WIP: cmdlib: Support deriving name from manifest, summary from image.yaml

[cgwalters]

Prep for variant builds:
#1459

This way variant builds can simply use separate filenames rather
than redundantly specifying the name in the rojig section (which
was always awkward to start with).

[openshift-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: cgwalters

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• OWNERS [cgwalters]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[cgwalters]

coreos/fedora-coreos-config#424

[jlebon]

Had a suggestion re. this in coreos/fedora-coreos-config#424 (comment).

[cgwalters]

Hmm, I wonder if it'd be simpler to just make #1474 use a name field in image.yaml instead? It's not strictly correct since e.g. the OSTree tarball has that name too, though it does seem cleaner to have it be more explicit and sitting alongside summary.

Yeah, the downside of going that path is it breaks the "division" we have where one can e.g. cosa build ostree and image.yaml isn't even read.

If image.yaml becomes the "source of truth"...how does it relate to manifest.yaml?

Here's another idea: we add to rpm-ostree manifests support for "extension blocks" like:

ext-meta-cosa:
  name: fedora-coreos
  summary: Fedora CoreOS
  size: 8
  extra-kargs:
    - mitigations=auto,nosmt
  ...

Basically we ignore (but pass through) all unknown keys starting with ext-meta?

Then image.yaml goes away entirely.

[jlebon]

If image.yaml becomes the "source of truth"...how does it relate to manifest.yaml?

This is why I'd prefer an assembly.yaml. :) It'd be the single entry point for cosa. E.g.

id: rhcos
name: Red Hat CoreOS
treefile: rhcos.yaml
image:
    size: 8
    ...

(And one can imagine folding the live/ convention into that file as well. And overlay.d too... though I think we should fold that into rpm-ostree instead eventually.)

Basically, we fold all the conventions into one convention: coreos-assembler will expect a file named assembly.yaml, which is the source of truth for how it shall assemble the OS and related artifacts.

We could then extend this pretty easily for variants too, e.g. riffing on #1459 (comment):

variants:
    - id: rhcos-8.1
      name: Red Hat CoreOS
      treefile: rhcos-8.1.yaml      

[openshift-ci-robot]

@cgwalters: PR needs rebase.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[openshift-ci]

@cgwalters: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
ci/prow/images	c11c67c	link	true	/test images
ci/prow/rhcos	c11c67c	link	true	/test rhcos

Full PR test history. Your PR dashboard.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here.

[dustymabe]

Closing as this is WIP for a long time. Re-open if it gets refreshed.