From 43aa97b7d2d7ece18efc3c3de62081e2cd6d1b9e Mon Sep 17 00:00:00 2001
From: Kian Jones <11655409+kianjones9@users.noreply.github.com>
Date: Wed, 10 Dec 2025 16:56:33 -0800
Subject: [PATCH] fix: prevent db connection pool exhaustion in MCP tool
 creation (#6621)

Problem: When creating an MCP server with many tools, the code used
asyncio.gather to create all tools concurrently. Each tool creation involves
database operations (INSERT with upsert logic), leading to N concurrent
database connections.

Example: An MCP server with 50 tools creates 50 simultaneous database
connections during server creation, exhausting the connection pool.

Root cause: asyncio.gather(*[create_mcp_tool_async(...) for tool in tools])
processes all tool creations concurrently, each opening a DB session.

Solution: Create tools sequentially instead of concurrently. While this takes
longer for server creation, it prevents database connection pool exhaustion
and maintains error handling by catching exceptions per tool.

Changes:
- apps/core/letta/services/mcp_manager.py:
  - Replaced asyncio.gather with sequential for loop
  - Maintained return_exceptions=True behavior with try/except
  - Added explanatory comment about db pool exhaustion prevention

Impact: With 50 MCP tools:
- Before: 50 concurrent DB connections (pool exhaustion)
- After: 1 DB connection at a time (no pool exhaustion)

Note: This follows the same pattern as PR #6617, #6619, and #6620 which
fixed similar issues in file operations, multi-agent execution, and file
status checks.
---
 letta/services/mcp_manager.py | 22 ++++++++++++++--------
 1 file changed, 14 insertions(+), 8 deletions(-)

diff --git a/letta/services/mcp_manager.py b/letta/services/mcp_manager.py
index b344ec3f..37f2d33d 100644
--- a/letta/services/mcp_manager.py
+++ b/letta/services/mcp_manager.py
@@ -492,17 +492,23 @@ class MCPManager:
             # Filter out invalid tools
             valid_tools = [tool for tool in mcp_tools if not (tool.health and tool.health.status == "INVALID")]
 
-            # Register in parallel
+            # Register tools sequentially to avoid exhausting database connection pool
+            # When an MCP server has many tools (e.g., 50+), concurrent tool creation can create
+            # too many simultaneous database connections, causing pool exhaustion errors
             if valid_tools:
-                tool_tasks = []
+                results = []
                 for mcp_tool in valid_tools:
                     tool_create = ToolCreate.from_mcp(mcp_server_name=created_server.server_name, mcp_tool=mcp_tool)
-                    task = self.tool_manager.create_mcp_tool_async(
-                        tool_create=tool_create, mcp_server_name=created_server.server_name, mcp_server_id=created_server.id, actor=actor
-                    )
-                    tool_tasks.append(task)
-
-                results = await asyncio.gather(*tool_tasks, return_exceptions=True)
+                    try:
+                        result = await self.tool_manager.create_mcp_tool_async(
+                            tool_create=tool_create,
+                            mcp_server_name=created_server.server_name,
+                            mcp_server_id=created_server.id,
+                            actor=actor,
+                        )
+                        results.append(result)
+                    except Exception as e:
+                        results.append(e)
 
                 successful = sum(1 for r in results if not isinstance(r, Exception))
                 failed = len(results) - successful