fix(reranker): log reranking failures instead of swallowing them silently (#5717)

tests/rerankers/test_reranker_failure_logging.py

@@ -0,0 +1,32 @@
+"""Reranker failures must be logged, not silently swallowed.
+
+Uses the LLMReranker because it is constructible without heavy ML deps (the
+``mock_llm`` fixture stubs the LLM factory). The fix under test is shared by all
+reranker providers: the ``except`` fallback now emits a ``logger.warning`` before
+degrading to the original order / a neutral score.
+"""
+
+import logging
+
+from mem0.reranker.llm_reranker import LLMReranker
+
+
+class TestRerankerFailureLogging:
+    def test_llm_failure_is_logged_and_falls_back(self, mock_llm, caplog):
+        _factory, llm_instance = mock_llm
+        llm_instance.generate_response.side_effect = RuntimeError("upstream 500")
+
+        reranker = LLMReranker({"provider": "openai"})
+        docs = [{"memory": "alpha"}, {"memory": "beta"}]
+
+        with caplog.at_level(logging.WARNING, logger="mem0.reranker.llm_reranker"):
+            result = reranker.rerank("q", docs)
+
+        # Graceful degradation preserved: every doc still comes back, scored neutral.
+        assert len(result) == 2
+        assert all(d["rerank_score"] == 0.5 for d in result)
+
+        # The failure is no longer silent.
+        warnings = [r for r in caplog.records if r.levelno == logging.WARNING]
+        assert warnings, "expected a warning to be logged on reranking failure"
+        assert "upstream 500" in caplog.text