Network timeouts when interacting with GitHub Enterprise Server

[jonathonlacher]

Describe the bug

I often see errors like this when interacting with GitHub Enterprise Server:

I attempted to create the pull request again, but the same network timeout issue occurred. Unfortunately, I couldn't complete the action programmatically.

And it often does work successfully on retry (clicking the retry button).

I've tried restarting the server through the UI, this doesn't seem to have much bearing on success.

Outside of MCP, getting errors while interacting with the GHES server isn't typical.

I'm also running the container with Colima (not Docker Desktop).

Perhaps exposing more details about the error could be helpful as well.

Affected version

Please run docker run -i --rm ghcr.io/github/github-mcp-server ./github-mcp-server --version and paste the output below

> docker run -i --rm ghcr.io/github/github-mcp-server ./github-mcp-server --version
server version v0.1.1 (7ab5d968e065f13a660cd123c2c5d0be6d8b6dbd) 2025-04-07T16:05:10Z

Steps to reproduce the behavior

1. Start Agentic chat
2. Ask to do something that interacts with one of the tools
3. Get a network timeout error like indicated above
4. Often does work correctly on retry, but sometimes not

[SamMorrowDrums]

Hi, that's interesting. The MCP Server itself doesn't have any timeout settings, so when calling the API itself to create the pull request, it appears to be legitimately timing out, and that would be the same if you did this manually via the API.

I wonder if there is another reason for this, but we will need to address this by improving the error reporting so that we get more detail in the error message.

Often does work correctly on retry, but sometimes not

That does sound like a legitimate timeout issue, although curious how it could take this long.