Add Flush Interval to default Buffered Logger

[jrmccluskey]

Adds a default flush interval of 15 seconds to the BufferedLogger type to avoid logs being held until the job ends.

---

Thank you for your contribution! Follow this checklist to help us incorporate your contribution quickly and easily:

• Mention the appropriate issue in your description (for example: addresses #123), if applicable. This will automatically add a link to the pull request in the issue. If you would like the issue to automatically close on merging the pull request, comment fixes #<ISSUE NUMBER> instead.
• Update CHANGES.md with noteworthy changes.
• If this contribution is large, please file an Apache Individual Contributor License Agreement.

See the Contributor Guide for more tips on how to make review process smoother.

To check the build health, please visit https://github.com/apache/beam/blob/master/.test-infra/BUILD_STATUS.md

GitHub Actions Tests Status (on master branch)

See CI.md for more information about GitHub Actions CI or the workflows README to see a list of phrases to trigger workflows.

[github-actions]

Assigning reviewers. If you would like to opt out of this review, comment assign to next reviewer:

R: @shunping for label python.

Available commands:

• stop reviewer notifications - opt out of the automated review tooling
• remind me after tests pass - tag the comment author after tests pass
• waiting on author - shift the attention set back to the author (any comment or push by the author will return the attention set to the reviewers)

The PR bot will only process comments in the main thread (not review comments).

[tvalentyn]

how about we make flush interval the default behavior, unless disabled, as opposed to having to instantiate a special instance of BufLogger?

[jrmccluskey]

Having looked at it we do use both instantiation methods in different places. We could clean things up and unify those paths if we really, really felt the need to but in this context that seems unnecessary. This kind of overloaded constructor pattern isn't atypical in Go, and I think the distinct methods provide a pretty clear indication of what you're getting when you use them (buffering vs buffering with a regular flush interval.)

If there was another code change to make it'd probably be to add some sort of flushing behavior every X logging entries stored, but that may be overcomplicating things in the case of dumping long tracebacks for errors.

[tvalentyn]

My point is less about code style but that we shouldn't be using buffered logger without flushing ever, looks like this is the last place where we do so: https://github.com/search?q=repo%3Aapache%2Fbeam%20NewBufferedLogger&type=code

[jrmccluskey]

Looking at the code more I'm a little surprised that there isn't any sort of flush mechanism beyond the time. I think we wound up dropping the buffered logging into the worker process startup to catch any early errors and it wound up swallowing too much. 10-15 seconds as a default should be fine then

I may try to come back to the buffered logging at some point when I have more time and try to do some other polishing, I have a few ideas as to why outputs from python segfaults get all screwy now.

[tvalentyn]

LGTM, thank you! Please test the change in Dataflow.

Optionally, consider using NewBufferedLogger where the other constructor is used, if it makes sense or keep as is.