Fix Proxmox live chart CPU fallback

Refs #1525

docs/release-control/v6/internal/subsystems/api-contracts.md

@@ -188,6 +188,10 @@ CPU fallback points use host-capacity-normalized percent, while Docker metadata
 may expose runtime-native per-core CPU percent plus the host CPU count as raw
 diagnostic evidence. API consumers must not treat raw per-core CPU as the
 threshold, chart, or canonical resource CPU value.
+Proxmox legacy model CPU fields have the inverse compatibility shape:
+`models.Node.CPU`, `models.VM.CPU`, and `models.Container.CPU` remain 0..1
+ratios, but `/api/metrics-store/history` live fallback points must convert
+those ratios to 0..100 chart percentages before serialization.
 
 Candidate import plans for PVE/PBS/PMG onboarding are API-contract consumers,
 not local UI estimates. Probe and Discovery candidates must build their
@@ -2459,8 +2463,10 @@ a new API state machine, queue contract, or verification-accounting field.
     and `POST /api/connections/probe` stay on one canonical payload shape
     instead of re-deriving state from per-type config stores in the frontend.
     State must remain a derived field sourced from in-memory scheduler health
-    (`monitoring.Monitor.SchedulerHealth()`) plus agent `Host.LastSeen`; the
-    endpoint must not introduce new persisted per-connection state. The probe
+    (`monitoring.Monitor.SchedulerHealth()`) plus agent `Host.LastSeen`; for
+    agent rows, `Host.LastSeen` is the monitoring-owned server receipt time,
+    not the agent payload timestamp. The endpoint must not introduce new
+    persisted per-connection state. The probe
     endpoint must remain admin-gated (`RequireAdmin` + `ScopeSettingsWrite`)
     to block unauthenticated SSRF against internal hosts. That same probe path
     must also validate user-supplied addresses before probing, reject metadata,

internal/api/contract_test.go

@@ -8,6 +8,7 @@ import (
 	"encoding/base64"
 	"encoding/json"
 	"fmt"
+	"math"
 	"net/http"
 	"net/http/httptest"
 	"net/url"
@@ -10802,6 +10803,57 @@ func TestContract_MetricsHistoryLiveFallbackJSONSnapshot(t *testing.T) {
 	assertJSONSnapshot(t, got, want)
 }
 
+func TestContract_MetricsHistoryProxmoxNodeCPUFallbackConvertsRatioToPercent(t *testing.T) {
+	state := models.NewState()
+	state.UpdateNodes([]models.Node{{
+		ID:       "pve1:node1",
+		Name:     "node1",
+		Instance: "pve1",
+		Status:   "online",
+		CPU:      0.37,
+	}})
+
+	monitor := &monitoring.Monitor{}
+	setUnexportedField(t, monitor, "state", state)
+	setUnexportedField(t, monitor, "metricsHistory", monitoring.NewMetricsHistory(10, time.Hour))
+
+	tempDir := t.TempDir()
+	mtp := config.NewMultiTenantPersistence(tempDir)
+	if _, err := mtp.GetPersistence("default"); err != nil {
+		t.Fatalf("failed to init persistence: %v", err)
+	}
+
+	router := &Router{
+		monitor:         monitor,
+		licenseHandlers: NewLicenseHandlers(mtp, false),
+	}
+
+	req := httptest.NewRequest(
+		http.MethodGet,
+		"/api/metrics-store/history?resourceType=node&resourceId=pve1:node1&metric=cpu&range=24h",
+		nil,
+	)
+	rec := httptest.NewRecorder()
+	router.handleMetricsHistory(rec, req)
+	if rec.Code != http.StatusOK {
+		t.Fatalf("status = %d, body=%s", rec.Code, rec.Body.String())
+	}
+
+	var resp metricsHistoryResponse
+	if err := json.Unmarshal(rec.Body.Bytes(), &resp); err != nil {
+		t.Fatalf("unmarshal metrics history response: %v", err)
+	}
+	if resp.Source != "live" {
+		t.Fatalf("source = %q, want live", resp.Source)
+	}
+	if len(resp.Points) != 1 {
+		t.Fatalf("points = %d, want 1", len(resp.Points))
+	}
+	if math.Abs(resp.Points[0].Value-37) > 0.001 {
+		t.Fatalf("cpu fallback value = %f, want 37", resp.Points[0].Value)
+	}
+}
+
 func TestContract_MetricsHistoryDockerContainerCPUFallbackUsesCapacityPercent(t *testing.T) {
 	state := models.NewState()
 	state.UpsertDockerHost(models.DockerHost{

internal/api/router.go

@@ -7540,6 +7540,10 @@ func clampWorkloadPercent(value float64) float64 {
 	return value
 }
 
+func proxmoxModelCPURatioPercent(value float64) float64 {
+	return clampWorkloadPercent(value * 100)
+}
+
 func clampNonNegativeWorkloadValue(value float64) float64 {
 	if value != value {
 		return 0
@@ -9010,7 +9014,7 @@ func (r *Router) handleMetricsHistory(w http.ResponseWriter, req *http.Request)
 			if vm == nil {
 				return points
 			}
-			points["cpu"] = monitoring.MetricPoint{Timestamp: now, Value: vm.CPU}
+			points["cpu"] = monitoring.MetricPoint{Timestamp: now, Value: proxmoxModelCPURatioPercent(vm.CPU)}
 			points["memory"] = monitoring.MetricPoint{Timestamp: now, Value: vm.Memory.Usage}
 			if vm.Disk.Usage >= 0 {
 				points["disk"] = monitoring.MetricPoint{Timestamp: now, Value: vm.Disk.Usage}
@@ -9024,7 +9028,7 @@ func (r *Router) handleMetricsHistory(w http.ResponseWriter, req *http.Request)
 			if ct == nil {
 				return points
 			}
-			points["cpu"] = monitoring.MetricPoint{Timestamp: now, Value: ct.CPU}
+			points["cpu"] = monitoring.MetricPoint{Timestamp: now, Value: proxmoxModelCPURatioPercent(ct.CPU)}
 			points["memory"] = monitoring.MetricPoint{Timestamp: now, Value: ct.Memory.Usage}
 			if ct.Disk.Usage >= 0 {
 				points["disk"] = monitoring.MetricPoint{Timestamp: now, Value: ct.Disk.Usage}
@@ -9038,7 +9042,7 @@ func (r *Router) handleMetricsHistory(w http.ResponseWriter, req *http.Request)
 			if node == nil {
 				return points
 			}
-			points["cpu"] = monitoring.MetricPoint{Timestamp: now, Value: node.CPU}
+			points["cpu"] = monitoring.MetricPoint{Timestamp: now, Value: proxmoxModelCPURatioPercent(node.CPU)}
 			points["memory"] = monitoring.MetricPoint{Timestamp: now, Value: node.Memory.Usage}
 			points["disk"] = monitoring.MetricPoint{Timestamp: now, Value: node.Disk.Usage}
 			if temperature := primaryNodeTemperatureCelsius(node.Temperature); temperature != nil {
@@ -9093,7 +9097,7 @@ func (r *Router) handleMetricsHistory(w http.ResponseWriter, req *http.Request)
 			if node == nil {
 				return points
 			}
-			points["cpu"] = monitoring.MetricPoint{Timestamp: now, Value: node.CPU}
+			points["cpu"] = monitoring.MetricPoint{Timestamp: now, Value: proxmoxModelCPURatioPercent(node.CPU)}
 			points["memory"] = monitoring.MetricPoint{Timestamp: now, Value: node.Memory.Usage}
 			points["disk"] = monitoring.MetricPoint{Timestamp: now, Value: node.Disk.Usage}
 			if temperature := primaryNodeTemperatureCelsius(node.Temperature); temperature != nil {

scripts/release_control/subsystem_lookup_test.py

@@ -2917,7 +2917,7 @@ class SubsystemLookupTest(unittest.TestCase):
                 {
                     "heading": "## Shared Boundaries",
                     "path": "internal/api/access_control_handlers.go",
-                    "line": 1168,
+                    "line": 1172,
                     "heading_line": 137,
                 }
             ],