Nik Bougalis

This was merged as 92d35e54c7de6bbe44ff6c7c52cc0765b3f78258

I don't know if we can just remove the deprecated field; someone might still be relying on it. Do we know how long it's been marked DEPRECATED for? @mDuo13, any...

> Thanks for the review, Nik. I will get on to these suggestions when back at work Thank me _after_ I'm done reviewing. This was just the opening salvo. Enjoy...

This looks interesting and it's something to consider; I like the performance results. But the reality is that the existing codebase is heavily invested in `boost::asio` and moving to a...

The reporting from the RPC should certainly be improved, and a new command showing the status of connections that were recently attempted or are in progress would be good.

So having looked at your proposed changes, I'm not sure I'm sold. I dislike the need for a `vector` to hold the `thread` instances. I suspect that simply changing the...

> One person's cake is another person's poison. Indeed, another approach would be to see if a `yield()` instead of a `sleep_for()` returns the performance. But I _like_ the vector...

@seelabs writes: >This is a large behavior change that probably should get it's own patch and not be folded into an unrelated change. There are times where we purposely crash...

The issue that @scottschurr encountered had to do with timing, but it did lead me to discover an implementation issue that could cause a worker thread to sleep despite the...

> With the most recent code I consistently get the following failure on the JobQueue unit test: > > ``` > $ build/rippled --unittest=JobQueue > ripple.core.JobQueue > Assertion failed: (!stopped_),...