forked from n9e/k8s-mon
-
Notifications
You must be signed in to change notification settings - Fork 0
/
kube_state_metrics.go
247 lines (214 loc) · 8.77 KB
/
kube_state_metrics.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
package collect
import (
"fmt"
"strings"
"time"
"github.com/didi/nightingale/src/common/dataobj"
"github.com/go-kit/kit/log"
"github.com/go-kit/kit/log/level"
"github.com/n9e/k8s-mon/config"
)
func DoKubeStatsMetricsCollect(cg *config.Config, logger log.Logger, funcName string) {
start := time.Now()
if len(cg.KubeStatsC.UserSpecifyAddrs) == 0 {
level.Error(logger).Log("msg", "DoKubeStatsMetricsEmptyUserSpecifyAddrs")
return
}
cg.KubeStatsC.Addr = cg.KubeStatsC.UserSpecifyAddrs[0]
metrics, err := CurlTlsMetricsApi(logger, funcName, cg.KubeStatsC, cg.AppendTags, cg.Step, cg.TimeOutSeconds, true)
if err != nil {
level.Error(logger).Log("msg", "DoKubeStatsMetricsCollectCurlTlsMetricsApiResError", "err", err)
return
}
if len(metrics) == 0 {
level.Error(logger).Log("msg", "DoKubeStatsMetricsCollectEmptyMetricsResult")
return
}
level.Debug(logger).Log("msg", "DoKubeStatsMetricsCollectCurlTlsMetricsApiRes", "resNum", len(metrics))
// kube_deployment_labels{deployment="test-server01-deployment", env="prod", instance="10.100.30.139:8080", job="business", label_app="test-server01",
//label_dtree_name="bive", label_dtree_nid="200", namespace="default", region="ap-south-1", service="k8s-mon"}
rm := map[string]struct{}{
"kube_daemonset_labels": {},
"kube_deployment_labels": {},
"kube_endpoint_labels": {},
"kube_namespace_labels": {},
"kube_node_labels": {},
"kube_persistentvolumeclaim_labels": {},
"kube_persistentvolume_labels": {},
"kube_pod_labels": {},
"kube_replicaset_labels": {},
"kube_secret_labels": {},
"kube_service_labels": {},
"kube_statefulset_labels": {},
"kube_storageclass_labels": {},
}
//
//rsm := map[string]struct{}{
// "daemonset": {},
// "deployment": {},
// "endpoint": {},
// "namespace": {},
// "node": {},
// "persistentvolume": {},
// "persistentvolumeclaim": {},
// "pod": {},
// "replicaset": {},
// "secret": {},
// "service": {},
// "statefulset": {},
// "storageclass": {},
//}
// tag 白名单
tagWhiteM := make(map[string]struct{})
if len(cg.KubeStatsC.TagsWhiteList) == 0 {
tagWhiteM = map[string]struct{}{
"namespace": {},
"pod": {},
"pod_name": {},
"container": {},
"container_name": {},
"reason": {},
"secret": {},
"condition": {},
"node": {},
"deployment": {},
"configmap": {},
"daemonset": {},
"endpoint": {},
"replicaset": {},
"statefulset": {},
"persistentvolumeclaim": {},
"persistentvolume": {},
"service": {},
"phase": {},
"type": {},
"status": {},
"resource": {},
"unit": {},
"host_ip": {},
"pod_ip": {},
"cluster_ip": {},
"external_name": {},
"load_balancer_ip": {},
"storageclass": {},
"kernel_version": {},
"os_image": {},
"role": {},
"volume": {},
"access_mode": {},
"volumename": {},
"provisioner": {},
"reclaimPolicy": {},
"volumeBindingMode": {},
"poddisruptionbudget": {},
}
} else {
for _, i := range cg.KubeStatsC.TagsWhiteList {
tagWhiteM[i] = struct{}{}
}
}
nidLabelName := fmt.Sprintf("label_%s", cg.N9eNidLabelName)
//insM := make(map[string]map[string]string)
nidM := make(map[string]string)
sepStr := "|||||"
// 为了防止 deployment daemonset重名问题
//deployment="test-server01" daemonset="test-server01"
for _, metric := range metrics {
if _, loaded := rm[metric.Metric]; loaded {
ss := strings.Split(metric.Metric, "_")
resouceName := ss[1]
resouceId := ""
nid := ""
for k, v := range metric.TagsMap {
if k == nidLabelName {
nid = v
}
if k == resouceName {
resouceId = resouceName + sepStr + v
}
}
if resouceId != "" && nid != "" {
nidM[resouceId] = nid
}
}
}
// 整理label:
var metricList []dataobj.MetricValue
kube_pod_container_resource_requests_cpu_cores_m := make(map[string]float64)
kube_pod_container_resource_limits_cpu_cores_m := make(map[string]float64)
kube_node_status_allocatable_cpu_cores_m := make(map[string]float64)
kube_pod_container_resource_requests_memory_bytes_m := make(map[string]float64)
kube_pod_container_resource_limits_memory_bytes_m := make(map[string]float64)
kube_node_status_capacity_memory_bytes_m := make(map[string]float64)
kube_pod_info_m := make(map[string]float64)
kube_node_status_capacity_pods_m := make(map[string]float64)
kube_pod_container_resource_requests_cpu_cores := "kube_pod_container_resource_requests_cpu_cores"
kube_pod_container_resource_limits_cpu_cores := "kube_pod_container_resource_limits_cpu_cores"
kube_node_status_allocatable_cpu_cores := "kube_node_status_allocatable_cpu_cores"
kube_pod_container_resource_requests_memory_bytes := "kube_pod_container_resource_requests_memory_bytes"
kube_pod_container_resource_limits_memory_bytes := "kube_pod_container_resource_limits_memory_bytes"
kube_node_status_capacity_memory_bytes := "kube_node_status_capacity_memory_bytes"
kube_pod_info := "kube_pod_info"
kube_node_status_capacity_pods := "kube_node_status_capacity_pods"
for _, metric := range metrics {
// 去掉kube_<>_labels
//if _, loaded := rm[metric.Metric]; loaded {
// continue
//}
for k, v := range metric.TagsMap {
kk := k + sepStr + v
if nid, loaded := nidM[kk]; loaded {
metric.Nid = nid
}
if metric.Nid == "" {
metric.Nid = cg.ServerSideNid
}
if _, loaded := tagWhiteM[k]; !loaded {
delete(metric.TagsMap, k)
}
}
labelNode := metric.TagsMap["node"]
switch metric.Metric {
// cpu
case kube_pod_container_resource_requests_cpu_cores:
kube_pod_container_resource_requests_cpu_cores_m[labelNode] += metric.Value
case kube_pod_container_resource_limits_cpu_cores:
kube_pod_container_resource_limits_cpu_cores_m[labelNode] += metric.Value
case kube_node_status_allocatable_cpu_cores:
kube_node_status_allocatable_cpu_cores_m[labelNode] += metric.Value
// mem
case kube_pod_container_resource_requests_memory_bytes:
kube_pod_container_resource_requests_memory_bytes_m[labelNode] += metric.Value
case kube_pod_container_resource_limits_memory_bytes:
kube_pod_container_resource_limits_memory_bytes_m[labelNode] += metric.Value
case kube_node_status_capacity_memory_bytes:
kube_node_status_capacity_memory_bytes_m[labelNode] += metric.Value
// pod num
case kube_pod_info:
kube_pod_info_m[labelNode] += metric.Value
case kube_node_status_capacity_pods:
kube_node_status_capacity_pods_m[labelNode] += metric.Value
}
if metric.CounterType == config.METRIC_TYPE_COUNTER {
metric.Metric = metric.Metric + config.COUNTER_TO_GAUGE_METRIC_NAME_SUFFIX
}
// tags string
metric.Tags = makeAppendTags(metric.TagsMap, cg.AppendTags)
metric.Step = cg.Step
metricList = append(metricList, metric)
}
newtagsm := map[string]string{
cg.MultiFuncUniqueLabel: funcName,
}
// 计算百分比
// cpu
metricList = PercentComputeForKsm(kube_pod_container_resource_requests_cpu_cores_m, kube_node_status_allocatable_cpu_cores_m, cg.ServerSideNid, "kube_node_pod_container_cpu_requests", "node", cg.Step, newtagsm, metricList)
metricList = PercentComputeForKsm(kube_pod_container_resource_limits_cpu_cores_m, kube_node_status_allocatable_cpu_cores_m, cg.ServerSideNid, "kube_node_pod_container_cpu_limits", "node", cg.Step, newtagsm, metricList)
// mem
metricList = PercentComputeForKsm(kube_pod_container_resource_requests_memory_bytes_m, kube_node_status_capacity_memory_bytes_m, cg.ServerSideNid, "kube_node_pod_container_memory_requests", "node", cg.Step, newtagsm, metricList)
metricList = PercentComputeForKsm(kube_pod_container_resource_limits_memory_bytes_m, kube_node_status_capacity_memory_bytes_m, cg.ServerSideNid, "kube_node_pod_container_memory_limits", "node", cg.Step, newtagsm, metricList)
// pod
metricList = PercentComputeForKsm(kube_pod_info_m, kube_node_status_capacity_pods_m, cg.ServerSideNid, "kube_node_pod_num", "node", cg.Step, newtagsm, metricList)
level.Debug(logger).Log("msg", "DoCollectSuccessfullyReadyToPush", "funcName", funcName, "metrics_num", len(metricList), "time_took_seconds", time.Since(start).Seconds(), "metric_addr", cg.KubeStatsC.Addr)
go PushWork(cg.PushServerAddr, cg.TimeOutSeconds, metricList, logger, funcName)
}