fix(otel): sync OTEL_METRICS_KNOWN_FIELD_LIST with emitted keys (#1723)

[prabhaks]

What

Fixes #1723.

compute_series_hash in src/otel/metrics.rs treats every top-level data-point key that is not in OTEL_METRICS_KNOWN_FIELD_LIST as a series label. The list had drifted from the keys the flatteners actually emit, in both directions, corrupting physical-series identity for metrics.

Emitted but missing from the list (series fragmentation)

These are emitted as top-level keys but were absent from the list, so their per-sample values were hashed into __series_hash, splitting one physical series into many:

• min, max — histogram per-sample statistics (flatten_histogram)
• data_point_quantile_values — summary quantile array (flatten_summary)

Listed but never emitted top-level (dead entries, with collision risk)

• value, quantile — only ever emitted nested inside the data_point_quantile_values objects, never as top-level keys. Because they were listed, a user attribute literally named value or quantile was silently excluded from series identity, merging distinct series. value is a common label name, so this was the higher-risk one.
• data_point_quantile_values_quantile, data_point_quantile_values_value — never emitted anywhere. Harmless dead entries, removed for correctness.

Change

• Add min, max, data_point_quantile_values to the list.
• Remove value, quantile, data_point_quantile_values_quantile, data_point_quantile_values_value.
• List length 37 -> 36.
• Tests: differing min/max (histogram) and data_point_quantile_values (summary) no longer change compute_series_hash; a data-point attribute named value or quantile now does; plus a guard asserting the list contents and length.

Verification

• cargo test --lib otel::metrics: 10 passed, 0 failed.
• cargo clippy --no-deps -- -D warnings: clean.
• Cross-checked the full set of top-level keys emitted by every flattener (number/gauge/sum/histogram/exp-histogram/summary, exemplars, flags, aggregation temporality, metric-level and resource/scope envelope fields) against the updated list: they now correspond 1:1, with no missing keys and no dead entries.

Note

min/max are emitted with bare (un-prefixed) names, unlike the other data_point_* stats. This PR only syncs the known-field list with what is currently emitted. Renaming them to data_point_min/data_point_max for consistency and to remove the (smaller) collision surface would be a separate, output-schema-changing follow-up worth discussing.

Summary by CodeRabbit

• Bug Fixes
  • Improved metric series identification for histogram and summary data.
  • Histogram minimum and maximum values, along with summary quantile values, are now handled correctly without creating unintended series.
  • Metric attributes named value or quantile are now preserved correctly when used as labels.
  • Added validation to ensure metric field handling remains consistent.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro

Run ID: 805279c0-d761-44e0-8606-2151eb40e84d

📥 Commits

Reviewing files that changed from the base of the PR and between 5acfcb6 and 5c90805.

📒 Files selected for processing (1)

• src/otel/metrics.rs

---

Walkthrough

The OTEL known-field list was synchronized with emitted metric keys. Histogram min/max and summary quantile values are excluded from series identity, while top-level value and quantile attributes remain series labels. Tests cover the updated behavior.

Changes

OTEL series identity

Layer / File(s)	Summary
Align known metric fields src/otel/metrics.rs	The known-field list length and entries now match emitted top-level metric fields, including histogram min/max and summary data_point_quantile_values.
Validate series-hash classification src/otel/metrics.rs	Tests verify sample fields do not alter series hashes, top-level value and quantile attributes do alter them, and the field list matches emitted keys.

Estimated code review effort: 2 (Simple) | ~10 minutes

Poem

I’m a rabbit with a tidy hash,
Sorting fields in a metrics stash.
Min and max now hop away,
Labels keep their names in play.
Quantiles bloom, collisions dash!

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title directly matches the main fix: syncing OTEL_METRICS_KNOWN_FIELD_LIST with emitted metric keys.
Description check	✅ Passed	The PR description covers the bug, fix strategy, key changes, and verification, with only the optional checklist omitted.
Linked Issues check	✅ Passed	The changes and tests align with #1723 by adding/removing the specified keys, updating the length, and covering the required hash behavior.
Out of Scope Changes check	✅ Passed	The patch stays focused on syncing the known-field list and related tests, with no unrelated code changes evident.
Docstring Coverage	✅ Passed	Docstring coverage is 83.33% which is sufficient. The required threshold is 80.00%.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}