ascend-device-plugin启动失败，无法找到NPU

一、问题现象（附报错日志上下文）：
ascend-device-plugin 启动失败
```
[INFO]     2025/04/21 05:59:02.188318 1       hwlog/api.go:108    devicePlugin.log's logger init success
[INFO]     2025/04/21 05:59:02.188920 1       main.go:180    ascend device plugin starting and the version is v5.0.1.1_linux-x86_64
[INFO]     2025/04/21 05:59:02.188987 1       main.go:181    ascend device plugin starting scene is center
[ERROR]    2025/04/21 05:59:09.193946 1       devmanager/devmanager.go:171    get error card quantity: 0
[ERROR]    2025/04/21 05:59:09.194237 1       main.go:196    init devmanager failed, err: auto init failed, err: get card list failed for init
```

二、软件版本:
-- CANN 版本 (e.g., CANN 3.0.x，5.x.x):  8.0.RC3.alpha002_linux-x86_64
--Tensorflow/Pytorch/MindSpore 版本:
--Python 版本 (e.g., Python 3.7.5):
-- Driver 版本 :23.0.0
--操作系统版本 (e.g., Ubuntu 18.04):Ubuntu 2004

三、测试步骤：
透传NPU卡到虚拟机，虚拟机系统为Ubuntu 20.04，将该虚拟机添加到集群，打上标签后部署ascend-device-plugin，启动失败，提示 get error card quantity: 0

四、日志信息:
```
[INFO]     2025/04/21 05:59:02.188318 1       hwlog/api.go:108    devicePlugin.log's logger init success
[INFO]     2025/04/21 05:59:02.188920 1       main.go:180    ascend device plugin starting and the version is v5.0.1.1_linux-x86_64
[INFO]     2025/04/21 05:59:02.188987 1       main.go:181    ascend device plugin starting scene is center
[ERROR]    2025/04/21 05:59:09.193946 1       devmanager/devmanager.go:171    get error card quantity: 0
[ERROR]    2025/04/21 05:59:09.194237 1       main.go:196    init devmanager failed, err: auto init failed, err: get card list failed for init
```

Ascend/ascend-device-plugin
暂停

内容风险标识

评论 (0)

Ascend/ascend-device-plugin暂停 .gitee-modal { width: 500px !important; }

内容风险标识