Skip to content

[Question]: Inspector Bug: Do RecordEvents recall after StopEvent? This results in a 'segfault encountered' error #1992

@adai995995

Description

@adai995995

How is this issue impacting you?

Application crash

Share Your Debug Logs

No response

Steps to Reproduce the Issue

I enabled the NCCL Profiler Plugin and specified the inspector. However, the application crashed during training, and the error log points to the plugin. The program runs normally after I disable the plugin

NCCL Version

2.28.9+CUDA12.8

Your platform details

H200

Error Message & Behavior

Image

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