Skip to content

Instantly share code, notes, and snippets.

@iikirilov
Last active March 14, 2019 14:13
Show Gist options
  • Save iikirilov/f2552d50a4b3a7a9b878bdef589802af to your computer and use it in GitHub Desktop.
Save iikirilov/f2552d50a4b3a7a9b878bdef589802af to your computer and use it in GitHub Desktop.
Eth 2.0 Implementers call #13
1. Testing Updates
https://github.com/ethereum/eth2.0-test-generators/issues/21
https://github.com/ethereum/eth2.0-test-generators/issues/12
https://github.com/prysmaticlabs/prysm/tree/master/beacon-chain/chaintest
Docker launcher for simulating blockchains - https://github.com/zscole/nonce
and example docker configs for various blockchains (ethereal, bitcoin, eos, miners, …): https://github.com/zscole/blockchain-dockerfiles
2. Client Updates
Pegasys:
- complete 0.4
- lmd ghost?
- bls
Trinity:
https://github.com/ethereum/trinity/pull/309
- almost complete 0.4
- start with devp2p
Prysmatic
- complete 0.4
- found tons of bugs
Nimbus
- targetting 0.3
- simple testnet 8 participants
- implementing libp2p - what language?
Woodstock?
- ssz?
- bls?
- prepping for simulation
Parity
- simple testnet using casperffg with 10 blocks/s
- they sound confused
Yeeth
- complete 0.4
- libp2p swift
- not sure which bls implementation to use
Lighthouse
- building fork-choice test framework
- focus on performance of client - sub 1s block import
- rust developer starting next week?
- rust libp2p != go lib2p2
3. Research (Mike protocol labs along with Raul)
- https://github.com/ethereum/eth2.0-specs/issues/675
- sha256 vs. keccak256 w.r.t. bls (wtf is bls?)
pro: most blockchain projects are on sha256
con: security concerns (mainly academic and not practicle)
- eth 2 light client
- potentially DoS on fork-choice rule
- attacker can have multiple forks and force full-nodes to evaluate all forks
- each fork has a minimum amount of work that needs to be done
- light clients would not be affected
- need to use sha256 or keccak256 so we can expose eth2 block state roots in eth1
- Vitalik
- https://github.com/ethereum/eth2.0-specs/pull/682
Phase 1 - https://github.com/ethereum/eth2.0-specs/issues/686
Phase 2 - https://github.com/ethereum/eth2.0-specs/issues/702
- Handel (https://github.com/ConsenSys/handel) presentation at SBC:
https://www.youtube.com/watch?v=HCd4fF9u664&list=PLxiUDysqkJ9xTGggKpeHwdANjPnM4Fxv3
4. Phase 0 Wire Protocol - https://github.com/ethereum/eth2.0-specs/issues/692
- basically they are talking about how to simplify the wire protocol because there is too much information
5. Serialization
SSZ spec - https://github.com/ethereum/eth2.0-specs/pull/696
- Wire protocol
- consensus objects should be encoded using ssz basically everyone
- how the actual message is encoded is up to anyones opinion
- RLP does not need a schema to decode
- SSZ does not have a self-describing structure
- you can decode specific field in SSZ - you must decode the whole rlp to get a specific field
- SOS: https://gist.github.com/karalabe/3a25832b1413ee98daad9f0c47be3632
- SSS: https://github.com/ethereum/bimini/blob/master/spec.md
Discorvey v5 - https://github.com/ethereum/devp2p/issues/48
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment