refactor(redfish): enhance power reading with PowerSubsystem API support

This refactoring improves reliability by supporting both new and legacy
Redfish implementations. This commit adds support for PowerSubsystem API
and uses deprecated/legacy Power API as fallback for old BMC

Ref:
https://dmtf.org/sites/default/files/standards/documents/DSP2046_2025.2.html#power-173-deprecated

Key Changes:
- Add PowerSubsystem API support as primary power reading method
- Fallback to deprecated Power API when needed
- Move client lifecycle management from Service to PowerReader
- Strategy pattern for determining optimal power reading approach
- redfish Init() fails if it can't create a gofish client that connects
  to BMC

Test changes:
- Reorganize test utilities into dedicated testutil package
- Add fallback scenario test
- JSON fixtures for PowerSubsystem and PowerSupplies responses
- Mock server handles configurable response scenarios
- Add validation helpers for power data verification

Signed-off-by: Sunil Thaha <[email protected]>

Author: Sunil Thaha

go.mod

@@ -14,7 +14,7 @@ require (
 	github.com/prometheus/client_model v0.6.1
 	github.com/prometheus/exporter-toolkit v0.14.0
 	github.com/prometheus/procfs v0.15.1
-	github.com/stmcginnis/gofish v0.15.0
+	github.com/stmcginnis/gofish v0.20.0
 	github.com/stretchr/testify v1.10.0
 	go.uber.org/zap v1.26.0
 	golang.org/x/sync v0.12.0

go.sum

@@ -135,8 +135,8 @@ github.com/rogpeppe/go-internal v1.12.0 h1:exVL4IDcn6na9z1rAb56Vxr+CgyK3nn3O+epU
 github.com/rogpeppe/go-internal v1.12.0/go.mod h1:E+RYuTGaKKdloAfM02xzb0FW3Paa99yedzYV+kq4uf4=
 github.com/spf13/pflag v1.0.5 h1:iy+VFUOCP1a+8yFto/drg2CJ5u0yRoB7fZw3DKv/JXA=
 github.com/spf13/pflag v1.0.5/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
-github.com/stmcginnis/gofish v0.15.0 h1:8TG41+lvJk/0Nf8CIIYErxbMlQUy80W0JFRZP3Ld82A=
-github.com/stmcginnis/gofish v0.15.0/go.mod h1:BLDSFTp8pDlf/xDbLZa+F7f7eW0E/CHCboggsu8CznI=
+github.com/stmcginnis/gofish v0.20.0 h1:hH2V2Qe898F2wWT1loApnkDUrXXiLKqbSlMaH3Y1n08=
+github.com/stmcginnis/gofish v0.20.0/go.mod h1:PzF5i8ecRG9A2ol8XT64npKUunyraJ+7t0kYMpQAtqU=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
 github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
 github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=

hack/gen-metric-docs/main.go

@@ -66,24 +66,27 @@ func (m *MockRedfishService) Power() (*redfish.PowerReading, error) {
 				ID: "System.Embedded.1",
 				Readings: []redfish.Reading{
 					{
-						ControlID: "PC1",
-						Name:      "System Power Control",
-						Power:     245.0 * device.Watt, // Dell 245W scenario
+						SourceID:   "PC1",
+						SourceName: "System Power Control",
+						SourceType: redfish.PowerControlSource,
+						Power:      245.0 * device.Watt, // Dell 245W scenario
 					},
 				},
 			},
 			{
 				ID: "Enclosure.Internal.0-1",
 				Readings: []redfish.Reading{
 					{
-						ControlID: "PC1",
-						Name:      "Enclosure Power Control",
-						Power:     189.5 * device.Watt, // HPE 189W scenario
+						SourceID:   "PC1",
+						SourceName: "Enclosure Power Control",
+						SourceType: redfish.PowerControlSource,
+						Power:      189.5 * device.Watt, // HPE 189W scenario
 					},
 					{
-						ControlID: "PC2",
-						Name:      "CPU Sub-system Power",
-						Power:     167.8 * device.Watt, // Lenovo 167W scenario
+						SourceID:   "PC2",
+						SourceName: "CPU Sub-system Power",
+						SourceType: redfish.PowerControlSource,
+						Power:      167.8 * device.Watt, // Lenovo 167W scenario
 					},
 				},
 			},

internal/exporter/prometheus/collector/platform_collector.go

@@ -53,8 +53,8 @@ func NewRedfishCollector(redfish RedfishDataProvider, logger *slog.Logger) *Plat
 		bmcID:    redfish.BMCID(),
 		wattsDesc: prometheus.NewDesc(
 			prometheus.BuildFQName(keplerNS, platformSubsystem, "watts"),
-			"Current platform power consumption in watts from BMC PowerControl entries",
-			[]string{"source", "node_name", "bmc_id", "chassis_id", "power_control_id", "power_control_name"},
+			"Current platform power in watts from BMC (PowerSubsystem or deprecated Power API)",
+			[]string{"source", "node_name", "bmc_id", "chassis_id", "source_id", "source_name", "source_type"},
 			nil,
 		),
 	}
@@ -80,13 +80,13 @@ func (c *PlatformCollector) Collect(ch chan<- prometheus.Metric) {
 		return
 	}
 
-	// Emit metrics for each PowerControl reading in each chassis
+	// Emit metrics for each power reading in each chassis (PowerSupply or PowerControl)
 	for _, chassis := range powerReading.Chassis {
 		for _, reading := range chassis.Readings {
-			// Label order must match the descriptor: source, node_name, bmc_id, chassis_id, power_control_id, power_control_name
-			labels := []string{"redfish", c.nodeName, c.bmcID, chassis.ID, reading.ControlID, reading.Name}
+			// Label order must match the descriptor: source, node_name, bmc_id, chassis_id, source_id, source_name, source_type
+			labels := []string{"redfish", c.nodeName, c.bmcID, chassis.ID, reading.SourceID, reading.SourceName, string(reading.SourceType)}
 
-			// Emit current power consumption metric (power-only approach)
+			// Emit current power metric (output from PowerSupply or consumption from PowerControl)
 			ch <- prometheus.MustNewConstMetric(
 				c.wattsDesc,
 				prometheus.GaugeValue,
@@ -98,8 +98,9 @@ func (c *PlatformCollector) Collect(ch chan<- prometheus.Metric) {
 				"node.name", c.nodeName,
 				"bmc.id", c.bmcID,
 				"chassis.id", chassis.ID,
-				"power_control.id", reading.ControlID,
-				"power_control.name", reading.Name,
+				"source.id", reading.SourceID,
+				"source.name", reading.SourceName,
+				"source.type", reading.SourceType,
 				"power.watts", reading.Power,
 				"age", time.Since(powerReading.Timestamp).Seconds())
 		}

internal/exporter/prometheus/collector/platform_collector_test.go

@@ -169,24 +169,24 @@ func TestPlatformCollector_Collect_Success(t *testing.T) {
 				ID: "System.Embedded.1",
 				Readings: []redfish.Reading{
 					{
-						ControlID: "PC1",
-						Name:      "Server Power Control",
-						Power:     450.5 * device.Watt,
+						SourceID:   "PC1",
+						SourceName: "Server Power Control",
+						Power:      450.5 * device.Watt,
 					},
 					{
-						ControlID: "PC2",
-						Name:      "CPU Sub-system Power",
-						Power:     85.2 * device.Watt,
+						SourceID:   "PC2",
+						SourceName: "CPU Sub-system Power",
+						Power:      85.2 * device.Watt,
 					},
 				},
 			},
 			{
 				ID: "Enclosure.Internal.0-1",
 				Readings: []redfish.Reading{
 					{
-						ControlID: "PC1",
-						Name:      "Enclosure Power Control",
-						Power:     125.3 * device.Watt,
+						SourceID:   "PC1",
+						SourceName: "Enclosure Power Control",
+						Power:      125.3 * device.Watt,
 					},
 				},
 			},
@@ -220,34 +220,34 @@ func TestPlatformCollector_Collect_Success(t *testing.T) {
 
 	// Verify first chassis, first PowerControl metric
 	chassis1PC1Value := findMetricValue(t, platformMetric, map[string]string{
-		"source":             "redfish",
-		"node_name":          "worker-1",
-		"bmc_id":             "bmc-1",
-		"chassis_id":         "System.Embedded.1",
-		"power_control_id":   "PC1",
-		"power_control_name": "Server Power Control",
+		"source":      "redfish",
+		"node_name":   "worker-1",
+		"bmc_id":      "bmc-1",
+		"chassis_id":  "System.Embedded.1",
+		"source_id":   "PC1",
+		"source_name": "Server Power Control",
 	})
 	assert.Equal(t, 450.5, chassis1PC1Value)
 
 	// Verify first chassis, second PowerControl metric
 	chassis1PC2Value := findMetricValue(t, platformMetric, map[string]string{
-		"source":             "redfish",
-		"node_name":          "worker-1",
-		"bmc_id":             "bmc-1",
-		"chassis_id":         "System.Embedded.1",
-		"power_control_id":   "PC2",
-		"power_control_name": "CPU Sub-system Power",
+		"source":      "redfish",
+		"node_name":   "worker-1",
+		"bmc_id":      "bmc-1",
+		"chassis_id":  "System.Embedded.1",
+		"source_id":   "PC2",
+		"source_name": "CPU Sub-system Power",
 	})
 	assert.Equal(t, 85.2, chassis1PC2Value)
 
 	// Verify second chassis metric
 	chassis2Value := findMetricValue(t, platformMetric, map[string]string{
-		"source":             "redfish",
-		"node_name":          "worker-1",
-		"bmc_id":             "bmc-1",
-		"chassis_id":         "Enclosure.Internal.0-1",
-		"power_control_id":   "PC1",
-		"power_control_name": "Enclosure Power Control",
+		"source":      "redfish",
+		"node_name":   "worker-1",
+		"bmc_id":      "bmc-1",
+		"chassis_id":  "Enclosure.Internal.0-1",
+		"source_id":   "PC1",
+		"source_name": "Enclosure Power Control",
 	})
 	assert.Equal(t, 125.3, chassis2Value)
 }
@@ -338,9 +338,9 @@ func TestPlatformCollector_Collect_SingleChassis(t *testing.T) {
 				ID: "System.Embedded.1",
 				Readings: []redfish.Reading{
 					{
-						ControlID: "PC1",
-						Name:      "Server Power Control",
-						Power:     300.0 * device.Watt,
+						SourceID:   "PC1",
+						SourceName: "Server Power Control",
+						Power:      300.0 * device.Watt,
 					},
 				},
 			},
@@ -389,9 +389,9 @@ func TestPlatformCollector_Collect_ParallelCollection(t *testing.T) {
 				ID: "System.Embedded.1",
 				Readings: []redfish.Reading{
 					{
-						ControlID: "PC1",
-						Name:      "Server Power Control",
-						Power:     200.0 * device.Watt,
+						SourceID:   "PC1",
+						SourceName: "Server Power Control",
+						Power:      200.0 * device.Watt,
 					},
 				},
 			},
@@ -479,9 +479,9 @@ func TestPlatformCollector_Collect_MetricLabelsValidation(t *testing.T) {
 						ID: tc.chassisID,
 						Readings: []redfish.Reading{
 							{
-								ControlID: "PC1",
-								Name:      "Server Power Control",
-								Power:     tc.power,
+								SourceID:   "PC1",
+								SourceName: "Server Power Control",
+								Power:      tc.power,
 							},
 						},
 					},