DesktopCommanderMCP: 'MCP error -32001: Request timed out' on startup with Claude Desktop
On macOS with node v22.14.0 (npm v10.9.2), DesktopCommanderMCP starts via npx and logs 'Server started and connected successfully', but about 58 seconds later Claude Desktop cancels the initialize request with 'Error: MCP error -32001: Request timed out' and the server disconnects. The timeout is intermittent. Is there a fix for this?
Originally asked on GitHub: https://github.com/wonderwhy-er/DesktopCommanderMCP/issues/19
3 answers
The maintainer closed this issue as fixed in release v0.1.25, but no user in the thread confirmed that the upgrade actually resolved the timeout.
What the logs show: the server initializes and connects ('Server started and connected successfully'), then roughly 58 seconds later Claude Desktop sends notifications/cancelled for requestId 0 with reason 'Error: MCP error -32001: Request timed out', and the transport closes. The server log also points to the official MCP debugging docs (modelcontextprotocol.io/docs/tools/debugging) for connection troubleshooting.
What was done in the thread:
- Upgrade to v0.1.25 or later. The maintainer released v0.1.25 (published to npm) with a batch of fixes and asked reporters to retry; he noted a Smithery install picks up the new version. Release notes: github.com/wonderwhy-er/DesktopCommanderMCP/releases/tag/v0.1.25.
- Check whether it is really this server. One commenter said the timeout was 'super intermittent' and also occurred with other MCP servers, and suspected Claude Desktop is the one timing out ("the mcp server hasn't responded to me in time") and that the session does not recover afterwards. That commenter said they would try v0.1.25 but did not post results.
If the timeout persists after upgrading, the evidence in the thread points at Claude Desktop's own request-timeout behavior as a likely cause rather than the server itself.
Did this work for you?What a GitHub user tried on GitHub (Didn't work)
Ran DesktopCommanderMCP via npx on macOS with node v22.14.0. Logs show the server started and connected, then the initialize request was cancelled about 58 seconds later with 'MCP error -32001: Request timed out' and the server transport closed.
Source: https://github.com/wonderwhy-er/DesktopCommanderMCP/issues/19 (opens external site)
Did this work for you?What @zudsniper tried on GitHub (Outcome unclear)
A commenter said the issue was intermittent and also appeared with other MCP servers, suspecting Claude Desktop's own response timeout; they said they would try v0.1.25 but did not report the result.
Did this work for you?
Related questions
- Is there a way to rollback changes made by DesktopCommanderMCP?
- DesktopCommanderMCP fails to attach under nvm: require() of ES modules is not supported
- search_code errors in DesktopCommanderMCP on Claude Desktop, then falls back to read_file/search_files
- DesktopCommanderMCP polls running processes in a loop and hits Claude limits — how to make it wait?
- DesktopCommanderMCP: how to read large HTML pages when results cap at 1048576 characters?