fix bug in metrics

CHANGELOG.md

@@ -148,3 +148,9 @@ Fix device memory calculation error after container crashloop
 Fix env cuda_oversubscribe not set properly when MemoryScaling < 1
 
 Fix MemoryScaling not working when set < 1
+
+## v2.2.15
+
+Move shared-memory from from /tmp/xxx.cache to /usr/local/vgpu/xxx.cache inside container
+
+Add Deviceidx to scheduler monitor apis(31993)

charts/vgpu/values.yaml

@@ -3,7 +3,7 @@
 nameOverride: ""
 fullnameOverride: ""
 imagePullSecrets: []
-version: "v2.2.14"
+version: "v2.2.15"
 
 #Nvidia GPU Parameters
 resourceName: "nvidia.com/gpu"

cmd/scheduler/metrics.go

@@ -149,17 +149,17 @@ func (cc ClusterManagerCollector) Collect(ch chan<- prometheus.Metric) {
 	ctrvGPUDeviceAllocatedDesc := prometheus.NewDesc(
 		"vGPUPodsDeviceAllocated",
 		"vGPU Allocated from pods",
-		[]string{"namespace", "nodename", "podname", "containeridx", "deviceuuid", "deviceidx", "deviceusedcore"}, nil,
+		[]string{"namespace", "nodename", "podname", "containeridx", "deviceuuid", "deviceusedcore"}, nil,
 	)
 	ctrvGPUdeviceAllocatedMemoryPercentageDesc := prometheus.NewDesc(
 		"vGPUMemoryPercentage",
 		"vGPU memory percentage allocated from a container",
-		[]string{"namespace", "nodename", "podname", "containeridx", "deviceuuid", "deviceidx"}, nil,
+		[]string{"namespace", "nodename", "podname", "containeridx", "deviceuuid"}, nil,
 	)
 	ctrvGPUdeviceAllocateCorePercentageDesc := prometheus.NewDesc(
 		"vGPUCorePercentage",
 		"vGPU core allocated from a container",
-		[]string{"namespace", "nmodename", "podname", "containeridx", "deviceuuid", "deviceidx"}, nil,
+		[]string{"namespace", "nmodename", "podname", "containeridx", "deviceuuid"}, nil,
 	)
 	schedpods, _ := sher.GetScheduledPods()
 	for _, val := range schedpods {