amesgen

This is a good point, this is definitely suboptimal and should be improved. It is actually somewhat documented here: https://github.com/http4s/http4s-jdk-http-client/blob/b360b86d31cf6e2617737e7edbeaca9f7a5ea027/docs/README.md?plain=1#L178-L180 Not sure why I thought this behavior is reasonable :sweat_smile:

Thanks for the report, I can reproduce this. This should definitely be fixed, but it should always be possible to work around this by consuming the response body, right?

I bisected this, which yields > 992bce4af032f0c07196acdc06e38dd6d0cc10c0 is the first bad commit ie #332. That is rather surprising to me. I wondered whether any changes in the generators could be...

Reverting the changes to `ouroboros-consensus/src/ouroboros-consensus/Ouroboros/Consensus/Storage/ImmutableDB/Impl/Index/Cache.hs` in #332 on top of 992bce4af032f0c07196acdc06e38dd6d0cc10c0 fixes the test failure, but the test is still failing when applying the same patch on top of current...

Rendered pdf for this PR: https://ci.iog.io/job/IntersectMBO-ouroboros-consensus/pullrequest-982/x86_64-linux.native.consensus-pdfs/latest/download/1/consensus-spec.pdf

Ah, the only thing was that the commits aren't signed (which is required). I rebased and squashed, should now enter the merge queue once CI passed :+1:

Ah, and the new golden files have to be checked in (that's why CI is red ATM). You can generate them locally via `cabal build all` and then `cabal test...

FTR: With #1173 merged, db-synthesizer will "understand" Conway by default.

> I'm surprised that it "only" took 70 seconds to write out the snapshot file, since I recall reports that the node was hampered for several minutes when writing a...

> The other idea is my "pulser" in the previous comment. [...] The wrinkle here is that there's no fundamental way to anticipate how many total tokens. What about separating...