diff --git a/adapters/llm.py b/adapters/llm.py
index 5bb2793..3e9223a 100644
--- a/adapters/llm.py
+++ b/adapters/llm.py
@@ -295,12 +295,10 @@ class GovernedLLMClient:
         if self._breaker.is_open(self._provider, time.monotonic()):
             raise CircuitOpenError(f"熔断器已开启，拒绝调用 provider={self._provider}")
 
-        # ② call_id 生成
-        call_id = str(uuid4())
-
-        # ③ 缓存查询（cache 为 None 时跳过）
+        # ② 缓存查询（cache 为 None 时跳过）— call_id 在缓存路径独立生成
         cached = await self._cache.get(self._model, messages) if self._cache is not None else None
         if cached is not None:
+            cache_call_id = str(uuid4())
             response = LLMResponse(
                 content=cached.content,
                 thinking=cached.thinking,
@@ -312,10 +310,10 @@ class GovernedLLMClient:
                 ttft_ms=None,
                 max_inter_token_ms=None,
                 cache_hit=True,
-                call_id=call_id,
+                call_id=cache_call_id,
             )
             await self._telemetry.record_llm_call(
-                call_id=call_id,
+                call_id=cache_call_id,
                 parent_call_id=parent_call_id,
                 session_id=session_id,
                 model_name=self._model,
@@ -333,9 +331,10 @@ class GovernedLLMClient:
             )
             return response
 
-        # ④ 重试循环 + 流式消费
+        # ③ 重试循环 + 流式消费（每次 attempt 独立 call_id）
         last_exc: Exception | None = None
         for attempt in range(self._max_retries):
+            call_id = str(uuid4())
             attempt_start = time.monotonic()
             try:
                 content, thinking_text, ttft_ms, max_itoken_ms, usage = await self._call_streaming(
diff --git a/tests/unit/test_governed_llm.py b/tests/unit/test_governed_llm.py
index 0f3146a..923b6ec 100644
--- a/tests/unit/test_governed_llm.py
+++ b/tests/unit/test_governed_llm.py
@@ -197,9 +197,9 @@ async def test_transient_error_retries_and_records_telemetry():
     success_calls = [c for c in telemetry.calls if c.get("error") is None]
     assert len(success_calls) == 1
 
-    # 所有遥测记录应使用同一个 call_id（Important 2 修复验证）
+    # 每次 attempt 应使用独立的 call_id（根因修复：防遥测主键冲突）
     call_ids = {c["call_id"] for c in telemetry.calls}
-    assert len(call_ids) == 1
+    assert len(call_ids) == 3  # 2 次失败 + 1 次成功 = 3 个独立 call_id
 
 
 @pytest.mark.asyncio