beacon-chain:peerDAS-01705d1 cl_max_mem: 4096 cl_extra_params: - --minimum-peers-per-subnet=1 - --subscribe-all-subnets # Teku - cl_type: teku cl_image: ethpandaops/teku:das cl_max_mem: 4096 cl_extra_params: - --p2p-subscribe-all-custody-subnets-enabled - cl_type: grandine cl_image: ethpandaops/grandine:peerdas-fulu cl_max_mem: 4096 cl_extra_params: - --subscribe-all-subnets # Nimbus - cl_type: nimbus cl_image: ethpandaops/nimbus-eth2:columns cl_extra_params: - --debug-peerdas-supernode=true - --sync-light-client=no # Lodestar - cl_type: lodestar cl_image: ethpandaops/lodestar:peerDAS cl_extra_params: - --persistNetworkIdentity - --supernode # Full nodes # ----------- # Lighthouse - cl_type: lighthouse cl_image: ethpandaops/lighthouse:peerdas-devnet-4 cl_max_mem: 4096 # Prysm - cl_type: prysm cl_image: ethpandaops/prysm-beacon-chain:peerDAS-01705d1 cl_max_mem: 4096 cl_extra_params: - --minimum-peers-per-subnet=1 # Teku - cl_type: teku cl_image: ethpandaops/teku:das # Grandine - cl_type: grandine cl_image: ethpandaops/grandine:peerdas-fulu # Nimbus - cl_type: nimbus cl_image: ethpandaops/nimbus-eth2:columns cl_extra_params: - --debug-peerdas-supernode=false - --sync-light-client=no # Lodestar - cl_type: lodestar cl_image: ethpandaops/lodestar:peerDAS cl_extra_params: - --persistNetworkIdentity global_log_level: debug network_params: electra_fork_epoch: 1 fulu_fork_epoch: 2 additional_services: - dora - spamoor_blob - prometheus_grafana snooper_enabled: false dora_params: image: ethpandaops/dora:fulu-support ``` ## Client Interop Readiness Manu's test scenarios: https://hackmd.io/@manunalepa/BJzNsCnvyx ### Prysm | Scenario | Grandine | Lighthouse | Lodestar | Nimbus | Prysm | Teku | | -------- | -------- | ---------- | -------- | ------ | ----- | ---- | | 1a | ✅ | ✅ | ❓ | ✅ | ✅ | ✅ | | 1b | ✅ | ✅ | ❓ | ✅ | ✅ | ✅ | | 2a | ✅ | ✅ | ❓ | ✅ | ✅ | ✅ | | 2b | ✅ | ✅ | ❓ | ✅ | ✅ | ✅ | | 3a | ❌ | ✅ | ❓ | ✅ | ✅ | ✅ | | 3b | ❌ | ✅ | ❓ | ✅ | ✅ | ✅ | | 4 | ❌ | ✅ | ❓ | ✅ | ✅ | ❌ | ### Prysm - Lighthouse - [x] 🐛 Lighthouse does not unsubscribe from blob topics after Fulu - [x] 🐛 FIXED: Prysm continue to send gossip blobs on Fulu slots - [x] ❓ Prysm not getting range data columns back from Lighthouse due to rate limiting. - This is because there's only one single node to request from and the requests were sent too quick in a short period, due to the size of the network - Jimmy: it's possible to add `--disable-rate-limiting` flag to the Kurtosis config for local testing purpose (@Manu suggested to leave it on by default, as node must be able to sync even when rate limiting is present) - [x] ❓ Lighthouse may need to review rate limits - [x] 🐛 Prysm sending data column by range requests for a block with no blobs (found the bug, fixing it) ### Prysm - Teku - [x] ✅ Blocks and blobs after Fulu fork - [ ] 🤨 Prysm super node crashes, but may not be PeerDAS related ### Prysm - Grandine - [x] ✅Blocks and blobs after Fulu fork ### Teku - Lighthouse - [x] ✅ Blocks and blobs after Fulu fork ### Teku - Grandine - [x] ✅ Blocks and blobs after Fulu fork ### Grandine - Lighthouse - [x] ✅ Blocks and blobs after Fulu fork ## Reference spec for previous devnet: https://notes.ethereum.org/@ethpandaops/peerdas-devnet-3
Last changed by  
 
ethpandaops For the ethpandaops team, devops for EL/CL layers
673

Read more

peerdas-devnet-6 specs

:::info :mega: Devnet-6 targets launch in mid/late Mar 2025 ::: :::info :mega: Note: PeerDAS is activated at the FULU_FORK_EPOCH ::: Note, changes include:

3/13/2025
osaka-devnet-0 specs

:::info :mega: Devnet-0 targets launch at ... ::: EIP List for osaka-devnet-0 The list below links the specific commit versions of the EIPs included in devnet-0. Spec changes for eof-devnet-0: Execution Specs

3/13/2025
Hoodi testnet

This write up will help you understand the initial distribution of validators among the new hoodi testnet. This testnet will launch in case we decide to do plan A of this writeup. Given: Ethpandaops controls 33% of the validators. The remaining 67% should be split among:11 client teams (each with approx. the same share). 5 staking operators (each with a slightly smaller share). Total staked ETH: At least 20M ETH. Each validator balance: Between 32 ETH and 2048 ETH.

3/12/2025
Post-finality Holesky path forward

The Problem Currently we have sepolia, a closed validator set testnet. This network enables application developers to test their applications in a closed environment. This network is not optimal for validators as they can't make deposits. We have holesky, which in it current state is not really suitable for testing validator related features. Due to the recent unfinality event, and the fact that the exit queue is full for the next ~1.5 years, full validator lifecycle tests are not possible. Consolidations rely on the exit queue as well, so this feature is currently untestable. We propose a few different plan options that the public,client teams,application developers,staking pools should consider. We would like to hear the feedback from everyone. Plan A: Plan for the future new testnet We are currently investigating the possibility of creating a new testnet, which will be more suitable for validators. This network could be big enough to enable consolidations but would not be too large so as to be expensive for each client team to run.

3/11/2025
Read more from ethpandaops
Published on HackMD