NVidia GPU operator - Cluster policy

[jnirmalraj]

Hi Team,

In our company, having 4.16 OpenShift cluster. We are using certified operator catalog (Nvidia GPU Operator - 25.3.1).

When try to configure custom policy(gpu-cluster-policy) for dcgmExporter for custom configMap the nvidia-dcgm-exporter pod struck in the init containter.

can you help us to resolve the issue.

Note:

dcgmExporter:
config:
name: dcp-metrics-included (which contains the dcp-metrics-included.csv)
enabled: true

when remove the name (dcp-metrics-included). The nvidia-dcgm-exporter pod working fine

Thanks,
Nirmalraj

[visheshtanksale]

@jnirmalraj Can you please provide some details of the dcgm exporter pod like describing the pods and getting its logs