Avoid mutating $stdout and $stderr

[karloscodes]

Description

Closes #1948

Uses flush after writing to stdout and stderr so we don't need to use sync=true

Your checklist for this pull request

• I have reviewed the guidelines for contributing to this repository.
• I have added an entry to History.md if this PR fixes a bug or adds a feature. If it doesn't need an entry to HISTORY.md, I have added [changelog skip] or [ci skip] to the pull request title.
• I have added appropriate tests if this PR fixes a bug or adds a feature.
• My pull request is 100 lines added/removed or less so that it can be easily reviewed.
• If this PR doesn't need tests (docs change), I added [ci skip] to the title of the PR.
• If this closes any issues, I have added "Closes #issue" to the PR description or my commit messages.
• I have updated the documentation accordingly.
• All new and existing tests passed, including Rubocop.

[nateberkopec]

May want to rebase against master as Greg just fixed an issue where we were using puts instead of stdout

[karloscodes]

I have tried with different apps locally and on Heroku and everything works fine. Maybe someone has a clear case that should be tested with this change? Ideas?

[karloscodes]

Also, I have mixed feelings now because I introduced some duplication, do you think it's a good idea to encapsulate the new writing logic for the code outside events and error_logger?

[nateberkopec]

Re: correctness, anything from #1906 could be extracted as a test

[nateberkopec]

You can leave the refactor/DRYness for another PR if you want, don't let it block this feature.

[nateberkopec]

Ideas for test re: comments in that thread:

output is no longer being flushed before Process.daemon calls fork

[karloscodes]

I've been checking the current integration tests and it seems that

puma/test/test_integration_single.rb

Line 126 in 441c474

def test_write_to_log

already covers #1906 am I missing something? I thought that maybe the same tests but with workers could be helpful to cover the issue we had with Process.daemon forks, WDYT?

[nateberkopec]

Took another look, I think this is good as written