SREP-4403: Add rosa-e2e nightly jobs to testgrid allow-list (#77976)

* SREP-4403: Add rosa-e2e nightly jobs to testgrid allow-list

Add rosa-e2e nightly and daily-status jobs so they get indexed
into BigQuery via testgrid, enabling Sippy to ingest test results
for the rosa-stage dashboard.

Jira: TRT-2588, SREP-4403

* SREP-4403: Use ci-status-report.sh for daily status job

Replace inline Sippy-only query with the ci-status-report.sh
script that checks Prow GCS directly and includes Sippy data
when available.

Jira: SREP-4403, TRT-2588

* SREP-4403: Update daily status to use script and enrich reporter template

Call ci-status-report.sh instead of inline Sippy query. The script
checks all tracked ROSA jobs via Prow GCS and adds Sippy data when
available. Enrich the slack_reporter_config template with links to
Sippy and the ROSA Prow jobs dashboard.

Jira: SREP-4403, TRT-2588

* Fix STS account-roles fallback to use published policy versions

The fallback was querying rosa list versions (installable cluster
versions) which is different from versions with published IAM
policies. Change to query rosa list account-roles instead, picking
the latest version that has existing policies in the AWS account.
Also add || true to prevent pipefail from crashing the script
before the fallback check runs.

Fixes Classic STS conformance failures on 4.19, 4.20, and 4.22
where nightly-channel policies aren't published in staging.

Jira: SREP-4403

* Fix testgrid allow-list: use 'informing' release_type

rosa-e2e is not a valid release_type. Valid values are: informing,
broken, generic-informing, osde2e, olm.

* Fix version sort to use sort -V for semantic ordering

jq's sort is lexicographic (4.9 > 4.10). Use sort -Vu for proper
version ordering when selecting the fallback version from existing
account roles.

Author: dustman9000

ci-operator/config/openshift-online/rosa-e2e/openshift-online-rosa-e2e-main.yaml

@@ -41,29 +41,7 @@ tests:
   container:
     from: src
 - as: rosa-ci-daily-status
-  commands: |
-    #!/bin/bash
-    set -euo pipefail
-    SIPPY_API="https://sippy.dptools.openshift.org/api/jobs"
-    RELEASE="rosa-stage"
-    echo "Querying Sippy for ${RELEASE} job results..."
-    response=$(curl -sf "${SIPPY_API}?release=${RELEASE}&limit=100" 2>/dev/null) || {
-      echo "Warning: Failed to fetch data from Sippy API"
-      response="[]"
-    }
-    total=$(echo "${response}" | jq 'length')
-    if [[ "${total}" == "0" ]]; then
-      echo "No jobs found in Sippy for ${RELEASE}"
-      exit 1
-    fi
-    passing=$(echo "${response}" | jq '[.[] | select(.current_pass_percentage >= 80)] | length')
-    failing=$((total - passing))
-    echo "ROSA CI Status: ${passing}/${total} jobs healthy, ${failing} failing"
-    echo "${response}" | jq -r '.[] | "\(.name): \(.current_pass_percentage)%"' | sort
-    if [[ "${failing}" -gt 0 ]]; then
-      echo "Some jobs are below threshold"
-      exit 1
-    fi
+  commands: scripts/ci-status-report.sh
   container:
     from: src
   cron: 0 14 * * *

ci-operator/jobs/openshift-online/rosa-e2e/openshift-online-rosa-e2e-main-periodics.yaml

@@ -343,9 +343,12 @@ periodics:
       - failure
       - error
       report_template: '{{if eq .Status.State "success"}} :white_check_mark: ROSA
-        CI Daily Status: all jobs healthy. <{{.Status.URL}}|View logs> | <https://sippy.dptools.openshift.org/sippy-ng/release/rosa-stage|Sippy>
-        {{else}} :x: ROSA CI Daily Status: some jobs failing. <{{.Status.URL}}|View
-        logs> | <https://sippy.dptools.openshift.org/sippy-ng/release/rosa-stage|Sippy>
+        CI Daily Status: all tracked jobs healthy. <{{.Status.URL}}|Full report> |
+        <https://sippy.dptools.openshift.org/sippy-ng/release/rosa-stage|Sippy> |
+        <https://prow.ci.openshift.org/?type=periodic&job=*rosa*|All ROSA jobs> {{else}}
+        :x: ROSA CI Daily Status: some jobs failing. <{{.Status.URL}}|Full report>
+        | <https://sippy.dptools.openshift.org/sippy-ng/release/rosa-stage|Sippy>
+        | <https://prow.ci.openshift.org/?type=periodic&job=*rosa*|All ROSA jobs>
         {{end}}'
   spec:
     containers:

ci-operator/step-registry/rosa/sts/account-roles/create/rosa-sts-account-roles-create-commands.sh

@@ -79,14 +79,15 @@ if [[ "$CHANNEL_GROUP" != "stable" ]]; then
   fi
 
   # Verify the version has published policies. If not, fall back to the latest
-  # available version. This handles cases where nightly builds for unreleased
-  # versions (e.g., 4.23, 5.0) don't have IAM policies published yet.
+  # version that has existing account-role policies in the AWS account.
   if ! rosa list account-roles --output json 2>/dev/null | jq -e --arg v "${OPENSHIFT_VERSION}" '.[] | select(.Version // "" | startswith($v))' > /dev/null 2>&1; then
-    echo "Version ${OPENSHIFT_VERSION} may not have published policies, checking available versions..."
-    available_version=$(rosa list versions --channel-group ${CHANNEL_GROUP} ${CLUSTER_SWITCH} -o json 2>/dev/null | jq -r '.[].raw_id' | cut -d'.' -f1,2 | sort -Vu | tail -1)
-    if [[ -n "${available_version}" ]]; then
-      echo "Falling back from ${OPENSHIFT_VERSION} to ${available_version}"
+    echo "Version ${OPENSHIFT_VERSION} may not have published policies, checking account roles for available versions..."
+    available_version=$(rosa list account-roles --output json 2>/dev/null | jq -r '.[].Version // empty' 2>/dev/null | cut -d'.' -f1,2 | sort -Vu | tail -1 || true)
+    if [[ -n "${available_version}" && "${available_version}" != "${OPENSHIFT_VERSION}" ]]; then
+      echo "Falling back from ${OPENSHIFT_VERSION} to ${available_version} (latest with published policies)"
       OPENSHIFT_VERSION="${available_version}"
+    else
+      echo "WARNING: No fallback version found in account roles, continuing with ${OPENSHIFT_VERSION}"
     fi
   fi
 

core-services/testgrid-config-generator/_allow-list.yaml

@@ -38,6 +38,10 @@ periodic-ci-openshift-osde2e-main-osd-aws-upgrade-latest-default-y-minus-1-to-la
 periodic-ci-openshift-osde2e-main-osd-aws-upgrade-latest-default-z-minus-1-to-latest-default-z: osde2e
 periodic-ci-openshift-osde2e-main-osd-aws-upgrade-latest-default-y-to-latest-y-plus-1: osde2e
 periodic-ci-openshift-osde2e-main-osd-aws-upgrade-latest-default-y-plus-1-to-latest-y: osde2e
+periodic-ci-openshift-online-rosa-e2e-main-periodics-rosa-hcp-e2e-nightly-4-20: informing
+periodic-ci-openshift-online-rosa-e2e-main-periodics-rosa-hcp-e2e-nightly-4-21: informing
+periodic-ci-openshift-online-rosa-e2e-main-periodics-rosa-hcp-e2e-nightly-4-22: informing
+periodic-ci-openshift-online-rosa-e2e-main-rosa-ci-daily-status: informing
 periodic-ci-redhat-chaos-krkn-hub-main-krkn-hub-tests: generic-informing
 periodic-ci-openshift-operator-framework-olm-master-periodics-e2e-gcp-console-olm: olm
 periodic-ci-openshift-operator-framework-olm-master-periodics-e2e-gcp-olm: olm