MCP server timeout

[steelp02]

Before submitting your bug report

• I believe this is a bug. I'll try to join the Continue Discord for questions
• I'm not able to find an open issue that reports the same bug
• I've seen the troubleshooting guide on the Continue Docs

Relevant environment info

- OS: macOS
- Continue version: 1.0.6
- IDE version: 1.99.3
- YAML config:
mcpServers:
  - name: snowflake_pip
    command: uvenv
    args:
      - run
      - mcp_snowflake_server
      - --
      - --authenticator
      - external browser

Description

ISSUE

When using an external MCP server like for instance map_snowflake_server, it takes a while to start up the server, especially when using external browser authentication (longer than the default time in the DEFAULT_MCP_TIMEOUT parameter in core/context/mcp/MCPConnection.ts). This results in the server being started multiple times and failing every time on a connection time out (even though the server did start fully, but is not recognised by continue).

RESOLUTION

Please add an option in the YAML config to set a longer mcp_timeout value to overcome this.

To reproduce

1. install uvenv on your computer
2. add the config as above under mcpServers in config.yaml, add a username and server address (I omitted this due to security reasons)
3. look at the MCP tab in the prompt window and observe the error.

Log output

[Patrick-Erichsen]

This makes sense, thanks for the heads up @steelp02 ! Any chance this is something you'd be interested in picking up? Otherwise, not sure how soon we'll be able to prioritize this, so I added a good first issue label if anyone else is interested.

[Patrick-Erichsen]

Alternatively I wonder if we should just bump the DEFAULT_MCP_TIMEOUT. Seems like there is little downside to bumping it.

[dora-wang-x]

you can set connectionTimeout in your yaml file, this blog gives details https://mp.weixin.qq.com/s/_1tVMzH9NQY5TUIsBKvS0Q