當一個叢集建立好之後總會需要安裝 Metrics 相關服務，這邊使用 ArgoCD 建立相關應用程式，讓 Cluster 加入之後就可以直接用 ArgoCD 部屬 Metrics-Server / Prometheus-adapter 等服務。
- CLI Command - 設定 Cluster/加入 Repository
- Metrics-Server 的 Application 設定 yaml
- Prometheus-adapter 的 Application 設定 yaml
1
| argocd cluster add srvk8s-admin@srvk8s --name Master-Cluster
|
1
2
3
| argocd repo add https://prometheus-community.github.io/helm-charts --type helm --name Prometheus
argocd repo add https://kubernetes.github.io/dashboard --type helm --name kubernetes-dashboard
argocd repo add https://kubernetes-sigs.github.io/metrics-server --type helm --name metrics-server
|
helm repository
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
| apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
name: metrics-server-master-cluster
spec:
destination:
name: Master-Cluster
namespace: kube-system
server: ''
source:
path: ''
repoURL: 'https://kubernetes-sigs.github.io/metrics-server/'
targetRevision: 3.8.2
chart: metrics-server
helm:
valueFiles:
- values.yaml
parameters:
- name: metrics.enabled
value: 'true'
values: |-
args:
- --kubelet-insecure-tls
project: cluster-infra
syncPolicy:
syncOptions:
- CreateNamespace=true
- PruneLast=true
|
警告
此方法搭配 Istio & Istio 其中佩附的 Prometheus，是 Metrics-Server 的替代作法。
由於 metrics-server 安裝後無法在 lens (kubernetes IDE) 上看到各個 pod 的資源狀態，如果 cluster 的資源夠多，我會建議使用這個做法。
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
| apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
name: prometheus-adapter-Master-Cluster
spec:
destination:
name: Master-Cluster
namespace: istio-system
server: ''
source:
path: ''
repoURL: 'https://prometheus-community.github.io/helm-charts'
targetRevision: 3.0.1
chart: prometheus-adapter
helm:
valueFiles:
- values.yaml
values: |-
rules:
default: true
resource:
cpu:
containerQuery: sum(rate(container_cpu_usage_seconds_total{<<.LabelMatchers>>, container!=""}[5m])) by (<<.GroupBy>>)
nodeQuery: sum(rate(container_cpu_usage_seconds_total{<<.LabelMatchers>>, id='/'}[5m])) by (<<.GroupBy>>)
resources:
overrides:
node:
resource: node
namespace:
resource: namespace
pod:
resource: pod
containerLabel: container
memory:
containerQuery: sum(container_memory_working_set_bytes{<<.LabelMatchers>>, container!=""}) by (<<.GroupBy>>)
nodeQuery: sum(container_memory_working_set_bytes{<<.LabelMatchers>>,id='/'}) by (<<.GroupBy>>)
resources:
overrides:
node:
resource: node
namespace:
resource: namespace
pod:
resource: pod
containerLabel: container
window: 5m
custom:
- seriesQuery: '{__name__=~"istio_requests_total"}'
seriesFilters: []
resources:
overrides:
kubernetes_namespace:
resource: namespace
kubernetes_pod_name:
resource: pod
destination_service_name:
resource: service
name:
matches: "^(.*)_total"
as: "${1}_per_min"
metricsQuery: sum(increase(<<.Series>>{<<.LabelMatchers>>}[1m])) by (<<.GroupBy>>)
parameters:
- name: prometheus.url
value: 'http://prometheus.istio-system.svc.cluster.local'
project: cluster-infra
syncPolicy:
syncOptions: []
|
危險
這個設定我尚未完成測試使用，僅成功安裝 & 利用 Lens 的 Forward 功能連進去，但是登入的部分碰到困難。
另外我的環境都有 istio，之後等測試 OK 的時候會再補上 istio 的 ingress 相關設定
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
| apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
name: kube-dashboard-master-cluster
spec:
destination:
name: Master-Cluster
namespace: kube-dashboard
server: ''
source:
path: ''
repoURL: 'https://kubernetes.github.io/dashboard/'
targetRevision: 5.3.1
chart: kubernetes-dashboard
helm:
valueFiles:
- values.yaml
project: cluster-infra
syncPolicy:
syncOptions:
- PruneLast=true
- CreateNamespace=true
|