Skip to content

Instantly share code, notes, and snippets.

Created September 17, 2020 14:41
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
Star You must be signed in to star a gist
What would you like to do?
2020-09-17 Eth2 Implementer's Call 48
- Last phase of audit started
- Validator core: secrets & keystores
- Attestation/block proposal + Slashing protection
- Gossipsub (1.1)
- Command-line + user’s instructions
- Several libp2p fixes
- async & memory leaks
- peer management
- Gossip 1.1 ready
- Slashing protection
- WIP aggregation through VC/BN split
- DB pruning
- Sync v2
Testing release updates
libp2p specs, modification of msgID in the gossipsub message
Proto: working on automated multinet, and unittest with specific test
Q from terence: merging fork choice spec test
Spadina, eth2.0-deposits-cli and configuration merged.
Client update
Improve sync to be more resilient to bad peers and better distribute on multiple core
Weak subjectivity work
ARM support for BLST
Discv5, fixed DOS vulnerabilities
Started on Discv5.1
Running in memory issues on Medalla
were storing all states in memory, now block processing broken up by checkpoints
can prune state more intelligently.
Allow also networking requests to complete as they were doing synchronous block processing and networking was interrupted during block processing
Regenerating state from checkpoint
Reimplemented fork choice in a separate package, were not using the right (?) balances
Gossipsub 1.1 integration
Fixed DOS vectors brought up by protocol
Core logic more resilient to timeout
BLST integrated in master branch
eth2-API integration. Protobuf done, endpoint to implement.
TODO: fix IPv6 integration and peer scoring
2 new team members
Progress on standard HTTP API
(lost connection)
Code freeze planned for audit on 1st october
Will test weak subjectivity by wiping DB
Refactor on forward and backward sync plan
Freeze on 0.2.x branch, only security updates
Post on weak subjectivity sync:
Reward accounting
Leo (BSC): moving forward on Gossipsub 1.1 testing:
BLS v4 incompatibility:
EIP2333: HKDF_mod_r <- minor change
EIP2386: NCC input validation concerns -
upgrading Medalla
-> testnet launch towards mainnet?
Mainnet launch parameter change
- Reduction of the proportional coef for slashing
- Reduce of inactivity
- Reduce minimal slashing quotient
Will be hardforked into “final parameters 5 months after now.
Get ready for Spadina
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment