-
Notifications
You must be signed in to change notification settings - Fork 30
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Sync resources from control plane to Kyma Runtimes #913
Comments
We would need some kind of design for the [KCP Sync Resource] (name TBD). I believe it should be a separate CRD that would function similarly to the Manifest CR, it would have an inline k8s manifest inside though. With a generic solution, we would be able to address the issue of Mandatory Modules as well, it would not require additional implementation on our side (i.e. Warden would be just a synced k8s manifest and not a module). This would fulfil almost all of the Mandatory Module ACs, so it is worth considering at least. |
A proposal for this KCP Sync Resource #991 |
A proposal for this resource sync setup design, I have a PR in kyma community as SDD document. |
We learned today that the VPC Peering use case requires a sync in the opposite direction as well. @pbochynski could you formulate that in the description additionally to not make it transparent for all. |
Description
Provide a way for central components to push some resources to Kyma Runtimes. It can be a secret, deployment, or any other k8s resource. The resource can be pushed to the individual cluster or to the group of clusters (e.g. all internal clusters, all beta clusters).
Reasons
Several central components need to create some resources in managed Kyma Runtimes. Use cases:
The NFS storage controller needs to create a storage class and persistent volume
Introducing a resource that contains a raw k8s manifest to install could be a generic solution. This way only one component would fetch the target cluster kubeconfig and use it to apply the resource. Together with the watcher component we can provide a tooling that will allow central components to avoid direct connection with thousands of Kyma Runtimes. As a result, we could achieve better performance (optimize the number of connections) and better security (fewer components have cluster-admin access to customer runtimes).
Acceptance Criteria
The text was updated successfully, but these errors were encountered: