Kai Mast

Would it not be better to add a separate feature for this? Or do you think the performance overhead for test networks that do not require timing information is minimal?

It might be good to postpone this until https://github.com/ProvableHQ/snarkOS/pull/3670 is done. There are some interesting subtleties to consider, such as differentiating between malicious nodes (sending invalid blocks) and other faulty/slow...

> Would the `value_to_json` and `mapping_to_json` functions be more useful in snarkVM? Yeah, that would make sense if you think there are other use cases for it. Maybe we should...

> However, I think the old format is potentially useful if people are manually executing transactions via `snarkos developer execute`, as the old output format is directly usable as inputs...

@vicsn this issue is tagged for Q4, but I am not sure what we decided on with regard to changing the API behavior. I think it should change, but we...

> If stake is spread among the validators unevenly, and a rich node is offline, then we might want to fetch from all of the remaining validators. Maybe the confusion...

I see similar issues when calling `.ci/generate_ledger.sh 40 250 1`. Here's the log of two validators in case it helps. Validator 0 does not encounter the issue, but validator 30...

It looks like they are still run in the `ledger-with-valid-solutions` job, e.g., [here](https://app.circleci.com/pipelines/github/ProvableHQ/snarkVM/15589/workflows/aee81965-5ef7-4752-9dd5-b083b1414fa8/jobs/775066).

In my opinion, the cleanest option, aside from removing the `test` feature altogether, is to have dedicated integration tests that require the `test` feature and other ones that conflict with...

The same should be done for consensus versions: ``` --- failure enum_variant_added: enum variant added on exhaustive enum --- Description: A publicly-visible enum without #[non_exhaustive] has a new variant. ref:...