Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

better embedded cluster usage metrics #4083

Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
67 changes: 49 additions & 18 deletions pkg/helmvm/helmvm_node.go
Original file line number Diff line number Diff line change
Expand Up @@ -60,9 +60,9 @@ func nodeMetrics(ctx context.Context, client kubernetes.Interface, metricsClient
return nil, fmt.Errorf("pods per node: %w", err)
}

cpuCapacity := types.CapacityAvailable{}
memoryCapacity := types.CapacityAvailable{}
podCapacity := types.CapacityAvailable{}
cpuCapacity := types.CapacityUsed{}
memoryCapacity := types.CapacityUsed{}
podCapacity := types.CapacityUsed{}

memoryCapacity.Capacity = float64(node.Status.Capacity.Memory().Value()) / math.Pow(2, 30) // capacity in GB

Expand All @@ -76,36 +76,67 @@ func nodeMetrics(ctx context.Context, client kubernetes.Interface, metricsClient
nodeUsageMetrics, err := metricsClient.MetricsV1beta1().NodeMetricses().Get(ctx, node.Name, metav1.GetOptions{})
if err == nil {
if nodeUsageMetrics.Usage.Memory() != nil {
memoryCapacity.Available = memoryCapacity.Capacity - float64(nodeUsageMetrics.Usage.Memory().Value())/math.Pow(2, 30)
memoryCapacity.Used = float64(nodeUsageMetrics.Usage.Memory().Value()) / math.Pow(2, 30)
}

if nodeUsageMetrics.Usage.Cpu() != nil {
cpuCapacity.Available = cpuCapacity.Capacity - nodeUsageMetrics.Usage.Cpu().AsApproximateFloat64()
cpuCapacity.Used = nodeUsageMetrics.Usage.Cpu().AsApproximateFloat64()
}
} else {
// if we can't get metrics, we'll do nothing for now
// in the future we may decide to retry or log a warning
}

podCapacity.Available = podCapacity.Capacity - float64(len(nodePods))
podCapacity.Used = float64(len(nodePods))

nodeLabelArray := []string{}
for k, v := range node.Labels {
nodeLabelArray = append(nodeLabelArray, fmt.Sprintf("%s:%s", k, v))
}

podInfo := []types.PodInfo{}

for _, pod := range nodePods {
newInfo := types.PodInfo{
Name: pod.Name,
Namespace: pod.Namespace,
Status: string(pod.Status.Phase),
}

podMetrics, err := metricsClient.MetricsV1beta1().PodMetricses(pod.Namespace).Get(ctx, pod.Name, metav1.GetOptions{})
if err == nil {
podTotalMemory := 0.0
podTotalCPU := 0.0
for _, container := range podMetrics.Containers {
if container.Usage.Memory() != nil {
podTotalMemory += float64(container.Usage.Memory().Value()) / math.Pow(2, 30)
}
if container.Usage.Cpu() != nil {
podTotalCPU += container.Usage.Cpu().AsApproximateFloat64()
}
}
newInfo.Memory = podTotalMemory
newInfo.CPU = podTotalCPU
}

podInfo = append(podInfo, newInfo)
}

return &types.Node{
Name: node.Name,
IsConnected: isConnected(node),
IsReady: isReady(node),
IsPrimaryNode: isPrimary(node),
CanDelete: node.Spec.Unschedulable && !isConnected(node),
KubeletVersion: node.Status.NodeInfo.KubeletVersion,
CPU: cpuCapacity,
Memory: memoryCapacity,
Pods: podCapacity,
Labels: nodeLabelArray,
Conditions: findNodeConditions(node.Status.Conditions),
PodList: nodePods,
Name: node.Name,
IsConnected: isConnected(node),
IsReady: isReady(node),
IsPrimaryNode: isPrimary(node),
CanDelete: node.Spec.Unschedulable && !isConnected(node),
KubeletVersion: node.Status.NodeInfo.KubeletVersion,
KubeProxyVersion: node.Status.NodeInfo.KubeProxyVersion,
OperatingSystem: node.Status.NodeInfo.OperatingSystem,
KernelVersion: node.Status.NodeInfo.KernelVersion,
CPU: cpuCapacity,
Memory: memoryCapacity,
Pods: podCapacity,
Labels: nodeLabelArray,
Conditions: findNodeConditions(node.Status.Conditions),
PodList: podInfo,
}, nil
}
43 changes: 26 additions & 17 deletions pkg/helmvm/types/types.go
Original file line number Diff line number Diff line change
@@ -1,31 +1,32 @@
package types

import corev1 "k8s.io/api/core/v1"

type HelmVMNodes struct {
Nodes []Node `json:"nodes"`
HA bool `json:"ha"`
IsHelmVMEnabled bool `json:"isHelmVMEnabled"`
}

type Node struct {
Name string `json:"name"`
IsConnected bool `json:"isConnected"`
IsReady bool `json:"isReady"`
IsPrimaryNode bool `json:"isPrimaryNode"`
CanDelete bool `json:"canDelete"`
KubeletVersion string `json:"kubeletVersion"`
CPU CapacityAvailable `json:"cpu"`
Memory CapacityAvailable `json:"memory"`
Pods CapacityAvailable `json:"pods"`
Labels []string `json:"labels"`
Conditions NodeConditions `json:"conditions"`
PodList []corev1.Pod `json:"podList"`
Name string `json:"name"`
IsConnected bool `json:"isConnected"`
IsReady bool `json:"isReady"`
IsPrimaryNode bool `json:"isPrimaryNode"`
CanDelete bool `json:"canDelete"`
KubeletVersion string `json:"kubeletVersion"`
KubeProxyVersion string `json:"kubeProxyVersion"`
OperatingSystem string `json:"operatingSystem"`
KernelVersion string `json:"kernelVersion"`
CPU CapacityUsed `json:"cpu"`
Memory CapacityUsed `json:"memory"`
Pods CapacityUsed `json:"pods"`
Labels []string `json:"labels"`
Conditions NodeConditions `json:"conditions"`
PodList []PodInfo `json:"podList"`
}

type CapacityAvailable struct {
Capacity float64 `json:"capacity"`
Available float64 `json:"available"`
type CapacityUsed struct {
Capacity float64 `json:"capacity"`
Used float64 `json:"used"`
}

type NodeConditions struct {
Expand All @@ -34,3 +35,11 @@ type NodeConditions struct {
PidPressure bool `json:"pidPressure"`
Ready bool `json:"ready"`
}

type PodInfo struct {
Name string `json:"name"`
Status string `json:"status"`
Namespace string `json:"namespace"`
CPU float64 `json:"cpu"`
Memory float64 `json:"memory"`
}
Loading