Kubernetes

Kubernetes has a powerful ability to alter schedule configurations, enabling it to do things such as bin-packing ( https://kubernetes.io/docs/concepts/scheduling-eviction/resource-bin-packing/ ), which can then make resource usage more efficient, or at least more predictable, when resources begin to spread across many nodes. There are other business cases, but these are all locked behind altering kube-scheduler's default configurations, or at least, adding new ones (under different scheduler names)

Containerd allows setting a directory for its CRI mirror configuration. Any configuration placed in this directory will be hot reloaded by Containerd. Unfortunately the directory setting requires a restart of Containerd to take effect. Including the following in the Containerd configuration by default would solve this. ```toml version = 2 [plugins."io.containerd.grpc.v1.cri".registry] config_path = "/etc/containerd/certs.d" ``` Having this set is useful for projects like Spegel which makes use of CRI mirror configuration to function. https://spegel.dev/docs/getting-started/

Provide a way to look up unused blobs using the REST API so we can make our own garbage-collection schedules/rules. Alternatively, stop billing us for unused space and just clean it up yourself when you see fit.

Currently DigitalOcean Managed Kubernetes doesn't provide any methods to track down control plane logs for auditing. Audit logs are highly necessary and useful for security-related issues and situations especially on production environments. This should be one of the first priority feature to be added on DOKS to let user keep their services safe and fully trust DOKS as a production-ready platform.

It would be really useful to have an option to schedule the Garbage Collector run every few days. And also be able to delete the images according to some criteria like: Last pull date Last push date Tag names or regex tag Examples Every day, delete all images that have the tag "green" and have not been pulled in the last 30 days Every Friday delete all images that have not been pulled in the last 45 days, except the tags ["latest", "1.*"]

It was surprising to me you can only have one registry per-account ( https://docs.digitalocean.com/products/container-registry/details/limits/ ). This is especially true when they are zoned in one region. For reliability, I'd like to create one in each region I run k8s clusters in. I would also like to separate permissions for say prod vs staging, by having different repositories. One related feature, which could be a solution to part of the problem - is multi-homing registries - so even one can survive clusters going down.

Was super excited to play around with cilium and hubble when the dashboard feature was expired. Meanwhile, I thought of: is there possibility of exporting the metrics to prometheus from the hubble! It would be nice to have that feature - which enable monitoring various aspect of the clusters.

Add support for static IP for egress traffic in kubernetes networking.

As of November 1, 2020, Docker Hub rate limits apply to unauthenticated or authenticated pull requests on the Docker Free plan. 100/6h anon200/6h auth'edhttps: //www.docker.com/blog/scaling-docker-to-serve-millions-more-developers-network-egress/It would be great if DO can add a registry / mirror to cache those requests. https://docs.docker.com/registry/recipes/mirror/With Kubernetes (and docker) in general, doing a lot of CI/CD will reach mentioned 100 requests limit very fast.