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>
56 lines
2.8 KiB
Markdown
56 lines
2.8 KiB
Markdown
---
|
|
kind: CONCEPT
|
|
slug: adapter-outbound-persistence-jpa-c13
|
|
title: 이미 소비한 시간을 트랜잭션 계층이 다시 주지 않는다
|
|
topic: state-machines-and-ownership
|
|
project: clean-architecture-backend-template
|
|
status: 게시 전
|
|
sourceRevision: 21234e38cdb9a926cbc92bb97a2aee2e4a7d2916
|
|
rootTreeNode: concept:adapter-outbound-persistence-jpa-c13
|
|
evidenceCapturedOn: 2026-09-01
|
|
assets:
|
|
- key: adapter-outbound-persistence-jpa-c13
|
|
file: ../../../final/evidence/rendered/adapter-outbound-persistence-jpa-c13.svg
|
|
evidence:
|
|
- ../../../final/evidence/raw/adapter-outbound-persistence-jpa-c13.txt
|
|
source:
|
|
- 원본 분석 절은 analysis/05-adapter-outbound-persistence-jpa.md#L879 이다.
|
|
module: adapter-outbound-persistence-jpa
|
|
---
|
|
|
|
# 이미 소비한 시간을 트랜잭션 계층이 다시 주지 않는다
|
|
|
|
application policy path는 timeout을 `TransactionDefinition.setTimeout()` 하나로 끝내지 않는다. CallBudget admission이 connection pool을 빌리기 전부터 시작한다.
|
|
|
|
## 본문
|
|
|
|
<!-- body:start -->
|
|
|
|
application policy path는 timeout을 단순히 `TransactionDefinition.setTimeout()` 하나로 끝내지 않는다. `ca-skeleton.jpa.transaction` settings는 transaction/resource-budget defaults를 가진다.
|
|
|
|
- duration positive
|
|
- duration <= 1 day
|
|
- retry max attempts 1..5
|
|
- statement timeout <= transaction timeout
|
|
- lock timeout < statement timeout
|
|
- completion/acquisition/action margin hierarchy
|
|
|
|
즉 runtime에서 무한 retry나 무한 transaction timeout을 property 하나로 열 수 없게 hard cap을 둔다.
|
|
|
|
## RetryProfile 참조 위치
|
|
|
|
:::evidence key="adapter-outbound-persistence-jpa-c13" alt="코드베이스에서 RetryProfile 를 검색한 출력 14줄. 이 기록이 세는 참조가 그 출력에 그대로 보인다." caption="RetryProfile 코드베이스 검색 — 14줄 · exit 0" zoom="true"
|
|
:::
|
|
|
|
이 점은 API `RetryProfile.maxAttempts`가 upper bound를 갖지 않는 것과 대비된다. canonical application path는 실제 deployment settings에서 최대 5회를 강제한다.
|
|
|
|
## 풀에서 기다린 시간이 다시 주어지지 않는다
|
|
|
|
CallBudget admission은 connection pool을 빌리기 **전**부터 시작한다. transaction을 열 가치가 있으려면 남은 budget이 최소한을 감당해야 하고, begin 후에는 실제 남은 budget으로 Spring whole-transaction timeout, statement timeout, lock timeout, idle-in-transaction timeout을 정한다. 따라서 pool에서 오래 기다린 요청이 "원래 5초 timeout이었으니 DB에서 다시 5초"를 받지 않는다.
|
|
|
|
## backoff도 budget을 본다
|
|
|
|
canonical path의 retry backoff도 CallBudget-aware다. 다음 attempt를 시작하기 전에 jitter delay, 다음 acquisition reserve, 다음 최소 transaction/action margin을 모두 감당할 수 있는지 확인한다.
|
|
|
|
<!-- body:end -->
|