Ethereum Core Devs Meeting 72 Agenda
Souptacular opened this issue · 15 comments
Ethereum Core Devs Meeting 72 Agenda
- Meeting Date/Time: Friday 4 October 2019, 14:00 UTC
- Meeting Duration 1.5 hours
- YouTube Live Stream
- Istanbul Meta EIP
- Istanbul EIP Implementation Tracker by @holiman
Agenda
- Ropsten Istanbul fork re-cap
- Reminder: "Ethereum Roadmap 2020: A Community Discussion" @ Devcon5
- Ice Age
- Testing updates
- Review previous decisions made and action items
- Client Updates (only if they are posted in the comments below)
a) Geth
b) Parity Ethereum
c) Aleth/eth
d) Trinity/PyEVM
e) EthereumJS
f) EthereumJ/Harmony
g) Besu
h) Turbo Geth
i) Nimbus
m) Nethermind - EWASM & Research Updates (only if they are posted in the comments below)
Ropsten Istanbul fork went "south" as we kind of expected. Most people didn't upgrade, big miners are on the non-updated chain, causing all kinds of nasty sync issues (hard to find good peers, hard to pick good peers from many bad ones, heaviest chain is bad, etc).
Our current network protocol is unable to split the peers into forking and non-forking sets, so the mess will only get sorted out once enough people update and the Istanbul chain beats the old one.
A few months ago we've proposed a forkid EIP to handle this exact scenario, whereby a node would not only advertise it's genesis hash (useless now), rather a combination of the genesis + all the forks afterwards. Such a mechanism would allow two nodes to cross validate each other and disconnect cleanly if on different configured forks, without wasting each other's time. https://eips.ethereum.org/EIPS/eip-2124
I've opened a PR to integrate this feature into Geth (ethereum/go-ethereum#20140), rolling out an eth/64 protocol update that just replaces the genesis hash field in the handshake with the fork id as specced above (current eth/63 would function as is for the foreseeable future). Anyone any objections against this being rolled out?
We should add an item to investigate when the next ice age is and if we need an EIP in Constantinople to delay it.
You mean Berlin?
@karalabe I think block times may increase before we do Berlin. However, now that we are going with an EIP centric release process we could just use the ice age EIP as our first one to use the new method.
@Souptacular so, you mean Istanbul, then?
@timbeiko it could be istanbul or a single purpose EIP centric fork after. Depends on when the ice age happens (I won't be on the call this Friday, but wanted it on the agenda).
Got it. Your first post said Constantinople, which I think is what caused the confusion @Souptacular 😛
Ohh lol. My bad.
However, now that we are going with an EIP centric release process
I wish Istanbul would be the first one of that kind.
The Ethereum 1.0/1.x Consensus test suite - maintained by @winsvega, @holiman, @shemnon and others - now has a release v7.0.0-beta.1 out with substantial Istanbul support.
This can now serve as a stable and comparable basis for clients to be run against the different Istanbul test types (BlockchainTests, StateTests, TransactionTests).
@timbeiko @Souptacular
5. Review previous decisions made and action items
- Call 71 in the agenda - Link not directed to correct page.
Correct link -
https://github.com/ethereum/pm/blob/1d012e2942234d6c035a95f8cd68b7f450b97de8/All%20Core%20Devs%20Meetings/Meeting%2071.md
Ropsten Istanbul with Geth can be stabilized with the --whitelist=6485846=0x43f0cd1e5b1f9c4d5cda26c240b59ee4f1b510d0a185aa8fd476d091b0097a80 flag.
I'm not sure if there's a more appropriate place to ask it. How can I follow Ropsten "right" chain?
Sooo... call tomorrow?