Skip to content

[Bug] when dp=2 on two nodes, cannot get content from /metrics on another node #7305

@cc8476

Description

@cc8476

Checklist

  • 1. I have searched related issues but cannot get the expected help.
  • 2. The bug has not been fixed in the latest version.
  • 3. Please note that if the bug-related issue you submitted lacks corresponding environment info and a minimal reproducible demo, it will be challenging for us to reproduce and resolve the issue, reducing the likelihood of receiving feedback.
  • 4. If the issue you raised is not a bug but a question, please raise a discussion at https://github.com/sgl-project/sglang/discussions/new/choose Otherwise, it will be closed.
  • 5. Please use English, otherwise it will be closed.

Describe the bug

Using sglang to deploy a large model with tp=16, dp=2 across two servers, so two requests will run on dp0-7 and dp8-15 respectively.

Then, executing curl -X GET http://0.0.0.0:40000/metrics -H "Authorization: Bearer xxxx" on servers A and B:

Server A (master node) has metrics, but Server B does not.
Both A and B have enable_metrics in their configurations.
What should I do?

Reproduction

curl -X GET http://0.0.0.0:40000/metrics -H "Authorization: Bearer xxxx" on Server B ,return {"detail":"Not Found"}

Environment

sglang v.0.4.6
linux :ubuntu
model:deepseek r1

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions