Merge remote-tracking branch 'origin/main' into codex/p0-obs-002-20260715

apps/api/src/jobs/incident_lifecycle_reconciler.py

@@ -23,6 +23,7 @@ import structlog
 from sqlalchemy import text
 
 from src.core.config import settings
+from src.core.context import clear_project_context, set_project_context
 from src.db.base import get_db_context
 from src.utils.timezone import now_taipei
 
@@ -31,6 +32,7 @@ logger = structlog.get_logger(__name__)
 BATCH_LIMIT = 100
 INTERVAL_SECONDS = 1800
 _PROMETHEUS_TIMEOUT_SECONDS = 5.0
+_PROJECT_ID = "awoooi"
 
 
 @dataclass(frozen=True)
@@ -43,20 +45,32 @@ class LifecycleCandidate:
 
 async def run_incident_lifecycle_reconciler_loop() -> None:
     """每 30 分鐘收斂一小批已有完成證據的 stuck incident。"""
-    while True:
-        try:
-            resolved, errors = await reconcile_stuck_incidents()
-            if resolved > 0 or errors > 0:
-                logger.info(
-                    "incident_lifecycle_reconciler_done",
-                    resolved=resolved,
-                    errors=errors,
-                    batch_limit=BATCH_LIMIT,
+    context_tokens = set_project_context(
+        _PROJECT_ID,
+        source="incident_lifecycle_reconciler",
+    )
+    try:
+        while True:
+            try:
+                resolved, errors = await reconcile_stuck_incidents()
+                if resolved > 0 or errors > 0:
+                    logger.info(
+                        "incident_lifecycle_reconciler_done",
+                        project_id=_PROJECT_ID,
+                        resolved=resolved,
+                        errors=errors,
+                        batch_limit=BATCH_LIMIT,
+                    )
+            except Exception as exc:
+                logger.warning(
+                    "incident_lifecycle_reconciler_loop_failed",
+                    project_id=_PROJECT_ID,
+                    error=str(exc),
                 )
-        except Exception as exc:
-            logger.warning("incident_lifecycle_reconciler_loop_failed", error=str(exc))
 
-        await asyncio.sleep(INTERVAL_SECONDS)
+            await asyncio.sleep(INTERVAL_SECONDS)
+    finally:
+        clear_project_context(context_tokens)
 
 
 async def reconcile_stuck_incidents(limit: int = BATCH_LIMIT) -> tuple[int, int]: