Replies: 1 comment
-
|
When I use the exact same code but only change the MCP server image from
to
the chatbot works fine and receives the responses correctly. However, my goal is to manage all AWS resources, not just ECS, so I specifically need the aws-api-mcp-server. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hello,
(This post was translated from Korean using GPT, so there may be slight wording inaccuracies — apologies in advance.)
I am using the following MCP server to query AWS resources through a Slack-based chatbot:
👉 https://github.com/awslabs/mcp/tree/main/src/aws-api-mcp-server
Environment: Python (LangChain)-based chatbot
Behavior:
When the chatbot asks, e.g., "Show me the list of EC2 instances", the MCP server (Docker) successfully executes the AWS CLI command.
The server logs a 200 response with the expected AWS resource data.
However, the chatbot never receives the response, and instead times out.
Timeout configuration: Even when I set the timeout to 30 minutes, no response ever arrives in the chatbot.
Important note: When I run the exact same MCP module inside Cursor IDE, everything works fine. The MCP settings are identical in both Cursor IDE and my chatbot app.
Here is my MCP configuration:
"aws-api": { "command": "docker", "args": [ "run", "--rm", "--interactive", "--env", "AWS_REGION=ap-northeast-2", "--volume", "C:\\Users\\User\\.aws:/app/.aws", "public.ecr.aws/awslabs-mcp/awslabs/aws-api-mcp-server:latest" ], "enabled": true, "description": "AWS API Integration" }| INFO | awslabs.aws_api_mcp_server.core.aws.service:_log_successful_execution:215 - AWS CLI command executed successfully: { "Reservations": [ { "ReservationId": "r-************", "Instances": [ { "InstanceId": "i-************", "PrivateIpAddress": "100.4.10.117", "SecurityGroups": [ {"GroupId": "sg-************", "GroupName": "mzcsg-dummy-dev"} ], "Tags": [{"Key": "Name", "Value": "mzcec2-aps-model"}], "State": {"Name": "running"} ... } ] } ] }Despite this successful execution log, the chatbot process never gets a response back from the MCP server.
Has anyone run into a similar issue?
Why would it work fine inside Cursor IDE but fail with timeout in a Python (LangChain) application?
Could this be related to stdio handling between Docker and the chatbot app?
Any guidance or suggestions would be greatly appreciated. Thanks in advance! 🙏
Beta Was this translation helpful? Give feedback.
All reactions