unrevert of read from cosmos by deads2k · Pull Request #4610 · Azure/ARO-HCP

David Eads (deads2k) · 2026-03-24T19:19:47Z

/hold

need to add fixes from #4412

Manyanda Chitimbo (machi1990) · 2026-03-24T21:55:12Z

Line 311 in 428d5ac

    
           func GetClusterServiceUserAssignedIdentities(clusterServiceCluster *arohcpv1alpha1.Cluster) map[string]*arm.UserAssignedIdentity {

is written will get us in a spot where we still can set identity with empty clientId and principalId

This

ARO-HCP/internal/ocm/convert.go

Line 318 in 428d5ac

clientID, _ := operatorIdentity.GetClientID()

needs to
be something like

clientID, ok := operatorIdentity.GetClientID() and we only set the cliendID to the returned response when ok is true.

We also need to ensure that the migration identity controller can handle the case of only some values being partially set

ARO-HCP/backend/pkg/controllers/clusterpropertiescontroller/identity_migration.go

Line 80 in 428d5ac

    
           func (c *identityMigrationSyncer) NeedsWork(ctx context.Context, existingCluster *api.HCPOpenShiftCluster) bool {

and resync anything.

i.e NeedsWork re-written in a way that it says work is needed if any of the identity elements have a nil clientId or principalId

David Eads (deads2k) · 2026-03-25T22:03:35Z

 	}
+
+	// Clear the user-assigned identities map since that is built by a controller. The defaults will be set next and valid until we know the actual values.
+	newInternalCluster.Identity.UserAssignedIdentities = nil


Behavior question I don't know the answer to. Do we need to validate the user provided .Identity.UserAssignedIdentities for some reason for ARM even though our long-held authoritative source is actually in .CustomerProperties.Platform.OperatorsAuthentication.UserAssignedIdentities.DataPlaneOperators (etc).

If we need to validate that every entry the user provided is present, we need to remove the defaulting I added in this PR and add the information only during reads.

Ben Vesel (@bennerv) Matthew Barnes (@mbarnes) will know more;
I believe we could also get away by not overwriting the values and the EnsureDefaults() method.

By that I mean persist whatever we received from the user for the .Identity property as ARM would've validate that prior to the frontend processing the request. We remove the EnsureDefaults().

Then in the controller we sync the properties for the .Identity only if the clientId or principalId are empty for a given entry. When syncing the properties, if CS hasn't set it i.e ok = false then we make sure to live that as nil so that it'll be omitted from the returned json.

I'm not sure I understand the scope of your question wrt to validation so forgive me if I overanswer.

The .Identity.UserAssignedIdentities part of the API is the means by which user-assigned managed identities actually get assigned to the cluster resource by ARM.

I believe ARM does some validation here before the request reaches the RP, insofar as it will catch references to identities that don't exist. That would be worth verifying though.

The ARM requirement that makes this section tricky is that we have to provide the client and principal ID for each key in .Identity.UserAssignedIdentities whenever we return the cluster resource in a response (aside from the initial PUT response). Those values must be obtained from the Managed Identity Azure service.

Currently Cluster Service obtains those values and supplies them to the RP, which means that in order to satisfy the ARM requirement there has to be a strict 1:1 mapping of keys in the .Identity.UserAssignedIdentities map to a key in either the ...OperatorsAuthentication.UserAssignedIdentities.ControlPlaneOperators map or to the ...OperatorsAuthentication.UserAssignedIdentities.ServiceManagedIdentity field.

Once the backend is obtaining the client and principal IDs itself, we could relax this constraint slightly by tolerating extraneous keys in .Identity.UserAssignedIdentities that don't appear under ...OperatorsAuthentication.UserAssignedIdentities. Vice versa, however, won't work. If a key in ...OperatorsAuthentication.UserAssignedIdentities.ControlPlaneOperators does not appear in .Identity.UserAssignedIdenties, then that would mean the identity is not actually assigned to the cluster resource and won't be any good to the control plane operator.

In terms of the content of the .Identity.UserAssignedIdentities map, the value for each key must be an empty JSON object or -- in the case of a cluster update -- I believe it can be a JSON object that matches exactly what would be returned in a GET request (i.e. a JSON object with clientId and principalId keys and values that match what Cluster Service provides).

Ok, I think the summary of all that is, "we must validate the user input as provided. we must return frontend values with all the keys and either accurate or empty values." Doable. Will update.

Manyanda Chitimbo (machi1990) · 2026-03-25T22:09:31Z

+	for _, operatorIdentityResourceID := range cluster.CustomerProperties.Platform.OperatorsAuthentication.UserAssignedIdentities.DataPlaneOperators {
+		if cluster.Identity == nil {
+			cluster.Identity = &arm.ManagedServiceIdentity{}
+		}
+		if cluster.Identity.UserAssignedIdentities == nil {
+			cluster.Identity.UserAssignedIdentities = make(map[string]*arm.UserAssignedIdentity)
+		}
+
+		if _, ok := cluster.Identity.UserAssignedIdentities[operatorIdentityResourceID.String()]; !ok {
+			cluster.Identity.UserAssignedIdentities[operatorIdentityResourceID.String()] = &arm.UserAssignedIdentity{}
+		}
+	}


It's not needed for data plane; only ControlPlane + SMI identities should be in the cluster.Identity. S
Something like

ARO-HCP/demo/bicep/cluster.bicep

Line 574 in 978e85c

identity: {

is what the user will provide which contains only the SMI + CP ones as those are treated differently from the DP ones.
For DP we don't request credentials directly from the MI RP but rather perform federations on them so they don't need to be in this list. However for the CP + SMI, do interact with the MI RP and hence the need for the .Identity stanza which makes ARM give the FPA credential ability to retrieve credentials for these identities from the MI RP

Manyanda Chitimbo (machi1990) · 2026-03-25T22:15:48Z

+			return true
+		}
+	}
+	for _, operatorIdentityResourceID := range existingCluster.CustomerProperties.Platform.OperatorsAuthentication.UserAssignedIdentities.DataPlaneOperators {


Same as #4610 (comment) we don't need to do this for DP identities

Manyanda Chitimbo (machi1990) · 2026-03-25T22:25:34Z

 				clientID, _ := operatorIdentity.GetClientID()
 				principalID, _ := operatorIdentity.GetPrincipalID()
-				internalCluster.Identity.UserAssignedIdentities[operatorIdentity.ResourceID()] = &arm.UserAssignedIdentity{
+				ret[operatorIdentity.ResourceID()] = &arm.UserAssignedIdentity{
 					ClientID:    &clientID,
 					PrincipalID: &principalID,
 				}


This should be written as so that we don't set the value to "" and potentially returning it to the end user when CS hasn't set it yet.

uai := &arm.UserAssignedIdentity{} if clientID, ok := operatorIdentity.GetClientID(); ok { uai.ClientID = &clientID } if principalID, ok := operatorIdentity.GetPrincipalID(); ok { uai.PrincipalID = &principalID } ret[operatorIdentity.ResourceID()] = uai

CS only sets this value async (as part of the cluster provisioning process and not during cluster creation).

This will avoid putting us in a situation where we return something like this

.. identity: { "..../kms": {"clientID": "", "principalID": ""} }

back to the end user when they do a GET and once they attempt a PUT using the gotten response it'll fail.

David Eads (@deads2k) bumping this ^

Manyanda Chitimbo (machi1990) · 2026-03-25T22:26:12Z

 				clientID, _ := mi.ServiceManagedIdentity().GetClientID()
 				principalID, _ := mi.ServiceManagedIdentity().GetPrincipalID()
-				internalCluster.Identity.UserAssignedIdentities[mi.ServiceManagedIdentity().ResourceID()] = &arm.UserAssignedIdentity{
+				ret[mi.ServiceManagedIdentity().ResourceID()] = &arm.UserAssignedIdentity{
 					ClientID:    &clientID,
 					PrincipalID: &principalID,
 				}


Same as #4610

David Eads (@deads2k)

David Eads (deads2k) · 2026-03-30T19:49:21Z

/retest

David Eads (deads2k) · 2026-03-30T21:39:53Z

/retest

David Eads (deads2k) · 2026-03-30T23:44:29Z

/retest

David Eads (deads2k) · 2026-03-31T13:09:42Z

/hold

Manyanda Chitimbo (machi1990)

Took another round of reviews.
Just 1 one comment and bumping 2 old related comments that I think will be critical to address

Manyanda Chitimbo (machi1990) · 2026-03-31T13:55:00Z

+
+		controlPlaneExists := false
+		for _, resourceID := range existingCluster.CustomerProperties.Platform.OperatorsAuthentication.UserAssignedIdentities.ControlPlaneOperators {
+			if resourceID != nil && resourceID.String() == operatorIdentityResourceIDString {


Suggested change

if resourceID != nil && resourceID.String() == operatorIdentityResourceIDString {

if resourceID.String() == operatorIdentityResourceIDString {

is the nil check needed?

Manyanda Chitimbo (machi1990) · 2026-03-31T14:01:40Z

 				clientID, _ := operatorIdentity.GetClientID()
 				principalID, _ := operatorIdentity.GetPrincipalID()
-				internalCluster.Identity.UserAssignedIdentities[operatorIdentity.ResourceID()] = &arm.UserAssignedIdentity{
+				ret[operatorIdentity.ResourceID()] = &arm.UserAssignedIdentity{
 					ClientID:    &clientID,
 					PrincipalID: &principalID,
 				}


David Eads (@deads2k) bumping this ^

Manyanda Chitimbo (machi1990) · 2026-03-31T14:01:54Z

 				clientID, _ := mi.ServiceManagedIdentity().GetClientID()
 				principalID, _ := mi.ServiceManagedIdentity().GetPrincipalID()
-				internalCluster.Identity.UserAssignedIdentities[mi.ServiceManagedIdentity().ResourceID()] = &arm.UserAssignedIdentity{
+				ret[mi.ServiceManagedIdentity().ResourceID()] = &arm.UserAssignedIdentity{
 					ClientID:    &clientID,
 					PrincipalID: &principalID,
 				}


David Eads (@deads2k)

Manyanda Chitimbo (machi1990)

/lgtm

David Eads (deads2k) · 2026-03-31T18:59:38Z

/retest

David Eads (deads2k) · 2026-03-31T19:00:57Z

/hold cancel

Manyanda Chitimbo (machi1990) · 2026-04-02T12:12:30Z

-		ServiceProviderProperties: api.HCPOpenShiftClusterServiceProviderProperties{
-			DNS: api.ServiceProviderDNSProfile{
-				BaseDomain: cluster.DNS().BaseDomain(),
-			},
-			Console: api.ServiceProviderConsoleProfile{
-				URL: cluster.Console().URL(),
-			},
-			API: api.ServiceProviderAPIProfile{
-				URL: cluster.API().URL(),
-			},
-			Platform: api.ServiceProviderPlatformProfile{
-				IssuerURL: cluster.Azure().OidcIssuerUrl(),
-			},
-		},


Looks like this removal is problematic

{fail [github.com/Azure/ARO-HCP/test/e2e/cluster_authorized_cidrs_connectivity.go:145]: cluster Properties.API.URL was nil Expected <*string | 0x0>: nil not to be nil fail [github.com/Azure/ARO-HCP/test/e2e/cluster_authorized_cidrs_connectivity.go:145]: cluster Properties.API.URL was nil Expected <*string | 0x0>: nil not to be nil}

it passed on a retest. Looks like it was just a flake

… mutation Rather than clearing entirely, this change has the frontend create a valid default. We also default on reading from storage so the return value is always valid for the RP.

If the strings aren't longer than zero, they aren't valid and confuse clients.

Copilot

Pull request overview

This PR shifts cluster reads/responses further toward Cosmos-only state (removing several legacy Cluster Service → RP conversion/merge paths) and updates integration fixtures accordingly, while adding logic to “complete” identity.userAssignedIdentities based on operator identity references.

Changes:

Remove legacy Cluster Service → RP cluster conversion and related tests/controllers; reduce frontend reliance on Cluster Service for reads/listing.
Introduce frontend identity “completion” to reconcile/prune identity.userAssignedIdentities from operator identity references.
Update integration test artifacts/fixtures to reflect the new identity payload shape and Cosmos document shape.

Reviewed changes

Copilot reviewed 20 out of 23 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
`frontend/pkg/frontend/cluster.go`	Removes CS merge on list/read; adds `completeClusterIdentity` and calls it during create/update/read.
`internal/ocm/convert.go`	Deletes legacy CS→RP conversion helpers; adds `GetClusterServiceUserAssignedIdentities` helper.
`internal/ocm/convert_test.go`	Removes tests tied to deleted legacy conversion path.
`internal/database/convert_defaults_consistency_test.go`	Removes defaults-consistency test that depended on legacy CS→RP conversion.
`internal/database/convert_cluster.go`	Adjusts Cosmos identity conversion to deep-copy values when present.
`backend/pkg/controllers/clusterpropertiescontroller/identity_migration.go`	Updates identity migration to use `GetClusterServiceUserAssignedIdentities`; expands `NeedsWork` logic.
`backend/pkg/controllers/clusterpropertiescontroller/identity_migration_test.go`	Updates identity migration tests for operator identity references.
`backend/pkg/controllers/clusterpropertiescontroller/cluster_customer_properties_migration*.go`	Removes the customer properties migration controller and its tests.
`backend/pkg/app/backend.go`	Stops registering/running the removed customer properties migration controller.
`test-integration/frontend/artifacts/**`	Updates fixtures for Cosmos cluster docs, identity payload shape, and expected validation errors.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

…e scanner

Miguel Soriano (miguelsorianod) · 2026-05-04T09:06:32Z

-	// TODO this is bad, see above TODOs. We want to validate what we store.
-	newInternalCluster.Identity.UserAssignedIdentities = nil
+	// we must validate using user provided .Identity.UserAssignedIdentities because that is the intent expressed by the user to allow
+	// us to use these identities. The information contained in those key is not trusted to be accurate, so we clear this field and set to


Could you expand on this? I thought the end-user needs to provide both .Identity.UserAssignedIdentities (as it's a requirement by ARM) as well as the .platform.operatorsAuthentication.* data and we would just limit what we do to validate all the expected data is there and consistent between them instead of modifying .Identity with something different than they have provided.

We validate the identity in the call above. We clear the input before storage and set it to an empty valid value.

Copilot

Pull request overview

Copilot reviewed 20 out of 23 changed files in this pull request and generated 5 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

 	if err != nil {
 		return utils.TrackError(err)
 	}
-	clustersByClusterServiceID := make(map[string]*api.HCPOpenShiftCluster)
 	for _, internalCluster := range internalClusterIterator.Items(ctx) {
-		if internalCluster.ServiceProviderProperties.ClusterServiceID == nil {
-			// TODO this will be removed during our switch to read only from cosmos.
-			// we can still merge now since the value will never be nil until both the read path is fixed and this PR makes it to prod.
-			continue
+		resultingExternalCluster := versionedInterface.NewHCPOpenShiftCluster(internalCluster)
+		jsonBytes, err := arm.MarshalJSON(resultingExternalCluster)
+		if err != nil {
+			return utils.TrackError(err)


+		if val, ok := cluster.Identity.UserAssignedIdentities[operatorIdentityResourceID.String()]; !ok || val == nil {
+			if existingValue, hasExisting := existingUserAssignedIdentity[operatorIdentityResourceID.String()]; hasExisting {
+				cluster.Identity.UserAssignedIdentities[operatorIdentityResourceID.String()] = existingValue.DeepCopy()
+			} else {
+				cluster.Identity.UserAssignedIdentities[operatorIdentityResourceID.String()] = &arm.UserAssignedIdentity{}
+			}


+		if val, ok := cluster.Identity.UserAssignedIdentities[serviceManagedIdentity.String()]; !ok || val == nil {
+			if existingValue, hasExisting := existingUserAssignedIdentity[serviceManagedIdentity.String()]; hasExisting {
+				cluster.Identity.UserAssignedIdentities[serviceManagedIdentity.String()] = existingValue.DeepCopy()
+			} else {
+				cluster.Identity.UserAssignedIdentities[serviceManagedIdentity.String()] = &arm.UserAssignedIdentity{}
+			}


+	if serviceManagedIdentity := existingCluster.CustomerProperties.Platform.OperatorsAuthentication.UserAssignedIdentities.ServiceManagedIdentity; serviceManagedIdentity != nil {
+		expectedIdentityResourceIDs[serviceManagedIdentity.String()] = struct{}{}
+	}


+        "message": "Invalid value: \"/subscriptions/different-sub/resourceGroups/some-resource-group/providers/Microsoft.Network/virtualNetworks/test-vnet/subnets/test-subnet\": must not be the same resource group name: \"some-resource-group\"",
+        "target": "properties.platform.subnetId"
+      }
      {
        "code": "InvalidRequestContent",


Miguel Soriano (miguelsorianod) · 2026-05-05T18:25:37Z

/lgtm

openshift-ci · 2026-05-05T18:25:45Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: deads2k, machi1990, miguelsorianod

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~~ [deads2k]

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

openshift-merge-bot · 2026-05-05T21:05:04Z

/retest-required

Remaining retests: 0 against base HEAD 2c8efd8 and 2 for PR HEAD daf3b71 in total

openshift-ci · 2026-05-05T22:56:54Z

David Eads (@deads2k): 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/cspr	`4252e83`	link	true	`/test cspr`
ci/prow/images-push	`4252e83`	link	true	`/test images-push`

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-sigs/prow repository. I understand the commands that are listed here.

openshift-merge-bot · 2026-05-06T01:06:08Z

/retest-required

Remaining retests: 0 against base HEAD c2abba6 and 1 for PR HEAD daf3b71 in total

Manyanda Chitimbo (machi1990) · 2026-05-06T10:34:55Z

/retest-required

openshift-ci Bot added the do-not-merge/hold label Mar 24, 2026

openshift-ci Bot requested review from Jan-Hendrik Boll (janboll) and Matthew Barnes (mbarnes) March 24, 2026 19:19

openshift-ci Bot added the approved label Mar 24, 2026

David Eads (deads2k) force-pushed the cs-147-read-from-cosmos branch from 428d5ac to 978e85c Compare March 25, 2026 22:00

David Eads (deads2k) commented Mar 25, 2026

View reviewed changes

Manyanda Chitimbo (machi1990) reviewed Mar 25, 2026

View reviewed changes

openshift-ci Bot added the needs-rebase label Mar 26, 2026

David Eads (deads2k) force-pushed the cs-147-read-from-cosmos branch from 978e85c to 9e05098 Compare March 27, 2026 18:36

openshift-ci Bot removed the needs-rebase label Mar 27, 2026

David Eads (deads2k) force-pushed the cs-147-read-from-cosmos branch from 9e05098 to 7f7d6e5 Compare March 30, 2026 18:33

Manyanda Chitimbo (machi1990) suggested changes Mar 31, 2026

View reviewed changes

Manyanda Chitimbo (machi1990) approved these changes Mar 31, 2026

View reviewed changes

openshift-ci Bot assigned Manyanda Chitimbo (machi1990) Mar 31, 2026

openshift-ci Bot added the lgtm label Mar 31, 2026

openshift-ci Bot removed the do-not-merge/hold label Mar 31, 2026

David Eads (deads2k) force-pushed the cs-147-read-from-cosmos branch from 53ffef9 to 3d07b56 Compare April 1, 2026 18:47

openshift-ci Bot removed the lgtm label Apr 1, 2026

David Eads (deads2k) added the lgtm label Apr 1, 2026

Manyanda Chitimbo (machi1990) reviewed Apr 2, 2026

View reviewed changes

David Eads (deads2k) added 3 commits April 30, 2026 09:28

Adjust frontend to build valid, empty default for Identity on cluster…

b22fce0

… mutation Rather than clearing entirely, this change has the frontend create a valid default. We also default on reading from storage so the return value is always valid for the RP.

Only create valid userassignedidentities

74879b9

If the strings aren't longer than zero, they aren't valid and confuse clients.

formating

83ab41e

Copilot AI review requested due to automatic review settings April 30, 2026 13:31

David Eads (deads2k) force-pushed the cs-147-read-from-cosmos branch from 7f85975 to 83ab41e Compare April 30, 2026 13:31

openshift-ci Bot removed the needs-rebase label Apr 30, 2026

Copilot started reviewing on behalf of David Eads (deads2k) April 30, 2026 13:32 View session

Copilot AI reviewed Apr 30, 2026

View reviewed changes

address a few nil checks that aren't a practical problem, but trip th…

068ef18

…e scanner

Miguel Soriano (miguelsorianod) reviewed May 4, 2026

View reviewed changes

Comment thread internal/ocm/convert.go

Miguel Soriano (miguelsorianod) reviewed May 4, 2026

View reviewed changes

Comment thread internal/ocm/convert.go

Miguel Soriano (miguelsorianod) reviewed May 4, 2026

View reviewed changes

return empty, but valid identities

2dba438

Copilot AI review requested due to automatic review settings May 4, 2026 13:37

Copilot started reviewing on behalf of David Eads (deads2k) May 4, 2026 13:38 View session

Copilot AI reviewed May 4, 2026

View reviewed changes

Miguel Soriano (miguelsorianod) reviewed May 4, 2026

View reviewed changes

Comment thread internal/ocm/convert.go

Miguel Soriano (miguelsorianod) mentioned this pull request May 4, 2026

feat: stop using clusers-service for frontend externalauth reads #5110

Merged

correct more small comments

daf3b71

Miguel Soriano (miguelsorianod) mentioned this pull request May 5, 2026

feat: move cluster creation from frontend to backend #5120

Closed

openshift-ci Bot assigned Miguel Soriano (miguelsorianod) May 5, 2026

openshift-ci Bot added the lgtm label May 5, 2026

openshift-merge-bot Bot merged commit 737b3da into main May 6, 2026
14 of 15 checks passed

openshift-merge-bot Bot deleted the cs-147-read-from-cosmos branch May 6, 2026 11:41

	if resourceID != nil && resourceID.String() == operatorIdentityResourceIDString {
	if resourceID.String() == operatorIdentityResourceIDString {

Uh oh!

Conversation

David Eads (deads2k) commented Mar 24, 2026

Uh oh!

Manyanda Chitimbo (machi1990) commented Mar 24, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

David Eads (deads2k) commented Mar 30, 2026

Uh oh!

David Eads (deads2k) commented Mar 30, 2026

Uh oh!

David Eads (deads2k) commented Mar 30, 2026

Uh oh!

David Eads (deads2k) commented Mar 31, 2026

Uh oh!

Manyanda Chitimbo (machi1990) left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Manyanda Chitimbo (machi1990) left a comment

Choose a reason for hiding this comment

Uh oh!

David Eads (deads2k) commented Mar 31, 2026

Uh oh!

David Eads (deads2k) commented Mar 31, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Miguel Soriano (miguelsorianod) commented May 5, 2026

Uh oh!

openshift-ci Bot commented May 5, 2026

openshift-ci Bot commented May 5, 2026 •

edited

Loading