fix(ops): harden cold-start schedule recovery

scripts/cron_km_vectorize.py

@@ -18,17 +18,21 @@ import httpx
 async def main() -> int:
     api_base = os.environ.get(
         "INTERNAL_API_URL",
-        "http://awoooi-api.awoooi-prod.svc.cluster.local:8000",
+        "http://awoooi-api-svc.awoooi-prod.svc.cluster.local:8000",
     )
     url = f"{api_base}/api/v1/knowledge/embed-all"
 
-    async with httpx.AsyncClient(timeout=120) as client:
+    async with httpx.AsyncClient(timeout=1800) as client:
         try:
             resp = await client.post(url)
             print(f"embed-all: {resp.status_code} {resp.text[:200]}")
             if resp.status_code >= 400:
                 print(f"ERROR: embed-all returned {resp.status_code}", file=sys.stderr)
                 return 1
+            result = resp.json()
+            if int(result.get("failed", 0)) > 0:
+                print(f"ERROR: embed-all failed rows: {result}", file=sys.stderr)
+                return 1
             return 0
         except httpx.RequestError as exc:
             print(f"ERROR: request failed — {exc}", file=sys.stderr)