[Epic] Artifact Management

[justaugustus]

What would you like to be added:

We need a cohesive story around artifact management.

Broadly, this will cover:

• File artifacts
• Container images
• deb/rpm packages

There are several issues which describe various aspects of developing a process, as well as user feature requests, so to start, we should do some issue collection.

• Create an "Epics" column in the SIG Release project board and add this epic to it
• Create a project board for artifact management: https://github.com/orgs/kubernetes/projects/48
• Add the area/artifacts label to repos which may contain artifact issues/PRs: labels: Add area/artifacts label to additional repos test-infra#20111
• Scrape the respective repos for artifact issues/PRs and tag w/ area/artifacts
• Add aforementioned issues/PRs to the Artifact Management project board
• KEP cleanup:
  • Image Promotion: Image Promotion enhancements#1734
  • Artifact Management: Artifact Management enhancements#1732
  • Publishing Packages: Publishing Kubernetes packages on community infrastructure enhancements#1731
• Further populate this list and cross link all artifact issues/PRs to the epic

/assign
/milestone v1.21
cc: @kubernetes/sig-release-leads @kubernetes/release-engineering

---

Tasks from kubernetes/k8s.io#170:

• Design Document describing how all this will work.
• GCR Repository for official published images
• GCR Repository for projects to use ([Umbrella Issue] Setup a GCR Repository for projects to use k8s.io#158)
• GCS Bucket for projects to use ([Umbrella Issue] Setup a GCS Bucket for projects to use k8s.io#153)
• Process for image promotion ([Umbrella Issue] Create a Image Promotion process k8s.io#157)
• ECR Repository mirror for Amazon
• ACR Repository mirror for Azure
• HTTP Redirector for image and artifact renaming.

and @spiffxp's notes:

/remove-help
I don't think this is appropriately scoped for help wanted

/priority important-longterm
Very important for the longterm health of this project. We've migrated image hosting (k8s.gcr.io), but we haven't migrated binary artifact hosting (dl.k8s.io). So we're not all the way done and already hosting/serving artifacts is a very non-trivial amount of our spend We will need to start the conversation back up on mirroring.

Issue covers too much, should break into:

* k8s.gcr.io mirroring
* binary artifact promotion
* artifact mirroring

Can close once those are created

[justaugustus]

(Tasks from kubernetes/k8s.io#170 moved to epic description.)

[justaugustus]

More ideas from the Twitterverse: https://twitter.com/pythomit/status/1353794885099245568?s=19

cc: @tpepper