[RCA] Create a reviewable evidence-selection plan
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
- 难度
- 5/5
- 预计耗时
- 一周以上
- 新手友好度
- 35/100
- Issue 类型
- 功能
- 描述清晰度
- 基本清楚
- 活跃度
- 活跃
- 领域
- cli, observability
调研方向
Start by checking the prerequisite issue #14223 and the existing bydbctl command structure under bydbctl/internal/cmd; the issue says the final test paths depend on the prerequisite seam. Implement and test the proposed selection operation through its real CLI entry point, covering budget refusal, pinned evidence, exact ledger reasons, coverage, unchanged raw hashes, and output conflicts. Done means the acceptance scenarios pass with focused behavioral tests and documented fixture provenance.
由索引模型根据 Issue 内容生成。
描述
Umbrella: #14222
Blocked by: #14223
Summary
Create a reviewable evidence-selection plan. Deliver one usable operation, not a framework-only foundation.
Boundary
Proposed activation caller: bydbctl rca curate --workspace … --selection …. Command names are proposed, not existing APIs.
Reporter collection runs with read-only source access; offline steps use local files only.
No source storage/wire-format changes are requested. The incident format is separately versioned.
Requirements
R1. Generate a candidate selection by incident interval, components, sources and byte budget; never mutate raw evidence or claim the selection is public-safe.
R2. Keep a selection ledger naming every included/excluded artifact, reason and coverage consequence. An over-budget selection is refused with the exact excess and an explanation; it is never silently reduced.
R3. Apply explicit reporter inclusion/exclusion rules and pins for contradictory evidence and control samples. No automatic relevance ranking. If the resulting selection exceeds the budget, refuse it with the excess and require a revised selection; never silently discard pinned evidence.
R4. Show size and source/time/component coverage before and after selection, including omitted crash/profile artifacts. Persist the approved selection separately from raw data.
Acceptance criteria
- A fixed inventory contains a small error log, a healthy control log, a contradictory event and a large profile. A budget too small for pinned evidence fails with the exact excess, not silent removal.
- Re-run with sufficient budget and check the exact selected IDs and ledger reasons; raw hashes remain unchanged.
- Use hand-authored or independently captured fixtures with documented provenance; do not generate expected values using the code under test.
- Add focused behavioral tests before implementation and record the actual RED command/failure, then GREEN results.
- Keep existing CLI/API behavior compatible. Never overwrite unrelated user files; output conflicts must be refused.
Scope and readiness audit
- Boundary and behavior: the single operation above; independently observable output and refusal behavior.
- Production activation: the named CLI or skill entry point above, included in the same PR.
- Packages/callers: workspace inventory and artifact metadata; local deterministic selection command.
- Dependencies: #14223
- Classification: planned blocked leaf; revalidate prerequisite seams after merge.
- RED evidence: not executed during this planning task; no claim of passing or failing implementation tests.
- End-to-end proof: the real command/skill scenarios above, not source-text assertions.
- Formats/lifecycle/fixtures: private workspace or public archive only; artifact-specific fixture family above.
- Focused tests: CLI leaves should add
TestRCA…coverage underbydbctl/internal/cmdand/or the introduced collector package;
rungo test ./bydbctl/internal/cmd -run TestRCA -count=1plus the new package tests. Skill leaves must add a replayable evaluation harness
with an exact command documented in the PR. Final test names/paths are fixed when the prerequisite seam merges. - Resource/concurrency checks: cancellation, bounded reads/output and cleanup on failure; use repository panic-aware goroutine wrappers if concurrency is introduced.
- Decision: one proposed PR. If the named behavior requires an unimplemented server API or another lifecycle change, stop and split that prerequisite.
Required-only scope audit
Reporter-controlled reduction is required to make large collections reviewable; relevance ranking is not.
Removed/deferred: Automatic relevance scoring, AI-selected publication, automatic incident detection and report dashboards.
Out of scope
LLM judgment as a safety gate, redaction, archive publication, auto-upload.
- 主要语言
- Java
- 星标
- 25k
- 派生
- 6.6k
- 平均合并
- 12 小时 1 分钟
- 30 天内合并 PR
- 43
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue,再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库,在一个分支上完成修改。
- 提交 Pull Request,并在描述里引用这个 Issue 编号。
apache/skywalking 的其他 Issue
-
bug
难度 2/5 1-3 小时 新手友好度 78/100
apache/skywalking#14239 ·
维护者通常 1 天内回复
-
AI Assistant database enhancement
难度 5/5 一周以上 新手友好度 24/100
apache/skywalking#14237 ·
维护者通常 1 天内回复
-
database enhancement
难度 5/5 一周以上 新手友好度 25/100
apache/skywalking#14236 ·
维护者通常 1 天内回复
-
AI Assistant database enhancement
难度 5/5 一周以上 新手友好度 25/100
apache/skywalking#14235 ·
维护者通常 1 天内回复
-
database enhancement
难度 5/5 一周以上 新手友好度 30/100
apache/skywalking#14234 ·
维护者通常 1 天内回复
查看 apache/skywalking 的全部 Issue
相似的 Issue
-
Re-executed query with IN condition on an index column returns deleted rows (stale IndexCursor)可能已有人在做 @dbusche 今天认领。 未关闭
难度 2/5 1-3 小时 新手友好度 82/100
h2database/h2database#4580 ·
维护者通常 1 天内回复
-
vips-drift
难度 2/5 1-3 小时 新手友好度 74/100
Stirling-Tools/JPDFium#300 ·
维护者通常 1 天内回复
-
hermes-queue
难度 1/5 1-3 小时 新手友好度 85/100
chrisalunlloyd2-sudo/mindpalace#403 ·
维护者通常 1 天内回复
-
Guidelines: document how to build a use case against a local, unpublished Flow/browserless-test change可能已有人在做 @totally-not-ai 今天认领。 未关闭
难度 1/5 1-3 小时 新手友好度 88/100
vaadin/use-cases#452 · 1 条评论 ·
维护者通常 1 天内回复
-
area/storage area/storage/sql
难度 2/5 1-3 小时 新手友好度 78/100
Apicurio/apicurio-registry#11159 ·
维护者通常 1 天内回复