DongHyeonkaandClaude Opus 5
b2963105a8
The keycloak project ended with four open questions that design could not
settle. A two-VM lab was built to answer them by measurement, and this is
that material: 26 experiments, 125 raw command outputs, 22 browser captures.
Follows the import procedure in README.md.
source/ the originating repository verbatim — 78 documents, 28 SVGs,
8 manifests, plus .source-revision recording the commit
final/ the SSOT
document.md 729 lines written from the 29 experiment documents, not
concatenated: what was predicted, what was measured, and
where the measurement itself was wrong
evidence/raw 125 outputs, flattened to <experiment>__<file> because
the originals collided (01-baseline.txt appeared three
times) and the audit only globs the top level
evidence/meta one per raw file; command and exitCode are null and the
README says why rather than inventing them
evidence/browser 22 captures
assets/ three diagrams through techviz
.techviz/ their VizSpecs
A separate project rather than an addition to keycloak: the B-layer answers
that project's four questions, but the A, C and D layers are about cluster
failure, SSO and operations, and one document.md should hold one subject.
The four question records there can point here through 관계.
Recorded rather than papered over: only three of the 28 diagrams were
remade. The repository forbids hand-drawn SVG and forbids titles inside the
canvas; all 28 originals carry both, so converting them is redrawing, not
reformatting. They stay in source/ and the gap is written into the document.
verify-pipeline.py passes. audit-records.py reports no issues.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
3.1 KiB
kind, slug, title, topic, project, status, sourceRevision, rootTreeNode, evidenceCapturedOn, assets, evidence, source, module
| kind | slug | title | topic | project | status | sourceRevision | rootTreeNode | evidenceCapturedOn | assets | evidence | source | module | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CONCEPT | messaging-core-api-c04 | 거절과 결과 모름을 하나로 합치면 중복 주문이 생긴다 | delivery-and-settlement-models | clean-architecture-backend-template | 게시 전 | 21234e38cdb9a926cbc92bb97a2aee2e4a7d2916 | concept:messaging-core-api-c04 | 2026-09-01 |
|
|
|
messaging-core-api |
거절과 결과 모름을 하나로 합치면 중복 주문이 생긴다
이 leaf의 실질은 여기 있다. 표현할 수 없는 상태를 생성자에서 거절하는 것이 설계의 축이다.
본문
이 leaf의 실질은 여기 있다. 표현할 수 없는 상태를 생성자에서 거절하는 것이 설계의 축이다.
발행 결과 세 상태
:::evidence key="messaging-core-api-c04-diagram" alt="발행 시도에서 CONFIRMED 와 REJECTED 와 AMBIGUOUS 세 갈래가 나온다" caption="발행 결과 세 상태" zoom="false" :::
PublishCompletion은 boolean이 아니라 3상태다. enum javadoc이 왜 셋인지 적는다 — "Collapsing 'the broker refused this' and 'we never learned what the broker did' into one failure is what produces duplicate orders"(publish/PublishCompletion.java:6-8). AMBIGUOUS인 호출자는 같은 messageId로만 재발행할 수 있다.
PublishCompletion 참조 위치
:::evidence key="messaging-core-api-c04" alt="코드베이스에서 PublishCompletion 를 검색한 출력 29줄. 이 기록이 세는 참조가 그 출력에 그대로 보인다." caption="PublishCompletion 코드베이스 검색 — 29줄 · exit 0" zoom="true" :::
생성자가 거절하는 열두 조합
PublishResult 생성자(publish/PublishResult.java:39-101)가 12가지를 거절한다. 11번과 14번에는 코드 주석이 직접 달려 있다. record가 public이고 모든 adapter가 이것을 만들기 때문에 호출부를 믿지 않고 여기서 검증한다는 것도 javadoc에 적혀 있다(PublishResult.java:18-20).
증거가 결론보다 먼저 기록된다
PublishEvidence(publish/PublishEvidence.java)는 queuedLocally, transmission, brokerAccepted, confirmationLevel 넷을 갖고, javadoc이 순서를 못 박는다 — "Evidence is recorded before a completion is chosen, not derived from it. That ordering is what lets an operator answer 'could the broker be holding this message?' from a stored result." TransmissionEvidence가 3상태(NOT_TRANSMITTED / MAY_HAVE_BEEN_TRANSMITTED / TRANSMITTED)인 것이 그 순서를 가능하게 한다.
정산 쪽도 같은 형태다
SettlementResult(settlement/SettlementResult.java:23-36)는 SETTLED인데 !brokerConfirmed이면 거절하고, SETTLED인데 redeliveryPossible이면 거절한다.