manifests: Add capability.openshift.io/name #414

wking · 2022-02-09T04:40:56Z

As described in the enhancement. The openshift-samples value matches our ClusterOperator name, and was registered with the API.

I've annotated these types:

$ for X in manifests/*.yaml; do yaml2json < "${X}" | jq -r '.[] | select(.metadata.annotations["capability.openshift.io/name"] == "openshift-samples").kind'; done | sort | uniq -c
      1 ClusterOperator
      4 ClusterRole
      3 ClusterRoleBinding
      2 Deployment
      1 Namespace
      1 PrometheusRule
      3 Role
      4 RoleBinding
      1 Service
      1 ServiceAccount
      1 ServiceMonitor

which seemed like the moving parts that folks who say "no samples, thanks" wouldn't want. I've left the annotation off the ImageStreams:

$ for X in manifests/*.yaml; do yaml2json < "${X}" | jq -r '.[] | select(.metadata.annotations["capability.openshift.io/name"] != "openshift-samples").kind'; done | sort | uniq -c
      9 ImageStream

since those ImageStreams are consumed by tests, the console downloads deployment, etc.

While I was touching annotations, I've also added a commit to drop the unused single-node-developer profile. Notes in the commit, and precedent in openshift/cluster-version-operator#685.

There's an enhancement proposal for this profile [1], and the Code Ready Containers folks took a run at using it in [2] before backing off in [3]. I don't have any problems with having a specific CRC profile, but if we end up going that way, we'll need a lot more (e.g. we'll need CVO manifests, and I dropped this profile there in [4]). Generated with: $ sed -i '/single-node-developer/d' manifests/*.yaml [1]: https://github.com/openshift/enhancements/blob/2911c46bf7d2f22eb1ab81739b4f9c2603fd0c07/enhancements/single-node/developer-cluster-profile.md [2]: crc-org/snc#338 [3]: crc-org/snc#373 (comment) [4]: openshift/cluster-version-operator#685

As described in [1]. The 'openshift-samples' value matches our ClusterOperator name, and was registered with the API in [2]. I've annotated these types: $ for X in manifests/*.yaml; do yaml2json < "${X}" | jq -r '.[] | select(.metadata.annotations["capability.openshift.io/name"] == "openshift-samples").kind'; done | sort | uniq -c 1 ClusterOperator 4 ClusterRole 3 ClusterRoleBinding 2 Deployment 1 Namespace 1 PrometheusRule 3 Role 4 RoleBinding 1 Service 1 ServiceAccount 1 ServiceMonitor which seemed like the moving parts that folks who say "no samples, thanks" wouldn't want. I've left the annotation off the ImageStreams: $ for X in manifests/*.yaml; do yaml2json < "${X}" | jq -r '.[] | select(.metadata.annotations["capability.openshift.io/name"] != "openshift-samples").kind'; done | sort | uniq -c 9 ImageStream since those ImageStreams are consumed by tests, the console downloads deployment, etc. I'm also adjusting the JSON Patch file to remove [3] the self-managed-high-availability annotation and add [4] the ibm-cloud-managed annotation, instead of clobbering the whole annotation set, to preserve additional annotations like the capability name. '~1` is the JSON Pointer escape for '/' [5]. [1]: https://github.com/openshift/enhancements/blame/27846285be01a2aebf8d3a04ebb8ed7f877e4959/enhancements/installer/component-selection.md#L106 [2]: https://github.com/openshift/api/blob/7e3ffb09accd36fb0536fa0e69bed5d70cccd6e5/config/v1/types_cluster_version.go#L232-L237 [3]: https://datatracker.ietf.org/doc/html/rfc6902#section-4.2 [4]: https://datatracker.ietf.org/doc/html/rfc6902#section-4.1 [5]: https://datatracker.ietf.org/doc/html/rfc6901#section-3

gabemontero · 2022-02-09T15:07:53Z

/assign @dperaza4dustbit

dperaza4dustbit

Just one question on profile-patches, not connecting the dots

dperaza4dustbit · 2022-02-10T22:23:27Z

profile-patches/ibm-cloud-managed/06-operator.yaml-patch

+  path: /metadata/annotations/include.release.openshift.io~1self-managed-high-availability
+- op: add
+  path: /metadata/annotations/include.release.openshift.io~1ibm-cloud-managed
+  value: "true"


Can you elaborate on what this is doing?

From the 024493c commit message:

I'm also adjusting the JSON Patch file to remove [3] the
self-managed-high-availability annotation and add [4] the
ibm-cloud-managed annotation, instead of clobbering the whole
annotation set, to preserve additional annotations like the capability
name. '~1` is the JSON Pointer escape for '/' [5].
...
[3]: https://datatracker.ietf.org/doc/html/rfc6902#section-4.2
[4]: https://datatracker.ietf.org/doc/html/rfc6902#section-4.1
[5]: https://datatracker.ietf.org/doc/html/rfc6901#section-3

dperaza4dustbit · 2022-02-10T22:32:17Z

/lgtm

dperaza4dustbit · 2022-02-10T22:32:57Z

/approve

dperaza4dustbit · 2022-02-10T22:33:17Z

/retest

openshift-ci · 2022-02-10T22:33:55Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: dperaza4dustbit, wking

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~~ [dperaza4dustbit]

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

openshift-bot · 2022-02-10T23:14:19Z

/retest-required