Just bump the value because predicting it is proving to be difficult and time consuming

Use a debug RPC. Waiting for getavalanchepeerinfo is not sufficient here because its possible for this value to be updated before the contender is added to the cache.

• move lastfailure to top of build_dir
• ignore OSError exceptions on windows (didn't use contextlib.suppress because I don't think we should ignore it on Linux)

In D17330#393720, @PiRK wrote:

Benchmark on my machine:

DEFAULT_JOBS =  (multiprocessing.cpu_count() // 3) + 1
ALL                                        | ✓ Passed  | 1603 s (accumulated)
Runtime: 179 s


DEFAULT_JOBS = multiprocessing.cpu_count()
ALL                                        | ✓ Passed  | 2187 s (accumulated)
Runtime: 95 s

DEFAULT_JOBS = 2 * multiprocessing.cpu_count()
ALL                                        | ✖ Failed  | 2583 s (accumulated)
Runtime: 68 s

With one port issue in the 2 * cpu_count attempt.

RuntimeError: Could not find available PortName.P2P port after 5 attempts.

feedback

fix potential deadlock

cleanup shebangs to make lint happy

@bot build-tsan

Better approach

@bot build-tsan

Added back an invalid case. It is incomplete since the local winner does not actually trigger an ACCEPT vote, but this should make it more obvious where that test is heading.

rebase on D17144 to fix early convergence issue

• set mock time and give some comments why we do this
• clarify a bunch of comments in the goofy loop

Split up into separate diffs:
D17027
D17075
D17117
D17121