From c658c5912c072c01ced8ea034abffc5ecb404a40 Mon Sep 17 00:00:00 2001 From: Ben Browning Date: Fri, 27 Jun 2025 11:43:39 -0400 Subject: [PATCH] fix: Gracefully handle errors when listing MCP tools When listing (and lazily indexing) tools, it's possible for an error to get thrown by individual toolgroups if for example an MCP toolgroup is unable to connect to its `mcp_endpoint`. This logs a warning in the server when that happens, logs a full stack trace of the error if debug logging is enabled, and just returns the list of tools from all working toolgroups instead of throwing an error to the client when a single toolgroup is temporarily or permanently misbehaving. Fixes #2540 A new unit test was added to test this exception handling, which is run as part of our regular test suite but also manually run to specifically verify this fix via: ``` uv run pytest -sv --asyncio-mode=auto \ tests/unit/distribution/routers/test_routing_tables.py ``` To verify the additional debug logging is printing properly: ``` LLAMA_STACK_LOGGING=core=debug \ uv run pytest -sv --asyncio-mode=auto \ tests/unit/distribution/routers/test_routing_tables.py ``` Signed-off-by: Ben Browning --- llama_stack/core/routing_tables/toolgroups.py | 11 ++++++--- .../routers/test_routing_tables.py | 24 +++++++++++++++++++ 2 files changed, 32 insertions(+), 3 deletions(-) diff --git a/llama_stack/core/routing_tables/toolgroups.py b/llama_stack/core/routing_tables/toolgroups.py index eeea406c1..91f8c4cd8 100644 --- a/llama_stack/core/routing_tables/toolgroups.py +++ b/llama_stack/core/routing_tables/toolgroups.py @@ -53,9 +53,14 @@ class ToolGroupsRoutingTable(CommonRoutingTableImpl, ToolGroups): all_tools = [] for toolgroup in toolgroups: - if toolgroup.identifier not in self.toolgroups_to_tools: - await self._index_tools(toolgroup) - all_tools.extend(self.toolgroups_to_tools[toolgroup.identifier]) + try: + if toolgroup.identifier not in self.toolgroups_to_tools: + await self._index_tools(toolgroup) + all_tools.extend(self.toolgroups_to_tools[toolgroup.identifier]) + except Exception as e: + logger.warning(f"Error listing tools for toolgroup {toolgroup.identifier}: {e}") + logger.debug(e, exc_info=True) + continue return ListToolsResponse(data=all_tools) diff --git a/tests/unit/distribution/routers/test_routing_tables.py b/tests/unit/distribution/routers/test_routing_tables.py index bbfea3f46..e09c8e305 100644 --- a/tests/unit/distribution/routers/test_routing_tables.py +++ b/tests/unit/distribution/routers/test_routing_tables.py @@ -10,6 +10,7 @@ from unittest.mock import AsyncMock import pytest +from llama_stack.apis.common.content_types import URL from llama_stack.apis.common.type_system import NumberType from llama_stack.apis.datasets.datasets import Dataset, DatasetPurpose, URIDataSource from llama_stack.apis.datatypes import Api @@ -645,3 +646,26 @@ async def test_models_source_interaction_cleanup_provider_models(cached_disk_dis # Cleanup await table.shutdown() + + +@pytest.mark.asyncio +async def test_tool_groups_routing_table_exception_handling(cached_disk_dist_registry): + """Test that the tool group routing table handles exceptions when listing tools, like if an MCP server is unreachable.""" + + exception_throwing_tool_groups_impl = ToolGroupsImpl() + exception_throwing_tool_groups_impl.list_runtime_tools = AsyncMock(side_effect=Exception("Test exception")) + + table = ToolGroupsRoutingTable( + {"test_provider": exception_throwing_tool_groups_impl}, cached_disk_dist_registry, {} + ) + await table.initialize() + + await table.register_tool_group( + toolgroup_id="test-toolgroup-exceptions", + provider_id="test_provider", + mcp_endpoint=URL(uri="http://localhost:8479/foo/bar"), + ) + + tools = await table.list_tools(toolgroup_id="test-toolgroup-exceptions") + + assert len(tools.data) == 0