fix: preserve per-series semantics for raw window queries - #529
Merged
Merged
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Before this PR
A bare per-series window query such as
sum_over_time(m[1m])could bind pooled raw state. With api=40 and worker=200, that state emits one value 240 instead of two labeled rows, violating Planner'sReduction::PerEntitycontract.After this PR
Backend-local planning retains native execution when a selected per-entity leaf cannot preserve source entities. An explicit direct SUM over additive Sum/Count state authorizes pooling, so global/grouped sums remain eligible for warm execution; permission does not cross binary arithmetic. Other independent roots and keyed TopK state remain available.
The compatibility demo now uses an explicit global sum and verifies exact forwarding for its bare quantile. Sharing/cost fixtures use semantically valid additive reductions. This change protects newly compiled plans; physical Sum state alone cannot identify whether an older artifact intended per-series or explicit aggregate output.
Validation: 663 control-plane tests, the concrete two-series worker regression, and all four compatibility process tests pass on current main with Planner abb2f20. The process fallback mock verifies forwarding, not numerical Prometheus agreement.