feat: upgrade node-exporter to v0.18.1

Signed-off-by: Feng Kun <[email protected]>

pkg/platform/controller/addon/prometheus/controller.go

@@ -1486,21 +1486,20 @@ func createDaemonSetForNodeExporter(components images.Components) *appsv1.Daemon
  Name: nodeExporterDaemonSet,
  Image: components.NodeExporterService.FullName(),
  Args: []string{
- "--path.procfs=/host/proc",
- "--path.sysfs=/host/sys",
+ "--path.rootfs=/host",
  "--no-collector.arp",
  "--no-collector.bcache",
  "--no-collector.bonding",
  "--no-collector.buddyinfo",
  "--no-collector.conntrack",
  "--no-collector.cpu",
+ "--no-collector.cpufreq",
  "--collector.diskstats",
  "--no-collector.drbd",
  "--no-collector.edac",
  "--no-collector.entropy",
  "--no-collector.filefd",
  "--collector.filesystem",
- "--no-collector.gmond",
  "--no-collector.hwmon",
  "--no-collector.infiniband",
  "--no-collector.interrupts",
@@ -1509,13 +1508,15 @@ func createDaemonSetForNodeExporter(components images.Components) *appsv1.Daemon
  "--no-collector.loadavg",
  "--no-collector.logind",
  "--no-collector.mdadm",
- "--no-collector.megacli",
  "--no-collector.meminfo",
  "--no-collector.meminfo_numa",
  "--no-collector.mountstats",
  "--collector.netdev",
  "--no-collector.netstat",
+ "--no-collector.netclass",
  "--no-collector.nfs",
+ "--no-collector.nfsd",
+ "--no-collector.pressure",
  "--no-collector.ntp",
  "--no-collector.qdisc",
  "--no-collector.runit",
@@ -1538,13 +1539,8 @@ func createDaemonSetForNodeExporter(components images.Components) *appsv1.Daemon
  },
  VolumeMounts: []corev1.VolumeMount{
  {
- MountPath: "/host/proc",
- Name: "proc",
- ReadOnly: true,
- },
- {
- MountPath: "/host/sys",
- Name: "sys",
+ MountPath: "/host",
+ Name: "root",
  ReadOnly: true,
  },
  },
@@ -1554,18 +1550,10 @@ func createDaemonSetForNodeExporter(components images.Components) *appsv1.Daemon
  HostPID: true,
  Volumes: []corev1.Volume{
  {
- Name: "proc",
- VolumeSource: corev1.VolumeSource{
- HostPath: &corev1.HostPathVolumeSource{
- Path: "/proc",
- },
- },
- },
- {
- Name: "sys",
+ Name: "root",
  VolumeSource: corev1.VolumeSource{
  HostPath: &corev1.HostPathVolumeSource{
- Path: "/sys",
+ Path: "/",
  },
  },
  },

pkg/platform/controller/addon/prometheus/images/images.go

@@ -57,7 +57,7 @@ var versionMap = map[string]Components{
  LatestVersion: {
  PrometheusService: containerregistry.Image{Name: "prometheus", Tag: "v2.16.0"},
  KubeStateService: containerregistry.Image{Name: "kube-state-metrics", Tag: "v1.9.5"},
- NodeExporterService: containerregistry.Image{Name: "node-exporter", Tag: "v0.15.2"},
+ NodeExporterService: containerregistry.Image{Name: "node-exporter", Tag: "v0.18.1"},
  AlertManagerService: containerregistry.Image{Name: "alertmanager", Tag: "v0.18.0"},
  ConfigMapReloadWorkLoad: containerregistry.Image{Name: "configmap-reload", Tag: "v0.1"},
  PrometheusOperatorService: containerregistry.Image{Name: "prometheus-operator", Tag: "v0.31.1"},

pkg/platform/controller/addon/prometheus/yamls.go

@@ -310,7 +310,7 @@ func scrapeConfigForPrometheus() string {
  regex: (.+)
  metric_relabel_configs:
  - source_labels: [ __name__ ]
- regex: 'scheduler_e2e_scheduling_latency_microseconds_sum|scheduler_e2e_scheduling_latency_microseconds_count|apiserver_request_duration_seconds_(.*)|node_sockstat_TCP_inuse|node_network_transmit_bytes|node_network_receive_bytes|node_filesystem_size|node_filesystem_avail|node_disk_bytes_written|node_disk_bytes_read|node_disk_writes_completed|node_disk_reads_completed'
+ regex: 'scheduler_e2e_scheduling_latency_microseconds_sum|scheduler_e2e_scheduling_latency_microseconds_count|apiserver_request_duration_seconds_(.*)|node_sockstat_TCP_inuse|node_network_transmit_bytes_total|node_network_receive_bytes_total|node_filesystem_size_bytes|node_filesystem_avail_bytes|node_disk_written_bytes_total|node_disk_read_bytes_total|node_disk_writes_completed_total|node_disk_reads_completed_total'
  action: keep
  - regex: "instance|job|pod_name|namespace|scope|subresource"
  action: labeldrop
@@ -643,28 +643,34 @@ groups:
  expr: sum(k8s_pod_gpu_memory_used) without(namespace,pod_name,workload_kind,workload_name) *100 / on(node) group_left() kube_node_status_capacity_gpu_memory
 
  - record: k8s_node_fs_write_bytes
- expr: (sum by (node) (irate(node_disk_bytes_written[4m]))) *on(node) group_left(node_role) kube_node_labels
+ expr: (sum by (node) (irate(node_disk_written_bytes_total[4m]))) *on(node) group_left(node_role) kube_node_labels
 
  - record: k8s_node_fs_read_bytes
- expr: (sum by (node) (irate(node_disk_bytes_read[4m])))*on(node) group_left(node_role) kube_node_labels
+ expr: (sum by (node) (irate(node_disk_read_bytes_total[4m])))*on(node) group_left(node_role) kube_node_labels
 
  - record: k8s_node_fs_write_times
- expr: (sum by (node) (irate(node_disk_writes_completed[4m])))*on(node) group_left(node_role) kube_node_labels
+ expr: (sum by (node) (irate(node_disk_writes_completed_total[4m])))*on(node) group_left(node_role) kube_node_labels
 
  - record: k8s_node_fs_read_times
- expr: (sum by (node) (irate(node_disk_reads_completed[4m])))*on(node) group_left(node_role) kube_node_labels
+ expr: (sum by (node) (irate(node_disk_reads_completed_total[4m])))*on(node) group_left(node_role) kube_node_labels
 
  - record: k8s_node_pod_num
  expr: count(k8s_pod_status_ready) without (pod_name,workload_kind,workload_name,namespace)
 
  - record: k8s_node_disk_space_rate
- expr: (100 - sum (node_filesystem_avail{fstype=~"ext3|ext4|xfs"}) by (node) / sum (node_filesystem_size{fstype=~"ext3|ext4|xfs"}) by (node) *100) *on(node) group_left(node_role) kube_node_labels
+ expr: (100 - sum (node_filesystem_avail_bytes{fstype=~"ext3|ext4|xfs"}) by (node) / sum (node_filesystem_size_bytes{fstype=~"ext3|ext4|xfs"}) by (node) *100) *on(node) group_left(node_role) kube_node_labels
+
+ - record: k8s_node_filesystem_avail_bytes
+ expr: node_filesystem_avail_bytes{fstype=~"ext3|ext4|xfs"}
+
+ - record: k8s_node_filesystem_size_bytes
+ expr: node_filesystem_size_bytes{fstype=~"ext3|ext4|xfs"}
 
  - record: k8s_node_network_receive_bytes_bw
- expr: (sum by (node) (irate(node_network_receive_bytes{device!~"lo|veth(.*)|virb(.*)|docker(.*)|tunl(.*)|v-h(.*)|flannel(.*)"}[5m])))*on(node) group_left(node_role) kube_node_labels
+ expr: (sum by (node) (irate(node_network_receive_bytes_total{device!~"lo|veth(.*)|virb(.*)|docker(.*)|tunl(.*)|v-h(.*)|flannel(.*)"}[5m])))*on(node) group_left(node_role) kube_node_labels
 
  - record: k8s_node_network_transmit_bytes_bw
- expr: (sum by (node) (irate(node_network_transmit_bytes{device!~"lo|veth(.*)|virb(.*)|docker(.*)|tunl(.*)|v-h(.*)|flannel(.*)"}[5m])))*on(node) group_left(node_role) kube_node_labels
+ expr: (sum by (node) (irate(node_network_transmit_bytes_total{device!~"lo|veth(.*)|virb(.*)|docker(.*)|tunl(.*)|v-h(.*)|flannel(.*)"}[5m])))*on(node) group_left(node_role) kube_node_labels
 
  - record: k8s_workload_abnormal
  expr: |-