Chip Zoller
Chip Zoller
The current solution requires users modify the ConfigMap AFTER it has been deployed whereas this solution allows them to define their customized list of metrics up front, at installation time.
Would certainly respect whatever decision is made, but I still question the validity of needlessly creating resources but not using them. #350 is fully backwards compatible from a user perspective...
> I liked the ability to create and deploy a custom list of metrics all at once and from within a single values file. How would you envision users doing...
>what happens if I want to exclude metrics or make changes to the ones "built-in"? You'd copy the contents of the [ConfigMap](https://github.com/NVIDIA/dcgm-exporter/blob/main/deployment/templates/metrics-configmap.yaml) which contains the default metrics and supply those...
Is it OK that I just incorporated them here? If so, I'll rename/update this PR and close #350.
Ok, done. Ready for review, @glowkey.
Checking back here to see if everything looks good.
Should I also submit a PR to the operator to be able to specify these custom metrics in its values?
> Feel free, though their deployment is slightly different. Also, can you [sign your commit](https://docs.github.com/en/authentication/managing-commit-signature-verification/signing-commits)? My commits are signed. There is no DCO check in this repo as I see....
Since it looks like everything in the operator is driven through the ClusterPolicy CR, updating all the relevant code to add this one field is probably going to be beyond...