feat(cloud-api): add correlated lifecycle logging

apps/cloud-api/cloud_api/app.py

@@ -8,6 +8,7 @@ from dataclasses import dataclass
 from typing import Any
 
 from fastapi import FastAPI, status
+from fastapi import Request
 from fastapi.responses import JSONResponse
 
 from cloud.auth import create_auth_provider
@@ -20,6 +21,13 @@ from cloud.control_config import (
 from cloud.database import CloudDatabase
 from cloud.internal_api.api import create_internal_router
 from cloud.plugins import PluginRegistry
+from cloud.observability import (
+    CORRELATION_HEADER,
+    bind_correlation_id,
+    new_correlation_id,
+    normalize_correlation_id,
+    reset_correlation_id,
+)
 from cloud.pool import DevicePool
 from cloud.scheduler import TaskScheduler
 from cloud.schema import require_current_schema
@@ -136,6 +144,28 @@ def create_app(
 
     app = FastAPI(title="Device Cloud API", lifespan=lifespan)
 
+    @app.middleware("http")
+    async def correlation_logging(request: Request, call_next):
+        correlation_id = normalize_correlation_id(
+            request.headers.get(CORRELATION_HEADER)
+        )
+        correlation_token = bind_correlation_id(correlation_id)
+        try:
+            response = await call_next(request)
+            logger.info(
+                "cloud request completed",
+                extra={
+                    "correlation_id": correlation_id,
+                    "method": request.method,
+                    "path": request.url.path,
+                    "status_code": response.status_code,
+                },
+            )
+            response.headers[CORRELATION_HEADER] = correlation_id
+            return response
+        finally:
+            reset_correlation_id(correlation_token)
+
     @app.get("/health/live")
     def health_live() -> dict[str, str]:
         return {"status": "live"}
@@ -198,6 +228,7 @@ async def _run_scheduler_loop(
     interval_seconds: float,
 ) -> None:
     while not stop.is_set():
+        correlation_token = bind_correlation_id(new_correlation_id())
         try:
             services.scheduler.assign()
         except Exception:
@@ -205,6 +236,8 @@ async def _run_scheduler_loop(
                 "cloud lifecycle iteration failed",
                 extra={"worker": "scheduler"},
             )
+        finally:
+            reset_correlation_id(correlation_token)
         if await _wait_for_stop(stop, interval_seconds):
             return
 
@@ -217,6 +250,7 @@ async def _run_lease_reaper_loop(
     max_attempts: int,
 ) -> None:
     while not stop.is_set():
+        correlation_token = bind_correlation_id(new_correlation_id())
         try:
             services.repository.reap_expired_leases(
                 now=utc_now(),
@@ -227,6 +261,8 @@ async def _run_lease_reaper_loop(
                 "cloud lifecycle iteration failed",
                 extra={"worker": "lease_reaper"},
             )
+        finally:
+            reset_correlation_id(correlation_token)
         if await _wait_for_stop(stop, interval_seconds):
             return