GHSA-jm63-cq47-j5jp: Race Condition
Summary In affected versions, an MCP server reached over HTTP could deliver one client's responses and notifications to another client. This happened when clients shared one stateless transport, or one McpServer or Server instance.
Messages were matched to a connection by JSON-RPC id alone. SDK clients number requests from 0, so ids collide in normal use.
Am I affected? Yes, if more than one client can reach your HTTP endpoint and your code creates the transport or the server once at startup, instead of once per request or per session. Until this release, the README examples of @modelcontextprotocol/hono, express, fastify and node showed that pattern.
Affected versions: @modelcontextprotocol/server 2.0.0 through 2.2.0.
Not affected:
- createMcpHandler() with a factory that builds a new server on every call - stdio servers - @modelcontextprotocol/sdk 1.26.0 and later
Fix Upgrade to @modelcontextprotocol/server 2.3.0 or later.
You also need to change your code. The SDK now refuses to be shared, so an affected app fails requests after upgrading: a second connect() rejects with ALREADYCONNECTED, and a stateless transport rejects its second request.
Change to one of these:
- createMcpHandler(): pass a factory that builds a new McpServer on every call. - Per request: build a new McpServer and a new transport inside the request handler. - Per session: set sessionIdGenerator and keep one McpServer and one transport per session.
If you cannot upgrade yet, the same change fixes affected versions.
Affected Software
Remediation
Recommended actions to resolve this vulnerability, in priority order.
- Upgrade
Upgrade
npm/@modelcontextprotocol/serverto a version that resolves this vulnerability.Fixed in 2.3.0 - Upgrade
Upgrade
@modelcontextprotocol/serverto a version that resolves this vulnerability.Fixed in 2.3.0 - Configuration
For each request, build a new McpServer and a new transport inside the request handler. For each session, set sessionIdGenerator and keep one McpServer and one transport per session. When using createMcpHandler(), pass a factory that builds a new McpServer on every call.
MCP server and transport lifecycle server/transport instantiation scope = per request or per session
Event History
Frequently Asked Questions
Which deployments are exposed?
HTTP MCP endpoints are affected when more than one client can reach them and the application creates a transport or server once at startup rather than once per request or session. The documented startup-sharing pattern affected examples for the hono, express, fastify, and node integrations.
What does an attacker need to exploit this?
An attacker needs network access to the HTTP MCP endpoint. Exploitation depends on sharing a stateless transport or McpServer/Server instance with another client; normal SDK request IDs can collide because clients begin numbering requests at 0.
How can we determine whether our application is affected?
Check whether it uses @modelcontextprotocol/server versions 2.0.0 through 2.2.0 and inspect server and transport lifecycle code. An application is affected if a single instance is reused across requests or sessions and multiple clients can access the HTTP endpoint.
What should we do if we cannot immediately upgrade?
Avoid sharing the transport or server between clients by creating a new server for every request or session. Using createMcpHandler() with a factory that creates a new server on each call is listed as not affected.
What changes are required when upgrading?
Upgrade to @modelcontextprotocol/server 2.3.0 or later and change the application to stop sharing instances. After upgrading, affected sharing patterns fail requests because a second connect() is rejected with ALREADY_CONNECTED.