stop building "scorecard-storage", "scorecard-untar" images

[acornett21]

Description of the change:
Remove "scorecard-storage", "scorecard-untar" from being built during the release process. I think this is the only place to change, though I am unsure if this is all the places.

Motivation for the change:
In order for scorecard tests to work in DCI and disconnected environments, images need to be referenced by sha and not a tag, this would stop these two testing images from getting rebuilt on release, since the rarely change.

• Relates: Operator-sdk 1.28.1 uses non-existing storage digest because the new release procedure rebuilds storage image #6417

Checklist

If the pull request includes user-facing changes, extra documentation is required:

• Add a new changelog fragment in changelog/fragments (see changelog/fragments/00-template.yaml)
• Add or update relevant sections of the docs website in website/content/en/docs

[tkrishtop]

Hi @acornett21, thank you for starting the PR.

It would also be nice to remove scorecard-untar and scorecard-storage from the Makefile to avoid rebuilding the images:

operator-sdk/Makefile

Line 95 in af14062

IMAGE_TARGET_LIST = operator-sdk helm-operator ansible-operator ansible-operator-2.11-preview scorecard-test scorecard-test-kuttl scorecard-untar scorecard-storage

We could also fix the digests at the same time by using the current digests for scorecard-storage and scorecard-untar:

quay.io/operator-framework/scorecard-storage@sha256:a3bfda71281393c7794cabdd39c563fb050d3020fd0b642ea164646bdd39a0e2

quay.io/operator-framework/scorecard-untar@sha256:2e728c5e67a7f4dec0df157a322dd5671212e8ae60f69137463bd4fdfbff8747

Here is our previous PR for changing the digests. We need more or less the same here.

[acornett21]

@tkrishtop I wasn't sure if the maintainers also wanted the makefile changed or not, was hoping for their feedback on this. I think the sha changes should be a separate PR for two reasons, 1) to not confuse the intent of this PR, 2) incase a release is cut before this gets merged and the shas change. So the intent is the sha changes will be in a separate PR. Hope this makes sense.

[tkrishtop]

@tkrishtop I wasn't sure if the maintainers also wanted the makefile changed or not, was hoping for their feedback on this. I think the sha changes should be a separate PR for two reasons, 1) to not confuse the intent of this PR, 2) incase a release is cut before this gets merged and the shas change. So the intent is the sha changes will be in a separate PR. Hope this makes sense.

1. @acornett21 agree, let's wait for @everettraven approval for Makefile.

2. The shas were already changed while cutting off the 1.28.1 release. The digests, currently used in the code, do not exist anymore, and the master branch is broken

$ podman pull quay.io/operator-framework/scorecard-storage@sha256:f7bd62664a0b91034acb977a8bb4ebb76bc98a6e8bdb943eb84c8e364828f056
Trying to pull quay.io/operator-framework/scorecard-storage@sha256:f7bd62664a0b91034acb977a8bb4ebb76bc98a6e8bdb943eb84c8e364828f056...
Error: initializing source docker://quay.io/operator-framework/scorecard-storage@sha256:f7bd62664a0b91034acb977a8bb4ebb76bc98a6e8bdb943eb84c8e364828f056: reading manifest sha256:f7bd62664a0b91034acb977a8bb4ebb76bc98a6e8bdb943eb84c8e364828f056 in quay.io/operator-framework/scorecard-storage: manifest unknown

Since we have nothing to lose, I proposed to fix everything at once, but please do as you think better.

[tkrishtop]

Hi @varshaprasad96, can you please check the question about the Makefile above? I usually ask Bryce to help us out but he is off this week.

[everettraven]

I wasn't sure if the maintainers also wanted the makefile changed or not, was hoping for their feedback on this. I think the sha changes should be a separate PR for two reasons, 1) to not confuse the intent of this PR, 2) incase a release is cut before this gets merged and the shas change. So the intent is the sha changes will be in a separate PR. Hope this makes sense.

@acornett21 @tkrishtop I think it is fine to update the Makefile as part of this PR although I don't think the Makefile has a direct impact on the release process in regards to building and pushing the images. Just to make sure I am following along though, I think the only change in the Makefile would be here:

operator-sdk/Makefile

Line 95 in af14062

IMAGE_TARGET_LIST = operator-sdk helm-operator ansible-operator ansible-operator-2.11-preview scorecard-test scorecard-test-kuttl scorecard-untar scorecard-storage

+1 on the sha changes being in a separate PR

[everettraven]

/lgtm

@acornett21 the Makefile can be updated as part of this PR if you would like but I don't think it should block this PR from making it in. From what I recall, the Makefile target for building the images is just used for local building and isn't a part of the release pipeline.

[acornett21]

@everettraven I updated the makefile as well.

[everettraven]

/lgtm

[everettraven]

Merging this despite test failures since this PR is part of helping fix these failures longer term. Thanks again @acornett21 for creating this PR!