Status: OUT
No subsystem at any architectural tier has a viable recovery strategy: storage-layer recovery is paradoxically over-engineered in infrastructure and under-implemented in usage (WAL builds complete but unused sequence/checkpoint machinery, with no safe crash recovery path), while application-layer event sourcing lacks durable persistence and uses conflated ID spaces — recovery fails both where infrastructure exists but is unused and where it was never built.
Two depth-3 conclusions spanning different architectural tiers (storage vs application) both conclude recovery is broken, but for opposite reasons — over-engineering vs under-engineering — revealing that no tier found the right calibration
Depends on (SL): recovery-simultaneously-over-and-under-engineered, event-sourcing-unreliable-in-persistence-and-addressing