Fix: e2e fails because not support latest bundles.

[xuezhaojun]

Fixes:

• https://github.com/open-cluster-management-io/clusteradm/actions/runs/5423857040/jobs/9862392036?pr=348
• https://github.com/open-cluster-management-io/clusteradm/actions/runs/5319342206/jobs/9631726852?pr=349

---

In clusteradm e2e, we should depend on a stable version of bundles, otherwise any bugs or new features imported in "registration-operator", "registration", "work" or other repos, will break or block the development of clusteradm.

Recommended practice: we can make a task to "Make clusteradm support 0.12.0" when we want to release 0.12.0 binaries.

Because we can't always promise the compatibility between latest clusteradm and the latest bundles, which would exhaust us.

So any developer who uses clusteradm to deploy latest bundles should know the risk that there might be compatibility issues.

[xuezhaojun]

/assign @qiujian16

[qiujian16]

what if we add a new feature in the latest version, and try to enable it in clusteradm? I think the root cause is we duplicate the deploy manifests in clusteradm.

[xuezhaojun]

what if we add a new feature in the latest version, and try to enable it in clusteradm? I think the root cause is we duplicate the deploy manifests in clusteradm.

That makes sense, I can fix the inconsistent part in this PR, but finally, I think we should have a way to trigger the e2e of clusteradm when manifests of registration, work, operator changes? That would be a motivation to migrate clusteradm into ocm repo.

[xuezhaojun]

what if we add a new feature in the latest version, and try to enable it in clusteradm? I think the root cause is we duplicate the deploy manifests in clusteradm.

Another thought of this is, maybe we can release a mirror version(for example, registration-operator 0.11.1) and update the clusteradm default bundle setting in e2e.

That will decouple the development work of registraiton-operator and clusteradm. after all we promote the concept of modular.

[qiujian16]

Yes, I think we need to some how trigger clusteradm test either periodically or per PR in ocm. But for this PR, I'd rather we do not change to default, but fix the permission issue

[xuezhaojun]

Yes, I think we need to some how trigger clusteradm test either periodically or per PR in ocm. But for this PR, I'd rather we do not change to default, but fix the permission issue

Got it.

[qiujian16]

/approve
/lgtm

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: qiujian16, xuezhaojun

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

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [qiujian16]

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