Repository navigation
docs: correct the S3 flush cadence in README (5 minutes → 60s) - #140
Merged
Merged
Conversation
README said logs flush to S3 "every 5 minutes". `_FLUSH_INTERVAL` defaults to 60 seconds and no manifest overrides it, so 60s is what actually runs in both deployments. LOGGING.md already said 60s, so the two docs contradicted each other and the wrong one was in the more-read file. The figure matters for log analysis: it sets how stale the freshest S3 data can be, and a 5x overstatement makes a gap look like missing data rather than a buffer not yet flushed. Names the env var so the number is traceable rather than another bare literal, and records the shutdown flush — relevant when deciding whether a restart is safe mid-session.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
README said logs flush to S3 "every 5 minutes".
_FLUSH_INTERVALdefaults to 60 seconds (llm_proxy.py:30) and no manifest overrides it, so 60s is what actually runs in both deployments.LOGGING.mdalready said 60s in two places, so the two docs contradicted each other — and the wrong one was in the more-read file.Why it is worth fixing rather than leaving: the figure sets how stale the freshest S3 data can be, so a 5x overstatement makes a gap look like missing data rather than a buffer that has not flushed yet. That is a live concern for anyone doing the log analysis this repo is built around.
Also names the env var, so the number is traceable instead of another bare literal (same failure mode as the 300s in #138), and records the shutdown flush —
lifespancalls_flush_to_s3()on exit, which is what makes a graceful restart safe mid-session.Docs-only;
python test_logging.pyexits 0.