getsentry / getsentry/sentry

Metric alert edit view: make the evaluation preview faithful to the equation (or clearly label its fidelity)

Open
#118,216 4 comments 1 reaction 1 assignee View on GitHub

@narsaynorath is already working on this.

Since Jul 9, 2026.

2. High Bug Product Area: Performance
Dominant language
Python
Stars
44.8k
Forks
4.9k
Avg merge
21h 23m
Merged PRs (30d)
607

Description

Problem Statement

The metric edit view includes an evaluation/preview graph intended to show how a metric will behave before it is enabled. Users rely on this to judge whether the metric will be noisy. Today the preview does not reliably reflect the exact equation entered. For example, an equation that is mathematically constant (A / A * 100, which should always equal 100) still shows dips in the preview, and editing the equation re-renders in ways users cannot fully reconcile with what they typed.

Why the current behavior is insufficient
Judging noisiness before enabling a metric is the entire purpose of the preview. If the preview diverges from the equation (or from the fidelity of the live alert window) without explanation, users cannot trust it and cannot safely decide whether to turn the metric on.

Solution Brainstorm

Evaluate the edit-view preview at the same fidelity as the live alert window, OR clearly label when the preview uses a lower-fidelity / longer (e.g., 7-day) query tier so expected divergence is explicit. Ensure the preview recomputes deterministically from the exact equation entered. (Note: some of the constant-equation dips are understood to be a lower-fidelity query-tier artifact of the preview window; this request is about making that behavior either faithful or transparent in the UI.)

Product Area

Alerts

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.