Project-HAMi / HAMi

Heterogeneous AI Computing Virtualization Middleware
http://project-hami.io/
Apache License 2.0
919 stars 194 forks source link

hami 适配 Ascend 910B 进容器执行 npu-smi 报错libraries: libc_sec.so: cannot open shared object file: No such file or directory #615

Open zhangpk1996 opened 5 hours ago

zhangpk1996 commented 5 hours ago

操作系统:Ubuntu 架构:arm k8s: v1.23.0 容器配置文件 { "default-runtime": "ascend", "insecure-registries": [ "0.0.0.0/0" ], "registry-mirrors": [ "https://docker.m.daocloud.io", "https://docker.rainbond.cc" ], "runtimes": { "ascend": { "path": "/usr/local/Ascend/Ascend-Docker-Runtime/ascend-docker-runtime", "runtimeArgs": [] }

hami版本:v2.4.0

1731639663151

运行了demo-nginx 分配显示正常,但是进容器执行npu-smi info 报错 npu-smi: error while loading shared libraries: libc_sec.so: cannot open shared object file: No such file or directory

demo-nginx.yaml apiVersion: v1 kind: Pod metadata: name: gpu-pod spec: containers:

archlitchi commented 2 hours ago

宿主机上npu-smi info是什么版本?

zhangpk1996 commented 2 hours ago
1731651324803

宿主机是这个版本

zhangpk1996 commented 1 hour ago

宿主机上npu-smi info是什么版本?

1731651539533

这个版本 npu-smi 24.1.rc2 Version: 24.1.rc2

archlitchi commented 1 hour ago

试试用docker直接运行vNPU

docker run -it --rm -e ASCEND_VISIBLE_DEVICES=0 -e=ASCEND_VNPU_SPECS=vir10_3c_32g ascendhub.huawei.com/public-ascendhub/ascend-mindspore:23.0.RC3-centos7-ssh bash

看看npu-smi info能否正常运行?