Title: Mcp-Param-* schema listing is not cancelled when a 2026-07-28 SSE client disconnects · Issue #3849 · modelcontextprotocol/python-sdk · GitHub
Open Graph Title: Mcp-Param-* schema listing is not cancelled when a 2026-07-28 SSE client disconnects · Issue #3849 · modelcontextprotocol/python-sdk
X Title: Mcp-Param-* schema listing is not cancelled when a 2026-07-28 SSE client disconnects · Issue #3849 · modelcontextprotocol/python-sdk
Description: Initial Checks I confirm that I'm using the newest release of my line (the latest 2.x, or the latest 1.x if I'm still on v1) I confirm that I searched for my issue in https://github.com/modelcontextprotocol/python-sdk/issues before openi...
Open Graph Description: Initial Checks I confirm that I'm using the newest release of my line (the latest 2.x, or the latest 1.x if I'm still on v1) I confirm that I searched for my issue in https://github.com/modelcontex...
X Description: Initial Checks I confirm that I'm using the newest release of my line (the latest 2.x, or the latest 1.x if I'm still on v1) I confirm that I searched for my issue in https://github.com/mod...
Opengraph URL: https://github.com/modelcontextprotocol/python-sdk/issues/3849
X: @github
Domain: github.com
{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Mcp-Param-* schema listing is not cancelled when a 2026-07-28 SSE client disconnects","articleBody":"### Initial Checks\n\n- [x] I confirm that I'm using the newest release of my line (the latest 2.x, or the latest 1.x if I'm still on v1)\n- [x] I confirm that I searched for my issue in https://github.com/modelcontextprotocol/python-sdk/issues before opening this issue\n\n### Release line\n\n2.x (current stable)\n\n### Description\n\n\nOn the 2026-07-28 Streamable HTTP path, a low-level `Server` without `get_tool_input_schema` finds the called\ntool's input schema for `Mcp-Param-*` validation by running its own `tools/list` handler before dispatch\n(`_mcp_param_rejection` → `_tool_input_schema`, through `serve_one`, up to `_MCP_PARAM_LIST_PAGE_CAP` pages).\n\nIn `handle_modern_request` that lookup runs before the task group that starts `watch_disconnect`. So in SSE\nmode, a client that goes away during the listing does not cancel it. The listing runs to completion; then the\n`tools/call` handler is started and immediately cancelled.\n\nThe spec says closing the SSE response stream **MUST** be treated as cancellation, and the server **SHOULD**\nstop work on the cancelled request as soon as practical (streamable-http, §Cancellation).\n\n#3630 lets a server avoid the listing with `get_tool_input_schema`. This affects servers that still take the\nlisting fallback: the default for the low-level `Server`, and frameworks built on it that do not set the hook\nyet (FastMCP 4.1.8 and 4.2.0a4, the latest releases; PrefectHQ/fastmcp#5666 adds it on their `main`). When `tools/list` is\nexpensive (per-user visibility, backend reads), every abandoned call still pays a full listing. We saw this in\nproduction on 2.2.0, with a client that fires many short calls and abandons them.\n\nExpected: a disconnect during the schema listing cancels it, and the call is never dispatched.\nActual: the listing runs to completion, then the call is dispatched and cancelled.\n\n### Example Code\n\n```python\nimport asyncio\nimport threading\nimport time\n\nimport uvicorn\nfrom mcp import types\nfrom mcp.client.client import Client\nfrom mcp.server.lowlevel.server import Server\n\nevents: list[str] = []\n\n\nasync def on_list_tools(ctx, params):\n events.append(\"tools/list started\")\n try:\n await asyncio.sleep(1.0) # a slow, per-user listing\n except asyncio.CancelledError:\n events.append(\"tools/list cancelled\")\n raise\n events.append(\"tools/list ran to completion\")\n return types.ListToolsResult(\n tools=[types.Tool(name=\"echo\", input_schema={\"type\": \"object\", \"properties\": {\"text\": {\"type\": \"string\"}}})]\n )\n\n\nasync def on_call_tool(ctx, params):\n events.append(\"tools/call dispatched\")\n try:\n await asyncio.sleep(1.0)\n except asyncio.CancelledError:\n events.append(\"tools/call cancelled\")\n raise\n return types.CallToolResult(content=[types.TextContent(type=\"text\", text=params.arguments[\"text\"])])\n\n\nserver = Server(\"repro\", on_list_tools=on_list_tools, on_call_tool=on_call_tool)\napp = server.streamable_http_app(stateless_http=True, json_response=False)\nthreading.Thread(target=uvicorn.Server(uvicorn.Config(app, port=8766, log_level=\"warning\")).run, daemon=True).start()\ntime.sleep(1)\n\n\nasync def main():\n async with Client(\"http://127.0.0.1:8766/mcp\", mode=\"2026-07-28\") as client:\n try:\n await client.call_tool(\"echo\", {\"text\": \"hi\"}, read_timeout_seconds=0.2)\n except Exception as exc:\n print(f\"client gave up: {type(exc).__name__}\")\n await asyncio.sleep(2.5)\n for event in events:\n print(\" \", event)\n\n\nasyncio.run(main())\n```\n\nOn `main` (77001c96):\n\n```text\nclient gave up: MCPError\n tools/list started\n tools/list ran to completion\n tools/call dispatched\n tools/call cancelled\n```\n\nThe client left after 0.2 s, but the listing ran its full second anyway. The fallback path is the same in\n2.2.0 and 2.3.0.\n\n### Python & MCP Python SDK\n\nPython 3.13, mcp 2.2.0 in production; reproduced on `main` @ 77001c96.\n\nRelated: #3565 / #3630 (the per-call cost of the listing), #3762 (JSON-response mode does not observe\ndisconnects).\n\nI have a small fix with tests on a branch and am happy to open a PR if assigned.\n\nDrafted with AI assistance (Claude Code), which also ran the reproduction above and wrote the fix.\n","author":{"url":"https://github.com/santangelx","@type":"Person","name":"santangelx"},"datePublished":"2026-11-11T13:00:47Z","url":"https://github.com/modelcontextprotocol/python-sdk/issues/3849"}
| react-profiling | 0 |
| react-import-map | react,react-dom,react-dom/client,react-dom/profiling,react-is,react/compiler-runtime,react/jsx-dev-runtime,react/jsx-runtime,scheduler@b8488ee2c322 |
| route-pattern | /_view_fragments/issues/show/:user_id/:repository/:id/issue_layout(.:format) |
| route-controller | voltron_issues_fragments |
| route-action | issue_layout |
| fetch-nonce | v2:3e1912fe-3e88-db6b-95b9-29968503ba0d |
| current-catalog-service-hash | 81bb79d38c15960b92d99bca9288a9108c7a47b18f2423d0f6438c5b7bcd2114 |
| request-id | 9646:117C62:11FF031:180E2FA:6B01218F |
| html-safe-nonce | 31b40ef2eaaa749f6f1f553d934ff556d9ca3cf902c362b9fa05bb136b07fea3 |
| visitor-payload | eyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiI5NjQ2OjExN0M2MjoxMUZGMDMxOjE4MEUyRkE6NkIwMTIxOEYiLCJ2aXNpdG9yX2lkIjoiODk5Nzc4MjA1OTcwMzQxMTA4NyIsInJlZ2lvbl9lZGdlIjoiaWFkIiwicmVnaW9uX3JlbmRlciI6ImlhZCJ9 |
| visitor-hmac | 07063e402cbe4ae7ee176487cc5cafc44ce67a7ff202201c942c85627f3800b6 |
| hovercard-subject-tag | issue:6328999703 |
| github-keyboard-shortcuts | repository,issues,copilot |
| google-site-verification | Apib7-x98H0j5cPqHWwSMm6dNU4GmODRoqxLiDzdx9I |
| octolytics-url | https://collector.github.com/github/collect |
| analytics-location | / |
| fb:app_id | 1401488693436528 |
| apple-itunes-app | app-id=1477376905, app-argument=https://github.com/_view_fragments/issues/show/modelcontextprotocol/python-sdk/3849/issue_layout |
| twitter:image | https://opengraph.githubassets.com/5893d58c96ad8fcf1eadd237671699d858f1422123a8ed84d67d1a03fb110e63/modelcontextprotocol/python-sdk/issues/3849 |
| twitter:card | summary_large_image |
| og:image | https://opengraph.githubassets.com/5893d58c96ad8fcf1eadd237671699d858f1422123a8ed84d67d1a03fb110e63/modelcontextprotocol/python-sdk/issues/3849 |
| og:image:alt | Initial Checks I confirm that I'm using the newest release of my line (the latest 2.x, or the latest 1.x if I'm still on v1) I confirm that I searched for my issue in https://github.com/modelcontex... |
| og:image:width | 1200 |
| og:image:height | 600 |
| og:site_name | GitHub |
| og:type | object |
| og:author:username | santangelx |
| hostname | github.com |
| expected-hostname | github.com |
| None | cc5dfdd32f62ebd3223cab3af3aa2b268d3605218237326a48a99d1efd89495c |
| turbo-cache-control | no-preview |
| go-import | github.com/modelcontextprotocol/python-sdk git https://github.com/modelcontextprotocol/python-sdk.git |
| octolytics-dimension-user_id | 182288589 |
| octolytics-dimension-user_login | modelcontextprotocol |
| octolytics-dimension-repository_id | 862584018 |
| octolytics-dimension-repository_nwo | modelcontextprotocol/python-sdk |
| octolytics-dimension-repository_public | true |
| octolytics-dimension-repository_is_fork | false |
| octolytics-dimension-repository_network_root_id | 862584018 |
| octolytics-dimension-repository_network_root_nwo | modelcontextprotocol/python-sdk |
| turbo-body-classes | logged-out env-production page-responsive |
| disable-turbo | false |
| browser-stats-url | https://api.github.com/_private/browser/stats |
| browser-errors-url | https://api.github.com/_private/browser/errors |
| release | 8c77c9e8439031ad54c75a250694842892f693e1 |
| ui-target | full |
| theme-color | #1e2327 |
| color-scheme | light dark |
| react-app-name | issues |
Links:
Viewport: width=device-width