monitoring: tune testing dashboard and gate rollups

ci/titan-iac-trivy-waivers.json

@@ -31,6 +31,7 @@
       "id": "KSV-0014",
       "targets": [
         "infrastructure/cert-manager/cleanup/cert-manager-cleanup-job.yaml",
+        "infrastructure/core/node-prefer-noschedule-cronjob.yaml",
         "infrastructure/core/ntp-sync-daemonset.yaml",
         "infrastructure/longhorn/adopt/longhorn-helm-adopt-job.yaml",
         "infrastructure/longhorn/core/longhorn-disk-tags-ensure-job.yaml",
@@ -98,6 +99,7 @@
         "services/keycloak/oneoffs/ldap-federation-job.yaml",
         "services/keycloak/oneoffs/logs-oidc-secret-ensure-job.yaml",
         "services/keycloak/oneoffs/mas-secrets-ensure-job.yaml",
+        "services/keycloak/oneoffs/metis-node-passwords-secret-ensure-job.yaml",
         "services/keycloak/oneoffs/metis-oidc-secret-ensure-job.yaml",
         "services/keycloak/oneoffs/metis-ssh-keys-secret-ensure-job.yaml",
         "services/keycloak/oneoffs/portal-admin-client-secret-ensure-job.yaml",
@@ -152,6 +154,7 @@
         "services/monitoring/platform-quality-gateway-deployment.yaml",
         "services/monitoring/platform-quality-suite-probe-cronjob.yaml",
         "services/monitoring/postmark-exporter-deployment.yaml",
+        "services/monitoring/vmalert-atlas-availability.yaml",
         "services/monitoring/vault-sync-deployment.yaml",
         "services/nextcloud-mail-sync/cronjob.yaml",
         "services/nextcloud/collabora.yaml",
@@ -247,6 +250,7 @@
       "targets": [
         "infrastructure/cert-manager/cleanup/cert-manager-cleanup-job.yaml",
         "infrastructure/core/coredns-deployment.yaml",
+        "infrastructure/core/node-prefer-noschedule-cronjob.yaml",
         "infrastructure/core/ntp-sync-daemonset.yaml",
         "infrastructure/longhorn/adopt/longhorn-helm-adopt-job.yaml",
         "infrastructure/longhorn/core/longhorn-disk-tags-ensure-job.yaml",
@@ -310,6 +314,7 @@
         "services/keycloak/oneoffs/ldap-federation-job.yaml",
         "services/keycloak/oneoffs/logs-oidc-secret-ensure-job.yaml",
         "services/keycloak/oneoffs/mas-secrets-ensure-job.yaml",
+        "services/keycloak/oneoffs/metis-node-passwords-secret-ensure-job.yaml",
         "services/keycloak/oneoffs/metis-oidc-secret-ensure-job.yaml",
         "services/keycloak/oneoffs/metis-ssh-keys-secret-ensure-job.yaml",
         "services/keycloak/oneoffs/portal-admin-client-secret-ensure-job.yaml",
@@ -364,6 +369,7 @@
         "services/monitoring/platform-quality-gateway-deployment.yaml",
         "services/monitoring/platform-quality-suite-probe-cronjob.yaml",
         "services/monitoring/postmark-exporter-deployment.yaml",
+        "services/monitoring/vmalert-atlas-availability.yaml",
         "services/monitoring/vault-sync-deployment.yaml",
         "services/nextcloud/collabora.yaml",
         "services/oauth2-proxy/deployment.yaml",

scripts/dashboards_render_atlas.py

@@ -3346,9 +3346,10 @@ def build_jobs_dashboard():
         f'/ clamp_min(sum by (suite) (max by (suite, check) (({{{checks_selector}}} > bool 0))), 1)) '
         f'or on(suite) ({selected_suite_zero})'
     )
+    success_history_runs = f'sum by (suite) (increase(platform_quality_gate_runs_total{{{runs_selector}}}[$__interval]))'
     success_history_by_suite = (
-        f'100 * (sum by (suite) (increase(platform_quality_gate_runs_total{{{runs_success_selector}}}[$__interval])) '
-        f'/ clamp_min((sum by (suite) (increase(platform_quality_gate_runs_total{{{runs_selector}}}[$__interval]))), 1))'
+        f'(100 * sum by (suite) (increase(platform_quality_gate_runs_total{{{runs_success_selector}}}[$__interval])) '
+        f'/ ({success_history_runs})) and on(suite) (({success_history_runs}) > 0)'
     )
     coverage_by_suite = (
         f'(max by (suite) ({{{coverage_metric_selector}}})) '
@@ -3357,7 +3358,6 @@ def build_jobs_dashboard():
     coverage_with_missing = (
         f"({coverage_by_suite}) or on(suite) ((0 * (sum by (suite) (increase(platform_quality_gate_runs_total{{{runs_selector}}}[30d])))) - 1)"
     )
-    coverage_gap = f"clamp_min(95 - ({coverage_by_suite}), 0)"
     smell_by_suite = f'max by (suite) (platform_quality_gate_source_lines_over_500_total{{{smell_selector}}})'
     loc_files_by_suite = f'max by (suite) (platform_quality_gate_source_files_total{{{smell_selector}}})'
     smell_with_missing = (
@@ -3384,6 +3384,14 @@ def build_jobs_dashboard():
         f"or on(suite) (100 * (1 - clamp_max(({loc_violations_history}), 1))) "
         f"or on(suite) ({selected_suite_zero})"
     )
+    coverage_loc_compliance_history = (
+        "min by (suite) ("
+        f'label_replace(max_over_time(platform_quality_gate_workspace_line_coverage_percent{{{workspace_coverage_selector}}}[$__interval]), '
+        '"dimension", "coverage", "__name__", ".*") '
+        "or "
+        f'label_replace(({loc_limit_compliance_history}), "dimension", "loc", "__name__", ".*")'
+        ")"
+    )
     average_coverage = f"(avg(({coverage_by_suite})) or on() vector(0))"
     suites_loc_violating = f'(sum((({smell_by_suite}) > bool 0)) or on() vector(0))'
 
@@ -3397,39 +3405,36 @@ def build_jobs_dashboard():
 
     def _check_state_series(regex: str, failed: bool) -> str:
         state = f'result!~"{non_failure}"' if failed else f'result=~"{success}"'
-        core = (
-            f'sum by (suite) (max_over_time(({{{checks_selector},check=~"{regex}",{state}}})[$__interval]))'
-        )
+        core = f'sum by (suite) (increase({{{checks_selector},check=~"{regex}",{state}}}[$__interval]))'
         return f'({core}) or on(suite) ({selected_suite_zero})'
 
     problematic_tests_history_core = (
-        f'topk(12, sum by (suite, test, jenkins_job) (increase(platform_quality_gate_test_case_result{{suite=~"{suite_var}",branch=~"{branch_var}",test!="__no_test_cases__",status="failed",{exported}}}[$__interval])))'
+        f'topk(12, sum by (suite, test) (platform_quality:test_case_status:count_1h{{suite=~"{suite_var}",branch=~"{branch_var}",test!="__no_test_cases__",status="failed"}}))'
     )
     problematic_tests_history = f"({problematic_tests_history_core}) or on() vector(0)"
     worst_test_per_suite_core = (
-        f'topk by (suite) (1, sum by (suite, test, jenkins_job) (increase(platform_quality_gate_test_case_result{{suite=~"{suite_var}",branch=~"{branch_var}",test!="__no_test_cases__",status="failed",{exported}}}[30d])))'
+        f'topk by (suite) (1, sum by (suite, test) (increase(platform_quality_gate_test_case_result{{suite=~"{suite_var}",branch=~"{branch_var}",test!="__no_test_cases__",status="failed",{exported}}}[30d])))'
     )
     worst_test_per_suite = f"({worst_test_per_suite_core}) or on() vector(0)"
     selected_test_pass_fail = [
         {
             "refId": "A",
-            "expr": f'sum by (suite, test, status, jenkins_job, build_number) (max_over_time(platform_quality_gate_test_case_result{{{test_case_selector},status="passed"}}[$__interval])) or on() vector(0)',
-            "legendFormat": "passed · {{suite}} · #{{build_number}}",
+            "expr": f'sum by (suite) (platform_quality:test_case_status:count_1h{{suite=~"{suite_var}",branch=~"{branch_var}",test=~"{test_var}",test!="__no_test_cases__",status="passed"}}) or on() vector(0)',
+            "legendFormat": "{{suite}} passed",
         },
         {
             "refId": "B",
-            "expr": f'sum by (suite, test, status, jenkins_job, build_number) (max_over_time(platform_quality_gate_test_case_result{{{test_case_selector},status="failed"}}[$__interval])) or on() vector(0)',
-            "legendFormat": "failed · {{suite}} · #{{build_number}}",
+            "expr": f'sum by (suite) (platform_quality:test_case_status:count_1h{{suite=~"{suite_var}",branch=~"{branch_var}",test=~"{test_var}",test!="__no_test_cases__",status="failed"}}) or on() vector(0)',
+            "legendFormat": "{{suite}} failed",
         },
         {
             "refId": "C",
-            "expr": f'sum by (suite, test, status, jenkins_job, build_number) (max_over_time(platform_quality_gate_test_case_result{{{test_case_selector},status="skipped"}}[$__interval])) or on() vector(0)',
-            "legendFormat": "skipped · {{suite}} · #{{build_number}}",
+            "expr": f'sum by (suite) (platform_quality:test_case_status:count_1h{{suite=~"{suite_var}",branch=~"{branch_var}",test=~"{test_var}",test!="__no_test_cases__",status="skipped"}}) or on() vector(0)',
+            "legendFormat": "{{suite}} skipped",
         },
     ]
     selected_test_pass_rate = (
-        f'100 * (sum by (suite, test) (max_over_time(platform_quality_gate_test_case_result{{{test_case_selector},status="passed"}}[$__interval]))) '
-        f'/ clamp_min((sum by (suite, test) (max_over_time(platform_quality_gate_test_case_result{{{test_case_selector},status=~"passed|failed|error|skipped"}}[$__interval]))), 1)'
+        f'avg by (suite) (platform_quality:test_case_pass_rate:percent_1h{{suite=~"{suite_var}",branch=~"{branch_var}",test=~"{test_var}",test!="__no_test_cases__"}})'
     )
     recent_branch_evidence = (
         f'sort_desc(count by (suite, branch) (max_over_time(platform_quality_gate_build_info{{{build_info_selector}}}[30d])))'
@@ -3442,70 +3447,70 @@ def build_jobs_dashboard():
         missing = f"(({suite_universe}) unless on(suite) {presence_expr})"
         return f"({missing}) or on(suite) (0 * ({suite_universe}))"
 
-    missing_tests_by_suite = _missing_suite_series(
+    def _present_suite_percent(presence_expr: str) -> str:
+        present = f"(({suite_universe}) and on(suite) {presence_expr})"
+        return f"(100 * ({present})) or on(suite) (0 * ({suite_universe}))"
+
+    present_tests_by_suite = _present_suite_percent(
         f'count by (suite) ({{__name__=~".*_quality_gate_tests_total",{exported}}})'
     )
-    missing_checks_by_suite = _missing_suite_series(
+    present_checks_by_suite = _present_suite_percent(
         f'count by (suite) ({{__name__=~".*_quality_gate_checks_total",{exported}}})'
     )
-    missing_coverage_by_suite = _missing_suite_series(
+    present_coverage_by_suite = _present_suite_percent(
         f"count by (suite) (platform_quality_gate_workspace_line_coverage_percent{{{exported}}})"
     )
-    missing_loc_by_suite = _missing_suite_series(
+    present_loc_by_suite = _present_suite_percent(
         f"count by (suite) (platform_quality_gate_source_lines_over_500_total{{{exported}}}) "
         f"and on(suite) count by (suite) (platform_quality_gate_source_files_total{{{exported}}})"
     )
-    missing_test_case_by_suite = _missing_suite_series(
+    present_test_case_by_suite = _present_suite_percent(
         f"count by (suite) (platform_quality_gate_test_case_result{{{exported}}})"
     )
-    placeholder_test_case_by_suite = _missing_suite_series(
+    real_test_case_by_suite = _present_suite_percent(
         f'count by (suite) (platform_quality_gate_test_case_result{{{exported},test!="__no_test_cases__"}})'
     )
 
+    dark_red = "dark-red"
+    dark_orange = "dark-orange"
+    dark_yellow = "dark-yellow"
+    dark_green = "dark-green"
+    dark_blue = "dark-blue"
     success_thresholds = {
         "mode": "absolute",
         "steps": [
-            {"color": "red", "value": None},
-            {"color": "orange", "value": 90},
-            {"color": "yellow", "value": 93},
-            {"color": "green", "value": 95},
-            {"color": "blue", "value": 100},
+            {"color": dark_red, "value": None},
+            {"color": dark_orange, "value": 90},
+            {"color": dark_yellow, "value": 93},
+            {"color": dark_green, "value": 95},
+            {"color": dark_blue, "value": 100},
         ],
     }
     coverage_thresholds = success_thresholds
     failures_thresholds = {
         "mode": "absolute",
         "steps": [
-            {"color": "green", "value": None},
-            {"color": "yellow", "value": 1},
-            {"color": "orange", "value": 3},
-            {"color": "red", "value": 5},
-        ],
-    }
-    coverage_gap_thresholds = {
-        "mode": "absolute",
-        "steps": [
-            {"color": "green", "value": None},
-            {"color": "yellow", "value": 1},
-            {"color": "orange", "value": 5},
-            {"color": "red", "value": 10},
+            {"color": dark_green, "value": None},
+            {"color": dark_yellow, "value": 1},
+            {"color": dark_orange, "value": 3},
+            {"color": dark_red, "value": 5},
         ],
     }
     smell_thresholds = {
         "mode": "absolute",
         "steps": [
-            {"color": "red", "value": None},
-            {"color": "green", "value": 0},
-            {"color": "yellow", "value": 1},
-            {"color": "orange", "value": 3},
-            {"color": "red", "value": 5},
+            {"color": dark_red, "value": None},
+            {"color": dark_green, "value": 0},
+            {"color": dark_yellow, "value": 1},
+            {"color": dark_orange, "value": 3},
+            {"color": dark_red, "value": 5},
         ],
     }
     missing_thresholds = {
         "mode": "absolute",
         "steps": [
-            {"color": "green", "value": None},
-            {"color": "red", "value": 1},
+            {"color": dark_green, "value": None},
+            {"color": dark_red, "value": 1},
         ],
     }
 
@@ -3554,7 +3559,7 @@ def build_jobs_dashboard():
             instant=True,
             thresholds={
                 "mode": "absolute",
-                "steps": [{"color": "red", "value": None}, {"color": "green", "value": 1}],
+                "steps": [{"color": dark_red, "value": None}, {"color": dark_green, "value": 1}],
             },
         )
     )
@@ -3620,21 +3625,6 @@ def build_jobs_dashboard():
         {"type": "value", "options": {"-1": {"text": "no runs"}}}
     ]
     panels.append(reliability_suite_panel)
-    coverage_gap_panel = bargauge_panel(
-        10,
-        "Coverage Gap to 95% by Suite",
-        coverage_gap,
-        {"h": 8, "w": 8, "x": 16, "y": 5},
-        unit="percent",
-        instant=True,
-        legend="{{suite}}",
-        sort_order="desc",
-        thresholds=coverage_gap_thresholds,
-        decimals=2,
-    )
-    coverage_gap_panel["description"] = "Gap from the 95% target. 0 means the suite is at or above target."
-    panels.append(coverage_gap_panel)
-
     history_panel = timeseries_panel(
         11,
         "Run Reliability History by Suite",
@@ -3676,15 +3666,10 @@ def build_jobs_dashboard():
                     "expr": f'sum(increase(platform_quality_gate_runs_total{{{runs_failure_selector}}}[$__interval])) or on() vector(0)',
                     "legendFormat": "Failure",
                 },
-                {
-                    "refId": "C",
-                    "expr": f'sum(increase(platform_quality_gate_runs_total{{{runs_selector}}}[$__interval])) or on() vector(0)',
-                    "legendFormat": "Total",
-                },
             ],
             legend_display="list",
             legend_placement="bottom",
-            legend_calcs=["lastNotNull", "sum"],
+            legend_calcs=[],
         )
     )
     panels.append(
@@ -3697,18 +3682,13 @@ def build_jobs_dashboard():
             targets=[
                 {
                     "refId": "A",
-                    "expr": f'max_over_time(platform_quality_gate_workspace_line_coverage_percent{{{workspace_coverage_selector}}}[$__interval])',
-                    "legendFormat": "{{suite}} coverage %",
-                },
-                {
-                    "refId": "B",
-                    "expr": loc_limit_compliance_history,
-                    "legendFormat": "{{suite}} files <=500 LOC %",
+                    "expr": coverage_loc_compliance_history,
+                    "legendFormat": "{{suite}}",
                 },
             ],
             legend_display="list",
             legend_placement="bottom",
-            legend_calcs=["lastNotNull", "max"],
+            legend_calcs=[],
         )
     )
     run_mix_panel = pie_panel(
@@ -3742,12 +3722,14 @@ def build_jobs_dashboard():
                 legend="{{suite}}",
                 legend_display="list",
                 legend_placement="bottom",
-                legend_calcs=["lastNotNull", "max"],
+                legend_calcs=[],
             )
             panel["description"] = (
-                "One line per selected suite. 1 means this check dimension was in that state during the bucket; "
-                "0 means the suite reported the dimension and it was not in that state."
+                "One line per selected suite, counting check events in this state during each bucket. "
+                "Intervals without runs stay at zero rather than being treated as failures."
             )
+            panel["fieldConfig"]["defaults"]["min"] = 0
+            panel["fieldConfig"]["defaults"].setdefault("custom", {})["spanNulls"] = True
             panels.append(panel)
         for index, (label, regex) in enumerate(check_dimensions[4:]):
             panel = timeseries_panel(
@@ -3759,12 +3741,14 @@ def build_jobs_dashboard():
                 legend="{{suite}}",
                 legend_display="list",
                 legend_placement="bottom",
-                legend_calcs=["lastNotNull", "max"],
+                legend_calcs=[],
             )
             panel["description"] = (
-                "One line per selected suite. 1 means this check dimension was in that state during the bucket; "
-                "0 means the suite reported the dimension and it was not in that state."
+                "One line per selected suite, counting check events in this state during each bucket. "
+                "Intervals without runs stay at zero rather than being treated as failures."
             )
+            panel["fieldConfig"]["defaults"]["min"] = 0
+            panel["fieldConfig"]["defaults"].setdefault("custom", {})["spanNulls"] = True
             panels.append(panel)
 
     _append_check_trends(130, "Failure Trend", True, 29)
@@ -3776,10 +3760,10 @@ def build_jobs_dashboard():
             problematic_tests_history,
             {"h": 8, "w": 12, "x": 0, "y": 57},
             unit="none",
-            legend="{{suite}} · {{test}}",
+            legend="{{suite}} - {{test}}",
             legend_display="list",
-            legend_placement="bottom",
-            legend_calcs=["lastNotNull", "max", "sum"],
+            legend_placement="right",
+            legend_calcs=[],
             links=jenkins_suite_links(),
             data_links=jenkins_latest_artifact_data_links(),
         )
@@ -3810,7 +3794,7 @@ def build_jobs_dashboard():
             targets=selected_test_pass_fail,
             legend_display="list",
             legend_placement="bottom",
-            legend_calcs=["lastNotNull", "sum"],
+            legend_calcs=[],
             links=jenkins_suite_links(),
             data_links=jenkins_artifact_data_links(),
         )
@@ -3821,10 +3805,10 @@ def build_jobs_dashboard():
         selected_test_pass_rate,
         {"h": 8, "w": 12, "x": 12, "y": 65},
         unit="percent",
-        legend="{{suite}} · {{test}}",
+        legend="{{suite}}",
         legend_display="list",
         legend_placement="bottom",
-        legend_calcs=["lastNotNull", "min"],
+        legend_calcs=[],
         links=jenkins_suite_links(),
         data_links=jenkins_artifact_data_links(),
     )
@@ -3871,56 +3855,56 @@ def build_jobs_dashboard():
     panels.append(
         bargauge_panel(
             27,
-            "Missing Tests Metrics by Suite",
-            missing_tests_by_suite,
+            "Tests Metrics Present by Suite",
+            present_tests_by_suite,
             {"h": 7, "w": 6, "x": 0, "y": 81},
-            unit="none",
+            unit="percent",
             instant=True,
             legend="{{suite}}",
             sort_order="desc",
-            thresholds=missing_thresholds,
+            thresholds=success_thresholds,
             decimals=0,
         )
     )
     panels.append(
         bargauge_panel(
             28,
-            "Missing Checks Metrics by Suite",
-            missing_checks_by_suite,
+            "Checks Metrics Present by Suite",
+            present_checks_by_suite,
             {"h": 7, "w": 6, "x": 6, "y": 81},
-            unit="none",
+            unit="percent",
             instant=True,
             legend="{{suite}}",
             sort_order="desc",
-            thresholds=missing_thresholds,
+            thresholds=success_thresholds,
             decimals=0,
         )
     )
     panels.append(
         bargauge_panel(
             29,
-            "Missing Coverage Metrics by Suite",
-            missing_coverage_by_suite,
+            "Coverage Metrics Present by Suite",
+            present_coverage_by_suite,
             {"h": 7, "w": 6, "x": 12, "y": 81},
-            unit="none",
+            unit="percent",
             instant=True,
             legend="{{suite}}",
             sort_order="desc",
-            thresholds=missing_thresholds,
+            thresholds=success_thresholds,
             decimals=0,
         )
     )
     panels.append(
         bargauge_panel(
             30,
-            "Missing LOC Compliance Metrics by Suite",
-            missing_loc_by_suite,
+            "LOC Compliance Metrics Present by Suite",
+            present_loc_by_suite,
             {"h": 7, "w": 6, "x": 18, "y": 81},
-            unit="none",
+            unit="percent",
             instant=True,
             legend="{{suite}}",
             sort_order="desc",
-            thresholds=missing_thresholds,
+            thresholds=success_thresholds,
             decimals=0,
         )
     )
@@ -3935,8 +3919,8 @@ def build_jobs_dashboard():
             thresholds={
                 "mode": "absolute",
                 "steps": [
-                    {"color": "red", "value": None},
-                    {"color": "green", "value": 1},
+                    {"color": dark_red, "value": None},
+                    {"color": dark_green, "value": 1},
                 ],
             },
         )
@@ -3988,28 +3972,28 @@ def build_jobs_dashboard():
     panels.append(
         bargauge_panel(
             148,
-            "Missing Test-Case Metrics by Suite",
-            missing_test_case_by_suite,
+            "Test-Case Metrics Present by Suite",
+            present_test_case_by_suite,
             {"h": 6, "w": 12, "x": 0, "y": 94},
-            unit="none",
+            unit="percent",
             instant=True,
             legend="{{suite}}",
             sort_order="desc",
-            thresholds=missing_thresholds,
+            thresholds=success_thresholds,
             decimals=0,
         )
     )
     panels.append(
         bargauge_panel(
             151,
-            "No Real Test Cases by Suite",
-            placeholder_test_case_by_suite,
+            "Real Test Cases Present by Suite",
+            real_test_case_by_suite,
             {"h": 6, "w": 12, "x": 12, "y": 94},
-            unit="none",
+            unit="percent",
             instant=True,
             legend="{{suite}}",
             sort_order="desc",
-            thresholds=missing_thresholds,
+            thresholds=success_thresholds,
             decimals=0,
         )
     )
@@ -4055,9 +4039,8 @@ def build_jobs_dashboard():
         5: {"h": 4, "w": 4, "x": 12, "y": 0},
         6: {"h": 4, "w": 4, "x": 16, "y": 0},
         7: {"h": 4, "w": 4, "x": 20, "y": 0},
-        8: {"h": 7, "w": 8, "x": 0, "y": 4},
-        9: {"h": 7, "w": 8, "x": 8, "y": 4},
-        10: {"h": 7, "w": 8, "x": 16, "y": 4},
+        8: {"h": 7, "w": 12, "x": 0, "y": 4},
+        9: {"h": 7, "w": 12, "x": 12, "y": 4},
         17: {"h": 7, "w": 12, "x": 0, "y": 11},
         18: {"h": 7, "w": 12, "x": 12, "y": 11},
     }
@@ -4070,6 +4053,24 @@ def build_jobs_dashboard():
     def children(ids):
         return [panel_by_id[panel_id] for panel_id in ids]
 
+    row_layout = {
+        11: {"h": 8, "w": 12, "x": 0, "y": 19},
+        12: {"h": 8, "w": 12, "x": 12, "y": 19},
+        13: {"h": 8, "w": 12, "x": 0, "y": 27},
+        14: {"h": 8, "w": 12, "x": 12, "y": 27},
+        145: {"h": 10, "w": 24, "x": 0, "y": 63},
+        147: {"h": 8, "w": 24, "x": 0, "y": 74},
+        146: {"h": 8, "w": 12, "x": 0, "y": 83},
+        152: {"h": 8, "w": 12, "x": 12, "y": 83},
+        31: {"h": 6, "w": 4, "x": 0, "y": 111},
+        32: {"h": 6, "w": 4, "x": 4, "y": 111},
+        33: {"h": 6, "w": 4, "x": 8, "y": 111},
+        34: {"h": 6, "w": 6, "x": 12, "y": 111},
+        35: {"h": 6, "w": 6, "x": 18, "y": 111},
+    }
+    for panel_id, grid in row_layout.items():
+        panel_by_id[panel_id]["gridPos"] = grid
+
     compact_panels.extend(
         [
             row_panel(500, "Reliability And Run History", 18, panels=children([11, 12, 13, 14])),
@@ -4093,9 +4094,15 @@ def build_jobs_dashboard():
             ),
             row_panel(
                 504,
-                "Telemetry Completeness, SonarQube, And Branches",
+                "Telemetry Completeness And Branches",
                 22,
-                panels=children([27, 28, 29, 30, 31, 32, 33, 34, 35, 148, 151, 149, 150]),
+                panels=children([27, 28, 29, 30, 148, 151, 149, 150]),
+            ),
+            row_panel(
+                505,
+                "SonarQube Project Health",
+                23,
+                panels=children([31, 32, 33, 34, 35]),
             ),
         ]
     )

scripts/tests/test_dashboards_render_atlas.py

@@ -133,8 +133,8 @@ def test_jobs_dashboard_bar_gauges_use_solid_threshold_colors():
     )
     threshold_steps = reliability_panel["fieldConfig"]["defaults"]["thresholds"]["steps"]
 
-    assert {"color": "yellow", "value": 93} in threshold_steps
-    assert {"color": "blue", "value": 100} in threshold_steps
+    assert {"color": "dark-yellow", "value": 93} in threshold_steps
+    assert {"color": "dark-blue", "value": 100} in threshold_steps
 
 
 def test_jobs_dashboard_collapses_heavy_drilldowns_for_light_first_paint():
@@ -150,19 +150,24 @@ def test_jobs_dashboard_collapses_heavy_drilldowns_for_light_first_paint():
     }
 
     assert len(panels) == 16
-    assert len(visible_query_panels) == 11
-    assert sum(len(panel.get("targets", [])) for panel in visible_query_panels) == 11
+    assert len(visible_query_panels) == 10
+    assert sum(len(panel.get("targets", [])) for panel in visible_query_panels) == 10
+    assert all(
+        panel["title"] != "Coverage Gap to 95% by Suite"
+        for panel in visible_query_panels
+    )
     assert [row["title"] for row in rows] == [
         "Reliability And Run History",
         "Failure Trends By Check",
         "Success Trends By Check",
         "Test Drilldowns And Problem Tests",
-        "Telemetry Completeness, SonarQube, And Branches",
+        "Telemetry Completeness And Branches",
+        "SonarQube Project Health",
     ]
     assert all(row["collapsed"] for row in rows)
 
     assert "Failure Trend: Coverage" in nested_panels_by_title
     assert "Success Trend: Supply Chain" in nested_panels_by_title
     assert "Selected Test Pass Rate History" in nested_panels_by_title
-    assert "Missing Coverage Metrics by Suite" in nested_panels_by_title
+    assert "Coverage Metrics Present by Suite" in nested_panels_by_title
     assert "SonarQube API Up" in nested_panels_by_title

services/monitoring/vmalert-atlas-availability.yaml

@@ -139,6 +139,34 @@ data:
             labels:
               scope: atlas
               rollup: yearly
+  platform-quality.yaml: |
+    groups:
+      - name: platform.quality
+        interval: 15m
+        rules:
+          - record: platform_quality:test_case_status:count_1h
+            expr: |
+              sum by (suite, branch, test, status) (
+                max_over_time(platform_quality_gate_test_case_result{job="platform-quality-ci",test!="__no_test_cases__"}[1h])
+              )
+            labels:
+              rollup: hourly
+          - record: platform_quality:test_case_pass_rate:percent_1h
+            expr: |
+              100 * (
+                sum by (suite, branch, test) (
+                  max_over_time(platform_quality_gate_test_case_result{job="platform-quality-ci",test!="__no_test_cases__",status="passed"}[1h])
+                )
+              )
+              /
+              clamp_min(
+                sum by (suite, branch, test) (
+                  max_over_time(platform_quality_gate_test_case_result{job="platform-quality-ci",test!="__no_test_cases__",status=~"passed|failed|error|skipped"}[1h])
+                ),
+                1
+              )
+            labels:
+              rollup: hourly
 
 ---
 
@@ -168,7 +196,7 @@ spec:
       labels:
         app: vmalert-atlas-availability
       annotations:
-        bstein.dev/rules-revision: "2026-05-10-availability-rollup-v6"
+        bstein.dev/rules-revision: "2026-05-15-platform-quality-rollups-v1"
     spec:
       serviceAccountName: vmalert-atlas-availability
       affinity: