Title: `stdio_server` uses unbuffered memory streams which can cause server to block and become unresponsive · Issue #1333 · modelcontextprotocol/python-sdk · GitHub
Open Graph Title: `stdio_server` uses unbuffered memory streams which can cause server to block and become unresponsive · Issue #1333 · modelcontextprotocol/python-sdk
X Title: `stdio_server` uses unbuffered memory streams which can cause server to block and become unresponsive · Issue #1333 · modelcontextprotocol/python-sdk
Description: Initial Checks I confirm that I'm using the latest version of MCP Python SDK I confirm that I searched for my issue in https://github.com/modelcontextprotocol/python-sdk/issues before opening this issue Description The MCP Python SDK's s...
Open Graph Description: Initial Checks I confirm that I'm using the latest version of MCP Python SDK I confirm that I searched for my issue in https://github.com/modelcontextprotocol/python-sdk/issues before opening this ...
X Description: Initial Checks I confirm that I'm using the latest version of MCP Python SDK I confirm that I searched for my issue in https://github.com/modelcontextprotocol/python-sdk/issues before opening t...
Opengraph URL: https://github.com/modelcontextprotocol/python-sdk/issues/1333
X: @github
Domain: github.com
{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"`stdio_server` uses unbuffered memory streams which can cause server to block and become unresponsive","articleBody":"### Initial Checks\n\n- [x] I confirm that I'm using the latest version of MCP Python SDK\n- [x] I confirm that I searched for my issue in https://github.com/modelcontextprotocol/python-sdk/issues before opening this issue\n\n### Description\n\nThe MCP Python SDK's `stdio_server` becomes unresponsive during slow message processing operations, causing ping and/or new requests to timeout.\n\n#### Observed Behaviour\n\n* Server appears to 'freeze' and becomes unresponsive after running for extended periods\n* Ping requests timeout during slow operations\n* New requests cannot be processed while the server is handling long-running operations\n* After 67 minutes of operation with requests every 10 seconds, the server became completely unresponsive\n\n#### Expected Behaviour\n\n* Server should remain responsive to new requests (like pings) even while processing slow operations\n* Ping requests should not timeout due to message processing delays\n* The server should handle concurrent requests without blocking\n\n### Suspected Root Cause\n\nThe `stdio_server` uses `max_buffer_size=0` (supplied to `anyio.create_memory_object_stream`) by default, which creates synchronous handoff between the stdin reader and message processor. \n\nSee: https://github.com/modelcontextprotocol/python-sdk/blob/543961968c0634e93d919d509cce23a1d6a56c21/src/mcp/server/stdio.py#L57-L58\n\nWhen the message processor is slow or blocked, the stdin reader cannot read new messages from stdin, making the server appear unresponsive.\n\n### Example Code\n\n```python\n import anyio\n import pytest\n\n\n @pytest.mark.anyio\n async def test_server_becomes_unresponsive_with_slow_processor():\n \"\"\"Demonstrates how server becomes unresponsive during slow processing.\"\"\"\n\n # Simulates stdio_server with default max_buffer_size=0 (synchronous handoff)\n send_stream, receive_stream = anyio.create_memory_object_stream(0)\n\n async def stdin_reader():\n # First message gets through\n await send_stream.send(\"request_1\")\n\n # Second message (like a ping) blocks until first is fully processed\n await send_stream.send(\"ping\") # This will block for entire processing time!\n\n async def message_processor():\n # Process first message\n msg = await receive_stream.receive()\n\n # Simulate slow processing (database query, API call, etc.)\n await anyio.sleep(0.1) # 100ms processing time\n\n # During this time, stdin_reader is completely blocked\n # No new messages (including pings) can be read!\n\n ping = await receive_stream.receive() # Finally unblocks stdin_reader\n\n async with anyio.create_task_group() as tg:\n tg.start_soon(message_processor)\n await anyio.sleep(0.01) # Let processor start waiting\n tg.start_soon(stdin_reader)\n\n send_stream.close()\n receive_stream.close()\n```\n\n### Proposed Solution\n\nAllow users to configure `max_buffer_size \u003e 0` to enable buffering, with a default value (`0`) that preserves the current behaviour.\n\n```python\nasync def stdio_server(\n stdin: anyio.AsyncFile[str] | None = None,\n stdout: anyio.AsyncFile[str] | None = None,\n max_buffer_size: int = 0, \n):\n # ...\n read_stream_writer, read_stream = anyio.create_memory_object_stream(max_buffer_size)\n write_stream, write_stream_reader = anyio.create_memory_object_stream(max_buffer_size)\n```\n\n### Python \u0026 MCP Python SDK\n\n - Python: 3.12\n - MCP Python SDK: 1.12.4\n - OS: macOS\n\n### Additional Context\n \nThe issue manifests in long-running servers where message processing can occasionally be slow (database queries, file operations, API calls). \n\nThe fix should make `max_buffer_size` configurable so users can add buffering to prevent the stdin reader from blocking during slow operations.","author":{"url":"https://github.com/peteski22","@type":"Person","name":"peteski22"},"datePublished":"2025-09-01T19:35:49.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":1},"url":"https://github.com/1333/python-sdk/issues/1333"}
| 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:d917862a-adb4-ffd6-18ee-cf59d988bf49 |
| current-catalog-service-hash | 81bb79d38c15960b92d99bca9288a9108c7a47b18f2423d0f6438c5b7bcd2114 |
| request-id | 95AA:A841C:1ED402:2BB765:6A59D5A6 |
| html-safe-nonce | c8a0f811e381988774f7c8475d20fb3e266c3371060f631911820587b14852ea |
| visitor-payload | eyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiI5NUFBOkE4NDFDOjFFRDQwMjoyQkI3NjU6NkE1OUQ1QTYiLCJ2aXNpdG9yX2lkIjoiMjA1OTMxMjI0NDY4OTg1MTgxNCIsInJlZ2lvbl9lZGdlIjoiaWFkIiwicmVnaW9uX3JlbmRlciI6ImlhZCJ9 |
| visitor-hmac | 64725c1f8d65e3350cbe52c4e664b1de363b67a0425e6c7603a9fc0bb77a6d12 |
| hovercard-subject-tag | issue:3373618859 |
| 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/1333/issue_layout |
| twitter:image | https://opengraph.githubassets.com/97bded2a02d59ef4fde7321cbe946e7d50bb544ea6ae3ac23c115fe4cacc8727/modelcontextprotocol/python-sdk/issues/1333 |
| twitter:card | summary_large_image |
| og:image | https://opengraph.githubassets.com/97bded2a02d59ef4fde7321cbe946e7d50bb544ea6ae3ac23c115fe4cacc8727/modelcontextprotocol/python-sdk/issues/1333 |
| og:image:alt | Initial Checks I confirm that I'm using the latest version of MCP Python SDK I confirm that I searched for my issue in https://github.com/modelcontextprotocol/python-sdk/issues before opening this ... |
| og:image:width | 1200 |
| og:image:height | 600 |
| og:site_name | GitHub |
| og:type | object |
| og:author:username | peteski22 |
| hostname | github.com |
| expected-hostname | github.com |
| None | ba3976babb66479b1c943a8edc0777d96157da48fadc0161f9ddb219deee8353 |
| 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 | bfacf98c3b7ad151665d6ddd216469389872b251 |
| ui-target | full |
| theme-color | #1e2327 |
| color-scheme | light dark |
Links:
Viewport: width=device-width