Andrew Toth

> `generateblock` is thought to be used always with a given set of txs and that may not always be the use case. Maybe would be worth to implement it...

Thanks to @l0rinc for discovering this behaviour, and @mzumsande for the suggested fix.

> is this comment still accurate? https://github.com/bitcoin/bitcoin/blob/master/src/validation.cpp#L2782 Guess it just refers to the size based trigger, not the time based - which is probably fine given its small context. I...

> It's unfortunate to lose this much performance, as suggested before, maybe the change from 24 hours to 1 hour was too big - what if we tried an average...

> Once we get to a state that is ACK-worthy, I'll enable hard assertions everywhere and run an IBD and check if it crashes anywhere. I think for a change...

Rebased after #30849 and #30906. We can now specify the errors that should be thrown for the tests. In follow up PR, we can remove `SetFresh` and combine with `SetDirty(bool...

@l0rinc rebased. Thank you for your reviews! I have tried to address all your comments. There are some older comments that I'm unsure are still relevant. Please let me know...

Closing in favor of https://github.com/bitcoin/bitcoin/pull/33018.

Concept ACK Indeed, the `FlushStateMode::ALWAYS` variant is now confusing. It could mean we want to always write the chainstate to disk, or we want to always erase the dbcache. Splitting...

@paplorinc I don't see anything in this PR that would affect reindexing. This PR seems like a refactor only. Oh, actually it will affect`scantxoutset` and `gettxoutsetinfo` rpcs.