swan-cern: Remove node-feature-discovery values #286
Conversation
The node-feature-discovery component values are no longer under the gpu-operator ones and are now defined seperately in the upstream cern-magnum chart. This commit reflects such change
| @@ -351,8 +351,3 @@ gpu-operator: | |||
| config: | |||
| name: nvidia-device-plugin-config | |||
| default: "default" | |||
| node-feature-discovery: | |||
| image: | |||
There was a problem hiding this comment.
Do we still need to include this somewhere, though? In the scope of which chart does cern-magnum define it?
There was a problem hiding this comment.
In the scope of their own chart, here: https://gitlab.cern.ch/kubernetes/automation/releases/cern-magnum/-/blob/master/values.yaml?ref_type=heads#L608
There was a problem hiding this comment.
But this means we do not deploy anymore any component that needs these settings? There's no node-feature-discovery anymore?
There was a problem hiding this comment.
I believe it is deployed with the CERN magnum chart, at cluster creation time
There was a problem hiding this comment.
I see, and before it was deployed by the gpu operator chart? So after this update we won't have the node-feature-discovery pods anymore in our cluster until we move to the new template?
There was a problem hiding this comment.
It will still be deployed, but through cern-magnum, not the gpu-operator
There was a problem hiding this comment.
But the cern-magnum of our current cluster is the old cern-magnum which does not have this setting, right? That's why I'm asking if before the node-feature-discovery was deployed via the gpu operator chart.
The node-feature-discovery component values are no longer under the gpu-operator ones and are now defined seperately in the upstream cern-magnum chart. This commit reflects such change