- 결정 17: 자체 JDK HTTP gateway·SigV4 서명기 삭제, SES를 AWS SDK v2 `sesv2`로(기존 설정 계약 유지, SDK 재시도 끔, 기본 자격 증명 체인 끔, endpoint 라우팅 검사 유지) — 로컬 가짜 서버로 실제 SDK client 계약 테스트 - S2-#1b: notification persistence를 `adapter:outbound:persistence-notification`으로(migration V1–V10 바이트·위치 동일, persistence-jpa → application-notification edge 제거), composition이 새 leaf를 가져옴 - 레지스트리 2개 삭제, provider 상태 전이표 하나, `ProviderCapabilities.dispatchOnly(...)`, dispatch record 전이 메서드, 미조립 회전 경로·중복 제출 진입점·테스트 전용 생성자 삭제 - 통합: at-rest 계약 테스트가 새 leaf 경로를 보고 `src/main/java`만 훑게(ORCH-new-2), app-bootstrap 알림 선택 주석(P23-new-3) 검증: ./gradlew check architectureCheck verifyDependencyLocks verifyPublicPathSnapshot :app-bootstrap:verifyEnvKeys configContractCheck :app-bootstrap:integrationTest :sample-portfolio:integrationTest persistence-jpa jpaPlatformContractTest jpaPlatformMigrationTest --continue 통과(704 task) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
47 lines
2.0 KiB
Markdown
47 lines
2.0 KiB
Markdown
# Operations
|
|
|
|
## Runtime shape
|
|
|
|
```text
|
|
durable queue (PostgreSQL, FOR UPDATE SKIP LOCKED)
|
|
→ expiry check
|
|
→ suppression and eligibility re-check
|
|
→ provider health gate
|
|
→ rate limiter
|
|
→ concurrency limiter
|
|
→ provider adapter
|
|
```
|
|
|
|
Provider calls run outside every database transaction. The attempt row is committed first, so after a
|
|
crash the row is either absent (nothing was sent) or present in `DISPATCHING` (reconciliation has
|
|
something to ask about).
|
|
|
|
## Guards that exist for specific incidents
|
|
|
|
| Guard | The incident it prevents |
|
|
|---|---|
|
|
| Credential failure opens the provider route | One expired key multiplied by a queue becomes a self-inflicted outage |
|
|
| Retry budget per provider profile | A provider outage turning every queued notification into its own retry loop |
|
|
| Ambiguous attempts block automatic fallback | A push whose response was lost arriving alongside the "just in case" SMS |
|
|
| Permits released during backoff | A slow provider pinning the whole concurrency budget on work that is only waiting |
|
|
| Fail-fast intake on capacity | An unbounded in-memory queue absorbing a burst it cannot survive |
|
|
|
|
## Scheduling
|
|
|
|
`scheduleAt` activates the job, `notBefore` is the earliest permitted provider submission, and
|
|
`expiresAt` blocks new attempts, retries and fallbacks. Suppression and expiry are re-checked
|
|
immediately before dispatch, because a scheduled notification can sit in the queue for hours and the
|
|
user may have opted out in the meantime.
|
|
|
|
## Redrive
|
|
|
|
A redrive preserves `NotificationId` and `RecipientDeliveryId`, creates a new `DeliveryAttemptId`, and
|
|
reuses the pinned template version and rendered digest. Sending different content is a new
|
|
notification, not a redrive. Redriving an ambiguous attempt requires explicit duplicate-risk approval,
|
|
because the platform genuinely cannot tell whether the first submission reached the user.
|
|
|
|
## Actuator surface
|
|
|
|
Provider runtime states and generations, queue depth and age, callback and reconciliation health.
|
|
Never addresses, never credentials.
|