Seems like it would increase adoption if the deployment experience was as easy as possible. A helm chart (or two) seems like the easiest way to go. A few thoughts on things we need to discuss or think about:
- How to package and manage the lifecycle of the CRD. See https://helm.sh/docs/topics/chart_best_practices/custom_resource_definitions/
- We talked about naming the chart "lemurnomic" since the "lemur" name was taken. Does "lemurturbo" make more sense to keep things consistent with kubeturbo?
- Will need to document values and defaults. This could go a long way in helping get this project "operationally ready" so it can be widely deployed in a large enterprise environment that may already have Istio or want to customize the deployment.
- Should the helm deployment use the operator to deploy like it does today or should we offer two separate deployment options like kubeturbo?
I'll probably think of more as I dig into it.
Seems like it would increase adoption if the deployment experience was as easy as possible. A helm chart (or two) seems like the easiest way to go. A few thoughts on things we need to discuss or think about:
I'll probably think of more as I dig into it.