INTEROP-9357: Point OPP observability tests at maintained main branch

[amp-rh]

Problem

The acm-tests-observability step in OPP workflows builds its test image from
deprecated, version-specific branches of
stolostron/multicluster-observability-operator (release-2.13 through release-2.16).
These branches are unmaintained for test development, and the deprecated test suite
has generated 6 spurious firewatch tickets since December 2025
(INTEROP-8656, ACM-32822, ACM-34163, ACM-34904, ACM-35122, ACM-38013).

The MCO team has confirmed that main is the actively maintained branch for the
test suite (see ACM-34163, ACM-34904, ACM-35122 comments).

Changes

1. New ci-operator config for stolostron/multicluster-observability-operator
main__opp variant — builds the OPP test image (Dockerfile.interop) from main
   and promotes it to the acm-qe/main image stream.

2. Updated 4 consumer configs to source multicluster-observability-operator-opp
   from acm-qe/main instead of version-specific streams:

   • stolostron-policy-collection-main__ocp4.22.yaml (was 2.16)
   • stolostron-policy-collection-main__ocp5.0.yaml (was 5.0)
   • RedHatQE-interop-testing-master__acm-cnv-ocp-4.22-p2p.yaml (was 2.16)
   • RedHatQE-interop-testing-master__acm-cnv-ocp-4.21-p2p.yaml (was 2.15)

3. Generated Prow job files for the new main__opp variant (presubmits + postsubmits).

Risk

The Dockerfile.interop base image is tagged go1.21 but go.mod on main
requires Go 1.25. If the Quay base image has not been updated behind the tag,
the image build will fail in CI. A follow-up PR to
stolostron/multicluster-observability-operator to update Dockerfile.interop
would then be needed.

Verification

• make ci-operator-config ✅
• make ci-operator-checkconfig ✅
• make ci-operator-prowgen ✅
• make sanitize-prow-jobs ✅

References

• INTEROP-9357 — parent ticket
• ACM-38013 — latest firewatch ticket with full root cause analysis
• Supersedes INTEROP-9357: Point OPP observability tests at maintained main branch #82347

Summary by CodeRabbit

• Switched OPP observability CI consumers to the maintained stolostron/multicluster-observability-operator main image stream for OCP 4.21, 4.22, and 5.0 configurations.
• Added the main__opp build and promotion configuration for the multicluster-observability-operator-opp image in acm-qe/main.
• Regenerated and validated the related Prow jobs and CI configuration.
• Updated OPP preflight compatibility checks to use full OCP version keys, such as 4.14, with improved version parsing and reporting.
• Producer rehearsal passed. Consumer rehearsals are expected to fail until the promoted image becomes available after merge.

[openshift-ci-robot]

@amp-rh: This pull request references INTEROP-9357 which is a valid jira issue.

Warning: The referenced jira issue has an invalid target version for the target branch this PR targets: expected the task to target the "5.0.0" version, but no target version was set.

Details

In response to this:

Problem

The acm-tests-observability step in OPP workflows builds its test image from
deprecated, version-specific branches of
stolostron/multicluster-observability-operator (release-2.13 through release-2.16).
These branches are unmaintained for test development, and the deprecated test suite
has generated 6 spurious firewatch tickets since December 2025
(INTEROP-8656, ACM-32822, ACM-34163, ACM-34904, ACM-35122, ACM-38013).

The MCO team has confirmed that main is the actively maintained branch for the
test suite (see ACM-34163, ACM-34904, ACM-35122 comments).

Changes

1. New ci-operator config for stolostron/multicluster-observability-operator
main__opp variant — builds the OPP test image (Dockerfile.interop) from main
   and promotes it to the acm-qe/main image stream.

2. Updated 4 consumer configs to source multicluster-observability-operator-opp
   from acm-qe/main instead of version-specific streams:

• stolostron-policy-collection-main__ocp4.22.yaml (was 2.16)
• stolostron-policy-collection-main__ocp5.0.yaml (was 5.0)
• RedHatQE-interop-testing-master__acm-cnv-ocp-4.22-p2p.yaml (was 2.16)
• RedHatQE-interop-testing-master__acm-cnv-ocp-4.21-p2p.yaml (was 2.15)

3. Generated Prow job files for the new main__opp variant (presubmits + postsubmits).

Risk

The Dockerfile.interop base image is tagged go1.21 but go.mod on main
requires Go 1.25. If the Quay base image has not been updated behind the tag,
the image build will fail in CI. A follow-up PR to
stolostron/multicluster-observability-operator to update Dockerfile.interop
would then be needed.

Verification

• make ci-operator-config ✅
• make ci-operator-checkconfig ✅
• make ci-operator-prowgen ✅
• make sanitize-prow-jobs ✅

References

• INTEROP-9357 — parent ticket
• ACM-38013 — latest firewatch ticket with full root cause analysis
• Supersedes INTEROP-9357: Point OPP observability tests at maintained main branch #82347

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 openshift-eng/jira-lifecycle-plugin repository.

[coderabbitai]

Note

Reviews paused

It looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the reviews.auto_review.auto_pause_after_reviewed_commits setting.

Use the following commands to manage reviews:

• @coderabbitai resume to resume automatic reviews.
• @coderabbitai review to trigger a single review.

Use the checkboxes below for quick actions:

• ▶️ Resume reviews
• 🔍 Trigger review

Walkthrough

Adds an OPP main image build and promotion configuration, updates image consumers to use main, and changes OCP compatibility checks to use full major.minor version keys.

Changes

OPP main image integration

Layer / File(s)	Summary
Configure and consume the OPP main image ci-operator/config/stolostron/multicluster-observability-operator/..., ci-operator/config/RedHatQE/interop-testing/..., ci-operator/config/stolostron/policy-collection/...	Adds the OPP main build and promotion configuration. Updates interop and policy-collection base-image selectors from versioned names to main.
Use full OCP version keys ci-operator/step-registry/interop/opp/preflight/interop-opp-preflight-commands.sh	Changes compatibility-matrix keys to full OCP versions such as 4.14. Adds an optional displayed version for API deprecation messages.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related PRs

• openshift/release#82372: Both PRs update OPP compatibility checks to use full OCP major.minor version keys.

Suggested reviewers: gparvin, ngraham20, jan-law

🚥 Pre-merge checks | ✅ 15

✅ Passed checks (15 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the primary change: updating OPP observability tests to use the maintained main branch.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Stable And Deterministic Test Names	✅ Passed	The full PR change set contains only CI configs, generated Prow jobs, and a shell preflight script; no Ginkgo test source or title declarations changed.
Test Structure And Quality	✅ Passed	The aggregate PR diff changes only YAML/Prow configuration and a shell preflight script; it adds no Ginkgo test code or It/Eventually operations to review.
Microshift Test Compatibility	✅ Passed	The PR changes only ci-operator configs, generated jobs, and a preflight shell script; it adds no Ginkgo e2e tests or new It/Describe/Context/When declarations.
Single Node Openshift (Sno) Test Compatibility	✅ Passed	The PR changes only YAML configs/generated Prow jobs and a shell preflight script; no Go/e2e test files or added Ginkgo declarations are present.
Topology-Aware Scheduling Compatibility	✅ Passed	The PR changes CI image-stream configs and preflight validation only; no deployment, operator, controller, or topology-sensitive scheduling constraints are introduced.
Ote Binary Stdout Contract	✅ Passed	The PR changes YAML/Prow configuration and a CLI preflight shell script; it adds no OTE binary, Go test suite setup, or process-level stdout code subject to this contract.
Ipv6 And Disconnected Network Test Compatibility	✅ Passed	The PR adds no Ginkgo e2e tests and no added IPv4 assumptions or public-network calls; the only curl in changed code targets cluster-local localhost.
No-Weak-Crypto	✅ Passed	The full PR diff contains no MD5, SHA1, DES, RC4, Blowfish, ECB, crypto APIs, or secret/token comparisons; added comparisons only check versions and cluster status.
Container-Privileges	✅ Passed	The PR adds no privileged, hostPID/hostNetwork/hostIPC, SYS_ADMIN, allowPrivilegeEscalation, or root security settings; existing USER root lines are unchanged and used for package installation.
No-Sensitive-Data-In-Logs	✅ Passed	The PR adds no credentials, tokens, PII, or customer data to logs; the preflight changes rename variables/functions and retain existing cluster diagnostics.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[redhat-chai-bot]

CI Config Verification — main__opp variant

Verified the new stolostron-multicluster-observability-operator-main__opp.yaml config against four criteria:

1. Image stream tag stolostron/builder:go1.25-linux — ✅ Exists

Confirmed live on app.ci:

name:      builder:go1.25-linux
namespace: stolostron
created:   2026-07-21T19:00:55Z

The upstream main config (stolostron-multicluster-observability-operator-main.yaml) uses the identical build_root, so the new __opp variant is aligned.

2. from: root omitted — ✅ Matches established pattern

Both existing __opp variants (release-2.16__opp, release-2.17__opp) omit the from field on their Dockerfile.interop image entry. The new main__opp config follows the same pattern.

3. Dockerfile.interop build without from: root — ✅ Correct

Dockerfile.interop on main specifies its own base image:

FROM quay.io/vboulos/acmqe-automation/go:go1.21-ginkgo2.17.1

Omitting from lets ci-operator honour the Dockerfile's native FROM, which is the intended behaviour — the interop test image needs the ACM QE Go/Ginkgo base, not the build root.

4. Regenerated Prow jobs — ✅ Included

The PR includes the generated job files:

• stolostron-multicluster-observability-operator-main-presubmits.yaml (+58 lines)
• stolostron-multicluster-observability-operator-main-postsubmits.yaml (+63 lines)

Both are purely additive (no deletions), consistent with a new variant config.

---

Note on the risk called out in the PR description: The Dockerfile.interop base image tag (go1.21) vs go.mod requiring Go 1.25 is a valid concern. If the Quay base image hasn't been updated behind that tag, the image build will fail at go build time. Worth confirming separately in a rehearsal.

[amp-rh]

/pj-rehearse

[openshift-merge-bot]

@amp-rh: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel.

[redhat-chai-bot]

Rehearsal Failure Analysis

The two failing rehearsal jobs are expected failures due to a chicken-and-egg dependency — not indicative of a problem with the PR.

Failed Jobs

Job	Duration	Link
acm-cnv-ocp-4.21-p2p-lp-interop-aws	19m28s	Prow
acm-cnv-ocp-4.22-p2p-lp-interop-aws	12m44s	Prow

Root Cause

Both jobs fail with the same error at the tagging_input_image step:

dockerimage.image.openshift.io "quay.io/openshift/ci:acm-qe_main_multicluster-observability-operator-opp" not found

ci-operator attempted to import the base image acm-qe/main:multicluster-observability-operator-opp 6 times and timed out because the image does not exist yet.

Why This Is Expected

This PR simultaneously:

1. Creates the producer — adds a new ci-operator config ([stolostron-multicluster-observability-operator-main__opp.yaml](https://github.com/openshift/release/pull/82365/files#diff-a77f43f2db5ab8db tried)) and postsubmit job that will build and promote the OPP image to acm-qe/main:multicluster-observability-operator-opp.
2. Migrates the consumers — updates the interop testing configs for 4.21 and 4.22 to reference name: main instead of versioned imagestreams (2.15, 2.16).

The consumer rehearsals (the failing jobs) now point at acm-qe/main, but the producer postsubmit hasn't run yet because this PR hasn't merged. Once the PR merges and the postsubmit branch-ci-stolostron-multicluster-observability-operator-main-opp-images executes, the image will be promoted and the periodic interop jobs will resolve correctly.

Successful Rehearsals

The three passing rehearsals confirm the new configurations are valid:

• stolostron/multicluster-observability-operator/main/opp-images ✅ — the producer job itself (builds the image, doesn't need it to already exist)
• acm-cnv-ocp-4.21-p2p-images ✅ — image build only, no dependency on the promoted OPP image
• acm-cnv-ocp-4.22-p2p-images ✅ — image build only, no dependency on the promoted OPP image

---

Analysis by Chai Bot

[amp-rh]

/pj-rehearse

[openshift-merge-bot]

@amp-rh: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel.

[amp-rh]

/pj-rehearse ack

[amp-rh]

/auto-cc

[openshift-merge-bot]

@amp-rh: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel.

[amp-rh]

/pj-rehearse ack

[openshift-merge-bot]

@amp-rh: your /pj-rehearse request was not processed because the request waited in queue for longer than 5 minutes. Please retry in a few minutes.

[amp-rh]

/pj-rehearse ack

[openshift-merge-bot]

@amp-rh: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel.

[redhat-chai-bot]

Code Review

Reviewed the full diff (8 files, +174 / −25). The change is well-motivated, well-structured, and low-risk. A few observations below.

Overall Assessment: ✅ Looks good

Consolidating on main is the right call — the version-specific branches (release-2.13 through release-2.17) are unmaintained for test development, and the 6 spurious firewatch tickets since December 2025 demonstrate the real cost of pointing at them.

Specific Observations

1. New ci-operator config (main__opp.yaml) — Clean and aligned with the existing release-2.16__opp / release-2.17__opp patterns. Build root uses rhel-9-release-golang-1.25-openshift-4.22 which is consistent with the upstream main config. Promotion to acm-qe/main is correct.

2. Consumer migrations — All four configs consistently switch from version-specific image stream names to main. Verified namespace (acm-qe) and tag (multicluster-observability-operator-opp) are unchanged.

3. Generated job files — Postsubmit has ci-operator.openshift.io/is-promotion: "true" and --promote flag, presubmit has pj-rehearse.openshift.io/can-be-rehearsed: "true". Both are purely additive. Consistent with existing jobs.

4. Preflight script changes — Good improvements:

• Switching OPP_COMPAT keys from short ("18") to full form ("4.14", "4.22", etc.) improves clarity and avoids ambiguity if OCP ever changes its versioning scheme (relevant now with OCP 5.0).
• REMOVED_APIS still uses short keys accessed via target_minor, while OPP_COMPAT now uses full keys accessed via ocp_xy — this is consistent and correct; each array is accessed through the appropriate variable.
• The version-unparseable case now increments the failed counter ((( failed += 1 ))) — this is a meaningful behavioral improvement. Previously, a completely unparseable CSV version would silently pass the compatibility check. Now it correctly counts as a failure.
• Switching the unparseable warning from echo "WARNING: ..." to echo >&2 is consistent with how other error-path messages in the same function are written.

5. Risk acknowledgment — The Go version mismatch (Dockerfile.interop base tagged go1.21 vs go.mod requiring Go 1.25) is clearly called out. The rehearsal of the producer job (opp-images) succeeded, which suggests the image build itself works — but the risk is valid if the Quay base image hasn't been updated behind the tag. Worth confirming in the first post-merge run.

6. Rehearsal failures — The two consumer rehearsal failures (acm-cnv-ocp-4.21-p2p and acm-cnv-ocp-4.22-p2p) are expected chicken-and-egg: consumers can't resolve acm-qe/main:multicluster-observability-operator-opp until the producer postsubmit has run post-merge.

Minor Note

OPP_COMPAT doesn't yet have an entry for OCP 5.0 — the script handles this gracefully (skips the compatibility check with a clear message), so it's not a blocker here. A follow-up to add the 5.0 mapping would complete the matrix when the corresponding operator versions are known.

---

Review by Chai Bot on behalf of a team member.

[amp-rh]

@gparvin @ngraham20 @jan-law Friendly ping, all checks are green (rehearsals-ack applied) and this is ready for review. This points OPP observability tests at the maintained main branch to fix deprecated step references.

[amp-rh]

Chai Bot Review

Overall Assessment: ✅ Approach is sound

Consolidating from unmaintained version-specific branches (release-2.13 through release-2.16) to main is the right call. Six spurious firewatch tickets since December 2025 demonstrate the real cost of pointing at dead branches, and the MCO team has confirmed main is the actively maintained branch.

---

⚠️ One issue to clarify: from: root in the new config

The new stolostron-multicluster-observability-operator-main__opp.yaml specifies:

images:
  items:
  - dockerfile_path: Dockerfile.interop
    from: root        # <--- this
    to: multicluster-observability-operator-opp

The existing __opp variants (release-2.16__opp, release-2.17__opp) omit the from field, which lets Dockerfile.interop use its native FROM base image (the ACM QE Go/Ginkgo image).

With from: root, ci-operator replaces the Dockerfile's FROM with the build root (openshift/release:rhel-9-release-golang-1.25-openshift-4.22). This would:

• Solve the Go version mismatch (build root has Go 1.25, matching go.mod)
• Potentially lose the Ginkgo binary and test framework from the ACM QE base image

If this is intentional (using the build root to solve the Go 1.25 requirement), the risk section in the PR description may need updating since from: root largely mitigates the Dockerfile.interop base image concern. If unintentional, it should be removed to match the existing variant pattern.

Please confirm which behavior is intended.

---

✅ Consumer migrations, clean

All four consumer configs consistently switch from version-specific image stream names to main:

• interop-testing/master__acm-cnv-ocp-4.21-p2p — "2.15" → main
• interop-testing/master__acm-cnv-ocp-4.22-p2p — "2.16" → main
• policy-collection/main__ocp4.22 — "2.16" → main
• policy-collection/main__ocp5.0 — "2.17" → main

Namespace (acm-qe) and tag (multicluster-observability-operator-opp) are unchanged. Promotion target acm-qe/main aligns with the consumers. Good.

✅ Generated Prow jobs, correct

Postsubmit has ci-operator.openshift.io/is-promotion: "true" and --promote. Presubmit has pj-rehearse.openshift.io/can-be-rehearsed: "true". Both are purely additive (+63 / +58 lines, 0 deletions). Consistent with existing jobs.

✅ Preflight script improvements, solid

• OPP_COMPAT keys changed from short form ("14") to full form ("4.14") — good future-proofing for OCP 5.x where minor-only keys would be ambiguous.
• REMOVED_APIS still uses short keys via target_minor, while OPP_COMPAT uses full keys via ocp_xy — each array is accessed through its correct variable. Consistent.
• New ocpDisplay parameter adds flexibility for version display strings without changing the lookup logic.
• Unparseable CSV versions now correctly increment the failure counter instead of silently passing — meaningful bug fix.
• Style changes (CamelCase functions, typeset, trailing true) are cosmetic and don't change behavior.

✅ Rehearsal failures, expected

The two consumer rehearsal failures (acm-cnv-ocp-4.21-p2p and 4.22-p2p) are a known chicken-and-egg: consumers point at acm-qe/main which won't be populated until the producer postsubmit runs post-merge. The producer rehearsal itself succeeded. This is fine.

📝 Minor note

OPP_COMPAT doesn't have an entry for OCP 5.0 yet. The script handles this gracefully (skips the check with a message), so it's not blocking. A follow-up to add the 5.0 mapping would complete the matrix.

⚙️ Housekeeping

The PR currently needs a rebase before it can merge.

---

Summary: The approach and motivation are sound. The one item to clarify before merge is whether from: root in the new config is intentional or should be removed to match existing variant patterns.

---

Review requested by PR author via Chai Bot ([SHIP Public])

[openshift-merge-bot]

[REHEARSALNOTIFIER]
@amp-rh: the pj-rehearse plugin accommodates running rehearsal tests for the changes in this PR. Expand 'Interacting with pj-rehearse' for usage details. The following rehearsable tests have been affected by this change:

Test name	Repo	Type	Reason
pull-ci-stolostron-multicluster-observability-operator-main-opp-images	stolostron/multicluster-observability-operator	presubmit	Presubmit changed
pull-ci-RedHatQE-interop-testing-master-acm-cnv-ocp-4.22-p2p-images	RedHatQE/interop-testing	presubmit	Ci-operator config changed
pull-ci-RedHatQE-interop-testing-master-acm-cnv-ocp-4.21-p2p-images	RedHatQE/interop-testing	presubmit	Ci-operator config changed
pull-ci-stolostron-policy-collection-main-ocp4.22-images	stolostron/policy-collection	presubmit	Ci-operator config changed
periodic-ci-stolostron-policy-collection-main-ocp4.22-interop-opp-vsphere	N/A	periodic	Ci-operator config changed
periodic-ci-stolostron-policy-collection-main-ocp4.22-upgrade-interop-opp-upgrade-aws	N/A	periodic	Ci-operator config changed
periodic-ci-RedHatQE-interop-testing-master-acm-cnv-ocp-4.21-p2p-lp-interop-aws	N/A	periodic	Ci-operator config changed
periodic-ci-stolostron-policy-collection-main-ocp5.0-upgrade-interop-opp-upgrade-aws	N/A	periodic	Registry content changed
periodic-ci-stolostron-policy-collection-main-ocp5.1-upgrade-interop-opp-upgrade-aws	N/A	periodic	Registry content changed
periodic-ci-RedHatQE-interop-testing-master-acm-cnv-ocp-4.22-p2p-lp-interop-aws	N/A	periodic	Ci-operator config changed
periodic-ci-stolostron-policy-collection-main-ocp4.22-interop-opp-aws	N/A	periodic	Ci-operator config changed

Interacting with pj-rehearse

Comment: /pj-rehearse to run up to 5 rehearsals
Comment: /pj-rehearse skip to opt-out of rehearsals
Comment: /pj-rehearse {test-name}, with each test separated by a space, to run one or more specific rehearsals
Comment: /pj-rehearse more to run up to 10 rehearsals
Comment: /pj-rehearse max to run up to 25 rehearsals
Comment: /pj-rehearse auto-ack to run up to 5 rehearsals, and add the rehearsals-ack label on success
Comment: /pj-rehearse list to get an up-to-date list of affected jobs
Comment: /pj-rehearse abort to abort all active rehearsals
Comment: /pj-rehearse network-access-allowed to allow rehearsals of tests that have the restrict_network_access field set to false. This must be executed by an openshift org member who is not the PR author

Once you are satisfied with the results of the rehearsals, comment: /pj-rehearse ack to unblock merge. When the rehearsals-ack label is present on your PR, merge will no longer be blocked by rehearsals.
If you would like the rehearsals-ack label removed, comment: /pj-rehearse reject to re-block merging.

[amp-rh]

/retest ci/prow/config

[amp-rh]

/retest-required

[amp-rh]

/pj-rehearse ack

[openshift-merge-bot]

@amp-rh: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel.

[amp-rh]

/auto-cc

[amiskin94]

/approve

[chaclark1974]

Looks good

[amp-rh]

/cc @bjoydeep

[amp-rh]

/assign @bjoydeep

[chaclark1974]

/approve

[redhat-chai-bot]

Hi 👋 — this PR has been review-ready with green CI for a while and just needs an /approve from a stolostron/multicluster-observability-operator OWNERS approver to merge.

@bjoydeep @thibaultmg @philipgough — could one of you take a quick look and /approve if it looks good? The changes add a new ci-operator config for the opp variant and update the interop OPP preflight script.

Thank you! 🙏

---

AI-generated. Review for accuracy.

[bjoydeep]

@jacobbaungard or @saswatamcode @philipgough @subbarao-meduri You guys should probably review this :)

[coleenquadros]

/lgtm

[guidonguido]

Hello @amp-rh, I only have a question regarding the consumer configs: why do they target acm-qe/2.17:multicluster-observability-operator-opp and not acm-qe/main:multicluster-observability-operator-opp?

[guidonguido]

Why not "main"?

[guidonguido]

Why not "main"?

[guidonguido]

Why not "main"?

[amiskin94]

Good question. Looking at the Jira history (INTEROP-9357, ACM-38013), the consumer configs are pinned to acm-qe/2.17 intentionally as a stability measure. The main branch tracks bleeding-edge and currently carries a known Thanos v0.41.0 regression (ACM-40349, fix targeted for 2.17.1). Pinning to 2.17 gives us the maintained test suite without pulling in that regression. The new producer config (main__opp) builds from main and promotes to acm-qe/main, so once the Thanos fix lands and main stabilizes, the consumers can be updated to acm-qe/main as a follow-up. @amp-rh can confirm if that's the reasoning.

[guidonguido]

I see, thank you @amiskin94

/lgtm

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: amiskin94, amp-rh, chaclark1974, coleenquadros, gparvin, guidonguido, vboulos

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:

• ci-operator/config/RedHatQE/interop-testing/OWNERS [amp-rh]
• ci-operator/config/stolostron/multicluster-observability-operator/OWNERS [coleenquadros,guidonguido]
• ci-operator/config/stolostron/policy-collection/OWNERS [gparvin]
• ci-operator/jobs/stolostron/multicluster-observability-operator/OWNERS [coleenquadros,guidonguido]
• ci-operator/step-registry/interop/opp/preflight/OWNERS [amp-rh,chaclark1974]

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