Skip to content

Instantly share code, notes, and snippets.

@mratsim
Created May 28, 2020 14:54
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save mratsim/813c041b58bfc2f2f34722c3cdcffac2 to your computer and use it in GitHub Desktop.
Save mratsim/813c041b58bfc2f2f34722c3cdcffac2 to your computer and use it in GitHub Desktop.
2020-05-28-Eth2-impl-call-40
https://github.com/ethereum/eth2.0-pm/issues/154
Nimbus
- Networking
- Investigating interop issues with lighthouse
- Protobuf parsing
- attestations received before blocks causing gossip issues
- Pushed this morning: multinets script that work with a 50/50 lighthouse/nimbus validator split
- Snappy fixes
- Gossipsub assumes that we ca
- Witti (and Schlesi):
- “make witty” toggle added
- syncing bugs
- investigating the “few peers” issue
- Core
- BLS changes merged in our crypto backend
- Can toggle between v0.11.3 or v0.12.0 for testnets or fuzzing purposes
- Audit:
- We received proposals from auditing firm
and are currently selecting vendors
—————————
Testing and releases
v0.12 released including BLS standard upgrade
Rewards & penalties updated with much more tests
Found bugs in ZRNT as well
Networking changes
Fork choice tests readying up for v0.12.1
No updates on Stethoscope atm
—————————
Testnets
Schlesi testnet couldn’t be relaunched.
Witti launched yesterday night, 3 clients at Genesis
- Teku, Lighthouse and Prysm each with 2 validators
Q: how to launch testnet at any other time than 2 AM: https://github.com/ethereum/eth2.0-specs/issues/1849
—————————
Client updates
Lodestar
Published a new version of BLS library
that uses the new Herumi’s code with WASM support
Updating Gossipsub to v1.1
Nimbus
Teku, Meredith
Releases every 2 weeks
Updated with witty
OpenAPI schema updated
BLS change under review
Can deal with fast sync Eth1 node
ENR
Fixed JVM libp2p issues regarding handling large stream packets
Prysm, Raul
v0.12 complete,
will restart Topa soon
memory improvements experimental features
progress on slasher
Fuzz testing
Quantstamp audit in-progress
Lighthouse
Audit with Trail of bits
2 parts: consensus and networking
networking is delayed
Metrics_logging in lighthouse
Issue with Discv5 using to much CPU, restructuring
WIP: peer reputation system
WIP on Gossip v1.1
Trinity
Py-libp2p
gossipsub
Spec updates todo to reach v0.11.3
Nethermind
- DB, merle trees
- testing validators onboarding
- WIP v0.11.x
- Moving a junior dev full-time for Eth2
- Nethermind Eth1 -> Eth2 interop
- Witty not before end of June
—————————
Research updates
Discovery peers efficiency analysis writeup
See ethresear.ch and hacked
Completed a write-up on Discv5 called "Comparing discovery advertisement features by efficiency: ENR attributes and topic advertisement"
-- https://ethresear.ch/t/comparing-discovery-advertisement-features-by-efficiency-enr-attributes-and-topic-advertisement/7448

Completed a write-up called "Discovery peer advertisement efficiency analysis"
-- https://hackmd.io/@zilm/BJGorvHzL
Transpiling Pyspec to Kotlin code
Testnet analysis
https://txrx-research.github.io/prkl/testnet-analysis.html
https://github.com/TXRX-Research/prkl
Cross shard
https://twitter.com/JonnyRhea/status/1263151048199372801?s=19
Cross-shard Transaction Probabilistic Simulation
-- https://ethresear.ch/t/cross-shard-transaction-probabilistic-simulation/7474
Cross-shard spec “alpha-alpha” will be started in the next coming week
Will:
account abstraction
Alex:
Eth1.x, and need for protocol support for yanking?
Yes, probably, lots of overhead
—> Issue with serializing everything in memory (expensive in EVM) and producing a receipt
Caching: shard committee cache if we have a standard witness format (few epochs long)
The cache would be restricted to block proposers
—————
Networking
Proposal to change Disc v5 wire protocol
(… long discussion …)
PR open, give opinion: https://github.com/ethereum/devp2p/issues/152
ENR
multi-addressing
—————
Spec discussion
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment