Description

OSC 1.6.0 with peer pod using GPU enabled AMI building job:

Steps to reproduce

<What actions did you take to hit the bug?>
1. Deploy OSC 1.6.0 with peer pod on OCP 4.15.11 based AWS
2. oc apply -n openshift-sandboxed-containers-operator -f https://raw.githubusercontent.com/bpradipt/coco-install/main/gpu/aws-gpu-image-create-job.yaml
3. Provide created AMI to peer-pods-cm configmap and in aws-podvm-image-cm ENABLE_NVIDIA_GPU=yes then restart caa-daemon

4. Run GPU capable workload (GPU-test)

5. connect to pod and execute: nvidia-smi -L

Expected result

To get GPU related output - Card model and statistics

Actual result

sh: 2: nvidia-smi: not found

Impact

GPU feature on AWS

Env

OCP 4.15.11 AWS

Additional helpful info

<logs, screenshot, doc links, etc.>