Shiva Krishna Merla

@k8s-gpubuilder you can pass `--set driver.enabled=false` during operator install. Same for toolkit if its pre-installed on the node.

yes, driver is the only component OS specific, rest all will work across ubuntu18.04 or ubuntu20.04. Not for mixed OS types though as container-toolkit is different for CentOS vs Ubuntu.

Thanks for reporting this, will fix this in upcoming patch. PR: https://gitlab.com/nvidia/kubernetes/gpu-operator/-/merge_requests/497

@adionisieva Please refer to [this](https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/getting-started.html#additional-controls-for-driver-upgrades) guide for info on driver upgrades. If you want to disable auto-drain of the node, you can set `driver.manager.env` with `ENABLE_AUTO_DRAIN` as `false`. GPU Operator...

@xhejtman Thanks for figuring this out. I will work with internal teams to get this addressed.

@heilerich Sorry we missed documenting this with vanilla K8s. With OpenShift 4.5/4.6/4.7 CRI-O config is set by default to include `/run/containers/oci/hooks.d` directory, but doesn't looks like with other flavors. Will...

@damora GPU Operator applies HTTP_PROXY, HTTPS_PROXY, NO_PROXY variables from the cluster-wide proxy object when deploying driver Daemonset. Do you see NO_PROXY was set correctly in cluster-wide proxy, but not with...

@damora we might need to add the kubernetes service cluster-ip to the NO_PROXY in cluster-wide proxy. ``` # oc get service kubernetes -n default NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE...

@damora if you have pulled this image manually, then you should be able to run with imagePullPolicy: IfNotPresent as set below. CRI-O should have been setup with proxy env when...

@damora you will need to add `imagePullPolicy: IfNotPresent` as i have added in my comment above.