Advertisement

亲测解决The client socket has failed to connect to

阅读量:

问题源于深度学习相关服务未能与本地主机建立连接。解决方案在于确保所有进程从rank 0开始启动。

报错原文

复制代码
    [W socket.cpp:663] [c10d] The client socket has failed to connect to [-xiaohu]:12345 (errno: 22 - Invalid argument).
    
    
    python

解决方法

Rank应该从0开始,Rank should start from 0。

复制代码
    opt.rank = kwargs.get("start_rank", 0) + opt.gpu_id
    
    
    python

To

复制代码
    opt.rank = kwargs.get("start_rank", 0) + i
    
    
    python

原版笔记

An invalid socket exists, indicating potential issues with connectivity.

全部评论 (0)

还没有任何评论哟~