Log refresh-cycle steps and snapshot-rebuild phases#583
Merged
Conversation
The entity-stats snapshot rebuild is silent until it completes, so a rebuild that never finishes (prod is stuck serving a stale snapshot) is undiagnosable from the logs. Add INFO logs that time each refresh-cycle step (charts prewarm, stats summary, leaderboard summary, then the snapshot rebuild) and each phase of the rebuild itself (loading run rows from the DB, walking the blob files, finalizing the Codex Elo fits). Whichever "done" log never arrives pinpoints where a stalled rebuild is stuck. Logging only — no behavior change.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The entity-stats snapshot rebuild is silent until it completes, so a rebuild that never finishes (prod is stuck serving snapshot v10, and no error is logged) can't be diagnosed from the logs.
Adds INFO logging only — no behavior change:
_run_refresh_cycle): times each step —charts prewarm done at Xs,stats summary done at Xs,leaderboard summary done at Xs, entering snapshot rebuild._build_cache_data): logs each phase —starting (loading run rows from the DB),loaded N run rows in Xs, now walking blob files,blob walk done in Xs (N entities), finalizing. Plus the existingsnapshot rebuilt: N entities across M runson completion.After deploying this, the logs will show exactly where the stalled rebuild stops — whichever phase's "done" line never appears is the culprit. My working theory is the
list(coll.find({}))full pull of ~740k run docs crawling under Mongo write-contention (heavy presence traffic), which these logs will confirm or refute.ruff + compile clean.