Huawei Ascend910B4 vnpu: “DrvMngGetConsoleLogLevel failed. (g_conLogLevel=3) dcmi module initialize failed. ret is -8005”

[Traumaa]

Please provide an in-depth description of the question you have:

What do you think about this question?:
In the Huawei Ascend910B4 vnpu instance, the following error occurred: "DrvMngGetConsoleLogLevel failed. (g_conLogLevel=3)
dcmi module initialize failed. ret is -8005".

(base) root@root:/data1/yaml/npu-test# kubectl get pod -A
NAMESPACE     NAME                                                     READY   STATUS                   RESTARTS         AGE
default       alertmanager-prometheus-kube-prometheus-alertmanager-0   0/2     Init:ErrImagePull        0                3d23h
default       npu-3                                                    1/1     Running                  0                9s
default       prometheus-grafana-6b88b59b7c-2lghj                      0/3     ContainerStatusUnknown   0                5d12h
default       prometheus-grafana-6b88b59b7c-q8qlw                      0/3     ImagePullBackOff         0                3d23h
default       prometheus-kube-prometheus-operator-745c84d57d-frwwp     0/1     ContainerStatusUnknown   0                5d12h
default       prometheus-kube-prometheus-operator-745c84d57d-lpwft     0/1     ImagePullBackOff         0                3d23h
default       prometheus-kube-state-metrics-596bc6cb7c-pvc68           0/1     ContainerStatusUnknown   0                5d12h
default       prometheus-kube-state-metrics-596bc6cb7c-zwhrh           0/1     ImagePullBackOff         0                3d23h
default       prometheus-prometheus-kube-prometheus-prometheus-0       0/2     Init:ImagePullBackOff    0                3d23h
default       prometheus-prometheus-node-exporter-rzxbw                0/1     ImagePullBackOff         0                3d23h
kube-system   calico-kube-controllers-658d97c59c-mpz4z                 1/1     Running                  2 (56d ago)      76d
kube-system   calico-node-gwmz6                                        1/1     Running                  1 (75d ago)      76d
kube-system   coredns-5884d58d84-82vwj                                 1/1     Running                  0                71d
kube-system   coredns-5884d58d84-zmn6k                                 1/1     Running                  0                71d
kube-system   etcd-root                                                1/1     Running                  2 (75d ago)      76d
kube-system   hami-ascend-device-plugin-hn9dp                          1/1     Running                  0                3d16h
kube-system   hami-scheduler-5654fd7b7-jmbh9                           2/2     Running                  0                9d
kube-system   kube-apiserver-root                                      1/1     Running                  32 (75d ago)     76d
kube-system   kube-controller-manager-root                             1/1     Running                  1495 (39d ago)   76d
kube-system   kube-proxy-x9p22                                         1/1     Running                  1 (75d ago)      76d
kube-system   kube-scheduler-root                                      1/1     Running                  1489 (39d ago)   76d
(base) root@root:/data1/yaml/npu-test# kubectl -n default exec -it npu-3 -- bash
HwHiAiUser@npu-3:~$ npu-smi info
DrvMngGetConsoleLogLevel failed. (g_conLogLevel=3)
dcmi module initialize failed. ret is -8005

Is it because of the following reasons?

(base) root@root:/data1/yaml# kubectl get node root -o json | jq -r '.status.allocatable | to_entries[] | select(.key|contains("huawei.com"))'
{
  "key": "huawei.com/Ascend910B4",
  "value": "32"
}
(base) root@root:/data1/yaml# cd npu-test/

Environment:

• HAMi version: 2.7.0
• Kubernetes version: v1.28.15
• Others: ascend-device-plugin: projecthami/ascend-device-plugin:v1.1.0

[archlitchi]

you need to consult your device provider for this error