George Knee

What's the correct long-term behaviour for this? How about we let the **defund** objective forcefully take over from the **fund** objective? Is there any point in the client in question...

Yeah I think sleeping is not the long term solution. I still think that abandoning the virtual fund objective might be the right thing to do in this case, if...

Discussed in standup today: abandoning the `virtualfund` objective is not a good approach, since validating a proposal to virtually defund requires inspecting **valid vouchers received**, which depends on having the...

We've seen the new behaviour play out on sepolia: ``` t = 0 | Requeue and start using CALLDATA dt = 11s | Publish tx dt = 3m | Requeue...

I believe the `Requeue` method is only ever going to requeue the blocks from a single channel. This is because the batcher now only requeues blocks from channels which didn't...

Sketch of a benchmark here https://github.com/ethereum-optimism/optimism/compare/gk/batcher-requeue-bench?expand=1: ``` go test -benchmem -run=^$ -bench ^BenchmarkChannelManager_Requeue$ github.com/ethereum-optimism/optimism/op-batcher/batcher -v goos: darwin goarch: arm64 pkg: github.com/ethereum-optimism/optimism/op-batcher/batcher BenchmarkChannelManager_Requeue BenchmarkChannelManager_Requeue-14 4 296146114 ns/op 74611654 B/op 32528 allocs/op...

Resolved conflicts over here https://github.com/ethereum-optimism/optimism/pull/12975

@tessr I believe we discussed that rollup stage = 1 might in the end be equivalent to "standard chain". Can you confirm? If so, shall we just point this out...

From call today, this is true, so we can just update the readme.

I think this adds extra weight to the new architecture https://github.com/ethereum-optimism/optimism/pull/12390, where we don't throw away a channel as soon as it is confirmed. To build this feature, we need...