k8s入門知識點
[TOC]
k8s并不神秘,你可以結(jié)合vm和redis之類的中間件
Q-1:為什么需要k8s
Q0:k8s架構(gòu),k8s基本概念入門
Q1:k8s的yml講解。
Q2:k8s能做什么?k8s如何做?
推薦官方文檔和中文文檔:
1.為什么需要k8s
來個官方鏈接:https://kubernetes.io/docs/concepts/overview/what-is-kubernetes/
2.基本架構(gòu)和基本概念入門
2.1基本架構(gòu)
對比幾張來自不同網(wǎng)站的架構(gòu)圖

[圖片上傳失敗...(image-eb049d-1582014664031)]
必須包含master和node 各至少一個。
-
master節(jié)點可認為是管家,包含核心組件
- <1>api-server(一組Restful的api接口,k8s集群控制的唯一入口和前端,是組件間交互的中樞,像是身份認證,授權(quán)等機制都由api提供)
- <2>數(shù)據(jù)管理(k8s的數(shù)據(jù)均存儲在etcd中,僅由apiserver操作數(shù)據(jù))
- <3>scheduler(pod調(diào)度,可自定義調(diào)度算法)
- <4>controller-manager(管理集群內(nèi)各類資源controller,如副本數(shù),endpoint,namespace,自愈等,使之始終處于“期望狀態(tài)”)等功能。
官方推薦master部署在獨立的物理主機上,該主機一般不會run其他服務(wù)。
-
node 節(jié)點就是用來run實際的pod,包含
- kubelet(systemd來進行管理,定期向master 匯報node狀態(tài),并接受apiserver的命令,負責node和master之間的通信,可認為是node的聲明周期管理器)
- kube-proxy(service的代理器,將到達service的request load到具體的endpoint去,由iptables mode實現(xiàn),比如常用的nodeport模式可實現(xiàn)外部訪問集群內(nèi)部service)
- container - runtime (還包含了各種容器,如docker的運行環(huán)境)。
如下單service的應(yīng)用外部訪問示意:hello service對外提供服務(wù),內(nèi)部負載了3個pod,controller 為deployment

2.2 k8s的對象模型
k8s可以看做是面向?qū)ο蟮?,每類服?wù)可看做是k8s的一個對象。這些對象由用戶定義yaml,k8s的api負責創(chuàng)建。所有對象包含spec(規(guī)范)+status兩類基本信息。
例如:k8s創(chuàng)建pod的api為:https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.17/#pod-v1-core
kubectl 也會將yaml轉(zhuǎn)成json發(fā)送到master節(jié)點。

k8s中常用的對象有pod、deployment、service、statefulSet等,每個對象至少包含3個metadata:namespace、name、uid。
2.2.1 k8s中常見的object kind:
-
pod
- 基本調(diào)度單元,特定關(guān)系的一組容器集合,最小的部署集合。
- 非持久性實體,調(diào)度失敗、自愈等情況會被終止,重建,不建議手動創(chuàng)建Pod,而是通過controller創(chuàng)建pod,如deployment。
-
Service
- pod 重啟之后ip會改變,多個相同服務(wù)的pod需要有discovery+loadbalance 。這一點與“微服務(wù)”中的概念一致。通過label selector 可輕松實現(xiàn)邏輯分組。
- service 的聲明周期內(nèi),ip不會改變??赏ㄟ^nodeport 暴露到外部。
- 有的service不需要ip、有的service 不需要負載均衡。
-
Controllers
Replicaset (rs)副本數(shù),用于loadbalance和冗余。(老版本是ReplicationController- rc,算是功能升級,官方推薦使用rs 代替rc)?,F(xiàn)階段非特殊情況如升級pod的操作,也不推薦單獨使用,
-
Deployment
使用并管理rs ,算是更高一層的概念,這是現(xiàn)在比較常用的部署app的方式。deployment為pod和rs提供聲明式更新(而非命令式)。支持滾動更新(rollingUpdate),支持回滾操作。

-
statefulSet(k8s 1.9 GA)
我們自己開發(fā)的應(yīng)用一般都是stateless的,像是redis、zk、kafka、mysql這類的中間件通常需要使用statefulSet。通常適用于穩(wěn)定的持久存儲、穩(wěn)定的網(wǎng)絡(luò)標識、有序部署有序擴展、有序收縮、有序滾動升級的場景。
pod的存儲一般都是volumes外掛到persistent介質(zhì)。并且伸縮或刪除不會刪除關(guān)聯(lián)的存儲。需要headless service負責pod的網(wǎng)絡(luò)身份。
<font color=red size=1>
"注意:statefulSet 對應(yīng)的service為headless servie,和普通的service的區(qū)別在于,普通的有cluster ip,通過只有service 有dns,通過iptables進行負載。headless service無cluster ip,endpoints是所有pod的dns地址。就意味著statefulset 創(chuàng)建pod的時候為pod生成了dns信息。如3 節(jié)點的mysql,會生成mysql-0 ... mysql-2三個pod(pod名稱生產(chǎn)規(guī)則為pod+遞增序號),并且會生成域: $(podname).(headless server name),完整FQDN為: $(podname).(headless server name).namespace.svc.cluster.local"</font>看一個完整的statefulSet + headless service 的示例。
apiVersion: v1 kind: Service metadata: name: nginx labels: app: nginx spec: ports: - port: 80 name: web clusterIP: None selector: app: nginx --- apiVersion: apps/v1beta1 kind: StatefulSet metadata: name: web spec: serviceName: "nginx" replicas: 1 template: metadata: labels: app: nginx spec: containers: - name: nginx image: nginx:1.11 ports: - containerPort: 80 name: web volumeMounts: - name: www mountPath: /usr/share/nginx/html nodeSelector: node: kube-node3 volumes: - name: www hostPath: path: /mydir -
DeamonSet
每個node都有一個pod副本運行。常用在后臺常駐程序如ceph,日志收集,心跳檢查,Prometheus exporter等。
-
ConfigMap
向pod 提供非敏感的配置信息,支持鍵值對,單個屬性,配置文件。
使用方式:環(huán)境變量,容器命令行參數(shù),數(shù)據(jù)卷掛載。
定義示例:
apiVersion:V1 kind:ConfigMap metadata: name:hello-config namespace:public-config data: title:hello everybody sex:girl引用示例:
apiVersion:V1 kind:deployment metadata: name:hello-dem spec: containers: - name:goodboy image:hub/images command:["/bin/bash","-c","echo ${envTitle} ${envSex}"] env: - name:envTitle valueFrom: configMapKeyRef: name:hello-config key:title - name:envSex valueFrom: configMapKeyRef: name:hello-config key:sex -
Secret
對應(yīng)于ConfigMap,用于敏感信息配置,如token,密碼,秘鑰等。
-
Endpoints
service 和 pod 之間的關(guān)系會創(chuàng)建endpoint對象,默認與service 同名。kube-proxy將會監(jiān)聽service 和 endpoint的變化,從而更新iptables的規(guī)則。
-
PodDisruptionBudget(pdb,主動驅(qū)逐保護)
- 沒有pdb。當進行節(jié)點維護時,如果某個服務(wù)的多個pod在該節(jié)點上,則節(jié)點的停機可能會造成服務(wù)中斷或者服務(wù)降級。舉個例子,某服務(wù)有5個pod,最低3個pod能保證服務(wù)質(zhì)量,否則會造成響應(yīng)慢等影響,此時該服務(wù)的4個pod在node01上,如果對node01進行停機維護,此時只有1個pod能正常對外服務(wù),在node01的4個pod遷移過程中,就會影響該服務(wù)正常響應(yīng);
- pdb能保證應(yīng)用在節(jié)點維護時不低于一定數(shù)量的pod運行,從而保持服務(wù)質(zhì)量;
等等其他的如 ingress
2.2.2 k8s中有常見的metadata
-
name && UID (uid是在k8s的整個聲明周期中均唯一,不會產(chǎn)生相同的uid,可對等為mysql的auto increment key)
來看一個api訪問對象的路徑:/api/{version}/namespaces/{namespace}/{object-kind}/{name} ,k8s通過層層限定來尋找唯一標識的name。
-
namespace
對一組資源和對象的抽象集合,從邏輯上劃分k8s實現(xiàn)分層管理,并實現(xiàn)一定層度上的資源和權(quán)限隔離。
內(nèi)置三個namespace:default,kube-system(存放api-server、dns插件等),kube-public(供所有用戶包含未經(jīng)過身份驗證的用戶使用,實現(xiàn)資源集群內(nèi)共享)
Node和persistentVolume(簡稱PV,通過PVC和pod綁定)不屬于任何namespace。
-
label
標識性的數(shù)據(jù),有嚴格的命名規(guī)范,k8s可通過標簽組合管理對象,達到松耦合。在spec中通過selectors進行匹配。
-
annotation(理解為java注釋吧)
非標識性的元數(shù)據(jù)附加到對象上。通常會有時間戳,版本號,用戶信息等輔助信息。
2.2.3 k8s spec中常見的參數(shù)
-
selectors(標簽和標簽選擇器)
對應(yīng)label對應(yīng)的key-value 進行對象的選擇
2.3 k8s的ip模型
k8s的ip:
- node Ip :node節(jié)點的ip,為物理ip.
- pod Ip:pod的ip,即docker 容器的ip,為虛擬ip。
- cluster Ip:service 的ip,為虛擬ip。提供一個集群內(nèi)部的虛擬IP以供Pod訪問。
2.4 k8s的volume
k8s的volume和docker有所不同,v是獨立于容器的,與pod聲明周期相同,即pod刪除空間也被刪除。有多重類型
-
emptydir
空目錄,pod中的容器會共享此目錄
如下所示,busybox的文件寫入,在nginx容器中能夠讀出來。
[root@master ~]# cat test.yaml apiVersion: v1 kind: Service metadata: name: serivce-mynginx namespace: default spec: type: NodePort selector: app: mynginx ports: - name: nginx port: 80 targetPort: 80 nodePort: 30080 --- apiVersion: apps/v1 kind: Deployment metadata: name: deploy namespace: default spec: replicas: 1 selector: matchLabels: app: mynginx template: metadata: labels: app: mynginx spec: containers: - name: mynginx image: lizhaoqwe/nginx:v1 volumeMounts: - mountPath: /usr/share/nginx/html/ name: share ports: - name: nginx containerPort: 80 - name: busybox image: busybox command: - "/bin/sh" - "-c" - "sleep 4444" volumeMounts: - mountPath: /data/ name: share volumes: - name: share emptyDir: {}
-
hostpath
使用node的目錄進行掛載
-
configmap && secret
前面講過,configmap和secret 都可以使用文件的方式,也可以把文件直接掛載到容器中。
如:
spec: containers: - name: nginx image: nginx ports: - name: nginx containerPort: 80 volumeMounts: - name: html-config mountPath: /nginx/vars/ readOnly: true volumes: - name: html-config configMap: name: nginx-var
-
外部存儲
如nfs,ceph等。外部存儲一般需要用到storageClass + pvc + pv。可參考blog:https://www.cnblogs.com/benjamin77/p/9944268.html
2.5 k8s port
Port,Targetport,Nodeport區(qū)別
service 暴露nodeport 到node上,node 接受request 之后通過kube-proxy 負載到具體的pod(iptables 或 ipvs)。

3.k8s yaml講解
一個yaml通常如下:其中apiVersion、kind、metadata、spec(規(guī)格)是必填的項目。通常spec對于對象而言是能起決定性作用的描述。且對于pod、deployment、service等對象有各自特殊格式的spec。
3.1 Depolyment的yaml示例
Deployment API 版本對照表
| Kubernetes 版本 | Deployment 版本 |
|---|---|
| v1.5-v1.15 | extensions/v1beta1 |
| v1.7-v1.15 | apps/v1beta1 |
| v1.8-v1.15 | apps/v1beta2 |
| v1.9+ | apps/v1 |
這是sre定義的內(nèi)容:
apiVersion: apps/v1beta2
kind: Deployment
metadata:
labels:
k8s-app: chainaccesspipe # lables k-v形式供service selector選擇
name: chainaccesspipe
namespace: chainaccess-dev # 命名空間
annotations:
title: "這是個測試title"
date: "2020/10/10 "
status: "be happy"
spec:
replicas: 1 # 副本數(shù)
selector:
matchLabels:
k8s-app: chainaccesspipe #template選擇器
template:
metadata:
labels:
k8s-app: chainaccesspipe
spec:
initContainers: #初始化容器
- image: www.xxx.com/middleware/sysctl:latest
imagePullPolicy: Always
name: sysctl
resources:
limits:
cpu: 20m
memory: 20Mi
requests:
cpu: 20m
memory: 20Mi
securityContext:
privileged: true
runAsUser: 0
containers:
- name: chainaccesspipe
image: www.xxx.com/baseserver-dev/pipe:integration.0.9.1.0.0
# resources:
# requests:
# memory: "2G"
# cpu: "350m"
# limits:
# memory: "2G"
# cpu: 1
# livenessProbe:
# httpGet:
# path: /actuator/health
# port: 8017
# initialDelaySeconds: 150
# periodSeconds: 5
# readinessProbe:
# httpGet:
# path: /actuator/health
# port: 8017
# initialDelaySeconds: 150
# periodSeconds: 3
env: # 環(huán)境變量
- name: LC_ALL
value: en_US.utf8
- name: JAVA_OPTS_OVERRIDE
value: |-
-Xms1024m
-Xmx1024m
-Dport=8017
-Dnacos-namespace=5ea6a570-379f-4d74-9553-066a2227720f
-Duser.timezone=GMT+08
args:
- -jar
- /app/app.jar
ports:
- containerPort: 8017 #容器暴露端口
imagePullPolicy: Always
securityContext:
runAsUser: 1000
fsGroup: 1000
imagePullSecrets:
- name: sec8084
---
apiVersion: v1
kind: Service
metadata:
labels:
k8s-app: chainaccesspipe
annotations:
prometheus.io/scrape: "true"
prometheus.io/app-metrics-path: /actuator/prometheus
name: chainaccesspipe
namespace: chainaccess-dev
spec:
ports:
- port: 8017
targetPort: 8017
nodePort: 28017
type: NodePort
selector:
k8s-app: chainaccesspipe
接下來我們登陸21 ,通過api獲取下完整的map:
curl -s 127.0.0.1:8080/apis/apps/v1/namespaces/chainaccess-dev/deployments/chainaccesspipe | jq
內(nèi)容如下,這個是v1版本下,完整的map,我們可以對比下定義的yaml。
{
"kind": "Deployment",
"apiVersion": "apps/v1",
"metadata": {
"name": "chainaccesspipe",
"namespace": "chainaccess-dev",
"selfLink": "/apis/apps/v1/namespaces/chainaccess-dev/deployments/chainaccesspipe",
"uid": "9ca01515-2aca-11ea-b7ca-005056aa5684",
"resourceVersion": "84030034",
"generation": 1,
"creationTimestamp": "2019-12-30T06:07:08Z",
"labels": {
"k8s-app": "chainaccesspipe"
},
"annotations": {
"deployment.kubernetes.io/revision": "1"
}
},
"spec": {
"replicas": 1,
"selector": {
"matchLabels": {
"k8s-app": "chainaccesspipe"
}
},
"template": {
"metadata": {
"creationTimestamp": null,
"labels": {
"k8s-app": "chainaccesspipe"
}
},
"spec": {
"initContainers": [
{
"name": "sysctl",
"image": "harbor.xxx.com/middleware/sysctl:latest",
"resources": {
"limits": {
"cpu": "20m",
"memory": "20Mi"
},
"requests": {
"cpu": "20m",
"memory": "20Mi"
}
},
"terminationMessagePath": "/dev/termination-log",
"terminationMessagePolicy": "File",
"imagePullPolicy": "Always",
"securityContext": {
"privileged": true,
"runAsUser": 0
}
}
],
"containers": [
{
"name": "chainaccesspipe",
"image": "harbor.xxx.com/baseserver-dev/pipe:release.0.9.1.1.0",
"args": [
"-jar",
"/app/app.jar"
],
"ports": [
{
"containerPort": 8080,
"protocol": "TCP"
}
],
"env": [
{
"name": "LC_ALL",
"value": "en_US.utf8"
},
{
"name": "JAVA_OPTS_OVERRIDE",
"value": "-Xms1024m\n-Xmx1024m\n-Dport=8080\n-Dnacos-server-addr=nacos-headless.nacos-server:8848\n-Dnacos-namespace=9482afb2-3267-45ea-92cd-48edaae5837b\n-Duser.timezone=GMT+08"
}
],
"resources": {},
"terminationMessagePath": "/dev/termination-log",
"terminationMessagePolicy": "File",
"imagePullPolicy": "Always"
}
],
"restartPolicy": "Always",
"terminationGracePeriodSeconds": 30,
"dnsPolicy": "ClusterFirst",
"securityContext": {
"runAsUser": 1000,
"fsGroup": 1000
},
"imagePullSecrets": [
{
"name": "sec8084"
}
],
"schedulerName": "default-scheduler"
}
},
"strategy": {
"type": "RollingUpdate",
"rollingUpdate": {
"maxUnavailable": "25%",
"maxSurge": "25%"
}
},
"revisionHistoryLimit": 10,
"progressDeadlineSeconds": 600
},
"status": {
"observedGeneration": 1,
"replicas": 1,
"updatedReplicas": 1,
"readyReplicas": 1,
"availableReplicas": 1,
"conditions": [
{
"type": "Progressing",
"status": "True",
"lastUpdateTime": "2019-12-30T06:07:17Z",
"lastTransitionTime": "2019-12-30T06:07:08Z",
"reason": "NewReplicaSetAvailable",
"message": "ReplicaSet \"chainaccesspipe-6bf9469bf4\" has successfully progressed."
},
{
"type": "Available",
"status": "True",
"lastUpdateTime": "2020-01-19T08:10:00Z",
"lastTransitionTime": "2020-01-19T08:10:00Z",
"reason": "MinimumReplicasAvailable",
"message": "Deployment has minimum availability."
}
]
}
}
3.2 pod完整yaml
apiVersion: v1 #必選,版本號,例如v1,版本號必須可以用 kubectl api-versions 查詢到 .
kind: Pod #必選,Pod
metadata: #必選,元數(shù)據(jù)
name: string #必選,Pod名稱
namespace: string #必選,Pod所屬的命名空間,默認為"default"
labels: #自定義標簽
- name: string #自定義標簽名字
annotations: #自定義注釋列表
- name: string
spec: #必選,Pod中容器的詳細定義
containers: #必選,Pod中容器列表
- name: string #必選,容器名稱,需符合RFC 1035規(guī)范
image: string #必選,容器的鏡像名稱
imagePullPolicy: [ Always|Never|IfNotPresent ] #獲取鏡像的策略 Alawys表示下載鏡像 IfnotPresent表示優(yōu)先使用本地鏡像,否則下載鏡像,Nerver表示僅使用本地鏡像
command: [string] #容器的啟動命令列表,如不指定,使用打包時使用的啟動命令
args: [string] #容器的啟動命令參數(shù)列表
workingDir: string #容器的工作目錄
volumeMounts: #掛載到容器內(nèi)部的存儲卷配置
- name: string #引用pod定義的共享存儲卷的名稱,需用volumes[]部分定義的的卷名
mountPath: string #存儲卷在容器內(nèi)mount的絕對路徑,應(yīng)少于512字符
readOnly: boolean #是否為只讀模式
ports: #需要暴露的端口庫號列表
- name: string #端口的名稱
containerPort: int #容器需要監(jiān)聽的端口號
hostPort: int #容器所在主機需要監(jiān)聽的端口號,默認與Container相同
protocol: string #端口協(xié)議,支持TCP和UDP,默認TCP
env: #容器運行前需設(shè)置的環(huán)境變量列表
- name: string #環(huán)境變量名稱
value: string #環(huán)境變量的值
resources: #資源限制和請求的設(shè)置
limits: #資源限制的設(shè)置
cpu: string #Cpu的限制,單位為core數(shù),將用于docker run --cpu-shares參數(shù)
memory: string #內(nèi)存限制,單位可以為Mib/Gib,將用于docker run --memory參數(shù)
requests: #資源請求的設(shè)置
cpu: string #Cpu請求,容器啟動的初始可用數(shù)量
memory: string #內(nèi)存請求,容器啟動的初始可用數(shù)量
livenessProbe: #對Pod內(nèi)各容器健康檢查的設(shè)置,當探測無響應(yīng)幾次后將自動重啟該容器,檢查方法有exec、httpGet和tcpSocket,對一個容器只需設(shè)置其中一種方法即可
exec: #對Pod容器內(nèi)檢查方式設(shè)置為exec方式
command: [string] #exec方式需要制定的命令或腳本
httpGet: #對Pod內(nèi)個容器健康檢查方法設(shè)置為HttpGet,需要制定Path、port
path: string
port: number
host: string
scheme: string
HttpHeaders:
- name: string
value: string
tcpSocket: #對Pod內(nèi)個容器健康檢查方式設(shè)置為tcpSocket方式
port: number
initialDelaySeconds: 0 #容器啟動完成后首次探測的時間,單位為秒
timeoutSeconds: 0 #對容器健康檢查探測等待響應(yīng)的超時時間,單位秒,默認1秒
periodSeconds: 0 #對容器監(jiān)控檢查的定期探測時間設(shè)置,單位秒,默認10秒一次
successThreshold: 0
failureThreshold: 0
securityContext:
privileged: false
restartPolicy: [Always | Never | OnFailure] #Pod的重啟策略,Always表示一旦不管以何種方式終止運行,kubelet都將重啟,OnFailure表示只有Pod以非0退出碼退出才重啟,Nerver表示不再重啟該Pod
nodeSelector: obeject #設(shè)置NodeSelector表示將該Pod調(diào)度到包含這個label的node上,以key:value的格式指定
imagePullSecrets: #Pull鏡像時使用的secret名稱,以key:secretkey格式指定
- name: string
hostNetwork: false #是否使用主機網(wǎng)絡(luò)模式,默認為false,如果設(shè)置為true,表示使用宿主機網(wǎng)絡(luò)
volumes: #在該pod上定義共享存儲卷列表
- name: string #共享存儲卷名稱 (volumes類型有很多種)
emptyDir: {} #類型為emtyDir的存儲卷,與Pod同生命周期的一個臨時目錄。為空值
hostPath: string #類型為hostPath的存儲卷,表示掛載Pod所在宿主機的目錄
path: string #Pod所在宿主機的目錄,將被用于同期中mount的目錄
secret: #類型為secret的存儲卷,掛載集群與定義的secre對象到容器內(nèi)部
scretname: string
items:
- key: string
path: string
configMap: #類型為configMap的存儲卷,掛載預定義的configMap對象到容器內(nèi)部
name: string
items:
- key: string
path: string
附錄:
附錄1:k8s常見api操作
# 查看當前集群支持的API版本
$ curl -s 127.0.0.1:8080/api/ | jq -r .versions
# 查看指定API的資源操作類型 返回一個字典列表
$ curl -s 127.0.0.1:8080/api/v1/ | jq keys
[
"groupVersion",
"kind",
"resources"
]
# 查看當前k8s集群支持的操作類型
curl -s 127.0.0.1:8080/api/v1/ | jq .resources | grep "name\b"
"name": "bindings",
"name": "componentstatuses",
"name": "configmaps",
"name": "endpoints",
"name": "events",
"name": "limitranges",
"name": "namespaces",
"name": "namespaces/finalize",
"name": "namespaces/status",
"name": "nodes",
"name": "nodes/proxy",
"name": "nodes/status",
"name": "persistentvolumeclaims",
"name": "persistentvolumeclaims/status",
"name": "persistentvolumes",
"name": "persistentvolumes/status",
"name": "pods",
"name": "pods/attach",
"name": "pods/binding",
"name": "pods/eviction",
"name": "pods/exec",
"name": "pods/log",
"name": "pods/portforward",
"name": "pods/proxy",
"name": "pods/status",
"name": "podtemplates",
"name": "replicationcontrollers",
"name": "replicationcontrollers/scale",
"name": "replicationcontrollers/status",
"name": "resourcequotas",
"name": "resourcequotas/status",
"name": "secrets",
"name": "serviceaccounts",
"name": "services",
"name": "services/proxy",
"name": "services/status",
# 常用的,需要關(guān)注的幾個資源類型 【namespaces|nodes|pods|podtemplates|replicationcontrollers|services|secrets】
# 查看某個namespace的URL
$ curl -s 127.0.0.1:8080/api/v1/namespaces/ | jq .items | jq values[0] | grep -E '(name|selfLink)'
"name": "default",
"selfLink": "/api/v1/namespaces/default",
# 查看namespace詳情
$ curl -s 127.0.0.1:8080/api/v1/namespaces/default | jq .spec
{
"finalizers": [
"kubernetes"
]
}
# 查看全部node
$ curl -s 127.0.0.1:8080/api/v1/nodes | jq .items | grep -E "\bname\b|selfLink"
"name": "172.25.47.138",
"selfLink": "/api/v1/nodes/172.25.47.138",
"name": "172.25.47.202",
"selfLink": "/api/v1/nodes/172.25.47.202",
"name": "172.25.47.75",
"selfLink": "/api/v1/nodes/172.25.47.75",
"name": "10.0.0.121",
"selfLink": "/api/v1/nodes/10.0.0.121",
"name": "10.0.0.122",
"selfLink": "/api/v1/nodes/10.0.0.122",
# 查看某個node詳情【當前該節(jié)點的資源信息以及設(shè)置限制條件,當前節(jié)點運行的pod容器以及鏡像相關(guān)信息】
$ curl -s 127.0.0.1:8080/api/v1/nodes/10.0.0.122 | jq .status
{
"capacity": {
"cpu": "4",
"ephemeral-storage": "14987616Ki",
"hugepages-2Mi": "0",
"memory": "6627524Ki",
"pods": "110"
},
"allocatable": {
"cpu": "4",
"ephemeral-storage": "13812586883",
"hugepages-2Mi": "0",
"memory": "6525124Ki",
"pods": "110"
},
"conditions": [
{
"type": "OutOfDisk",
"status": "False",
"lastHeartbeatTime": "2018-07-19T03:15:50Z",
"lastTransitionTime": "2018-06-28T01:59:06Z",
"reason": "KubeletHasSufficientDisk",
"message": "kubelet has sufficient disk space available"
},
{
"type": "MemoryPressure",
"status": "False",
"lastHeartbeatTime": "2018-07-19T03:15:50Z",
"lastTransitionTime": "2018-06-28T01:59:06Z",
"reason": "KubeletHasSufficientMemory",
"message": "kubelet has sufficient memory available"
},
{
"type": "DiskPressure",
"status": "False",
"lastHeartbeatTime": "2018-07-19T03:15:50Z",
"lastTransitionTime": "2018-06-28T01:59:06Z",
"reason": "KubeletHasNoDiskPressure",
"message": "kubelet has no disk pressure"
},
{
"type": "PIDPressure",
"status": "False",
"lastHeartbeatTime": "2018-07-19T03:15:50Z",
"lastTransitionTime": "2018-06-28T01:59:06Z",
"reason": "KubeletHasSufficientPID",
"message": "kubelet has sufficient PID available"
},
{
"type": "Ready",
"status": "True",
"lastHeartbeatTime": "2018-07-19T03:15:50Z",
"lastTransitionTime": "2018-07-18T08:18:31Z",
"reason": "KubeletReady",
"message": "kubelet is posting ready status"
}
],
"addresses": [
{
"type": "InternalIP",
"address": "10.0.0.122"
},
{
"type": "Hostname",
"address": "10.0.0.122"
}
],
"daemonEndpoints": {
"kubeletEndpoint": {
"Port": 10250
}
},
"nodeInfo": {
"machineID": "cd7f16da1665b7cee74f46122e2d7cdd",
"systemUUID": "0E96FBAF-C214-18F8-D9E6-8B65B9B5A413",
"bootID": "f67f0185-4be6-41d0-b5e0-d126165f6986",
"kernelVersion": "3.10.0-327.el7.x86_64",
"osImage": "CentOS Linux 7 (Core)",
"containerRuntimeVersion": "docker://1.12.6",
"kubeletVersion": "v1.10.4",
"kubeProxyVersion": "v1.10.4",
"operatingSystem": "linux",
"architecture": "amd64"
},
"images": [
{
"names": [
"dockerhub.jd.com/wolf/fe-wolf@sha256:0b155e99f27de990cfd3a5f96bcc13b820530cf1236b96ca5d87efae4f65f62a",
"dockerhub.jd.com/wolf/fe-wolf:latest"
],
"sizeBytes": 352597644
},
{
"names": [
"xxbandy123/go-web@sha256:c375b479a74b0ec77608a221a28ddc4589149d55a04845eac697903add557c30",
"xxbandy123/go-web:latest"
],
"sizeBytes": 338268706
},
{
"names": [
"dockerhub.jd.com/gcr_mirror/pause-amd64@sha256:fcaff905397ba63fd376d0c3019f1f1cb6e7506131389edbcb3d22719f1ae54d",
"dockerhub.jd.com/gcr_mirror/pause-amd64:3.1"
],
"sizeBytes": 742472
}
]
}
# 查看pod相關(guān)信息【由于一般k8s是不建議使用pod類型來直接控制容器的,因此這個接口直接查沒什么意義】
# 查看apis group相關(guān)信息
curl -s 127.0.0.1:8080/apis/extensions/v1beta1 | grep "\bname\b"
"name": "daemonsets",
"name": "daemonsets/status",
"name": "deployments",
"name": "deployments/rollback",
"name": "deployments/scale",
"name": "deployments/status",
"name": "ingresses",
"name": "ingresses/status",
"name": "networkpolicies",
"name": "podsecuritypolicies",
"name": "replicasets",
"name": "replicasets/scale",
"name": "replicasets/status",
"name": "replicationcontrollers",
"name": "replicationcontrollers/scale",
$ curl -s 127.0.0.1:8080/apis/apps/v1beta1 | grep "\bname\b"
"name": "controllerrevisions",
"name": "deployments",
"name": "deployments/rollback",
"name": "deployments/scale",
"name": "deployments/status",
"name": "statefulsets",
"name": "statefulsets/scale",
"name": "statefulsets/status",
# 查看控制器對應(yīng)的控制器信息【statefulset,deployment】 這些控制器的信息一般在/apis/apps/下
# job和cronjob一般會在/apis/batch/下
$ curl -s 127.0.0.1:8080/apis/batch/v1 | grep "\bname\b"
"name": "jobs",
"name": "jobs/status",
curl -s 127.0.0.1:8080/apis/batch/v1/jobs | jq -r .items | grep name
"name": "image-build-service",
"namespace": "default",
"selfLink": "/apis/batch/v1/namespaces/default/jobs/image-build-service",
"job-name": "image-build-service"
"name": "image-build-service",
"job-name": "image-build-service"
"name": "hosts",
"name": "image-build-service",
"name": "dockerhost",
"name": "branch",
"name": "giturl",
"name": "app",
"name": "dockerfiledir",
"name": "hosts",
# 查看當前/apis/apps/支持的apiversion
$ curl -s 127.0.0.1:8080/apis/apps | jq .versions |grep "groupVersion"
"groupVersion": "apps/v1",
"groupVersion": "apps/v1beta2",
"groupVersion": "apps/v1beta1",
# 查看應(yīng)用的api支持的相關(guān)資源類型 【daemonsets|deployments|replicasets|statefulsets】
$ curl -s 127.0.0.1:8080/apis/apps/v1 | jq .resources | grep "name\b"
"name": "controllerrevisions",
"name": "daemonsets",
"name": "daemonsets/status",
"name": "deployments",
"name": "deployments/scale",
"name": "deployments/status",
"name": "replicasets",
"name": "replicasets/scale",
"name": "replicasets/status",
"name": "statefulsets",
"name": "statefulsets/scale",
"name": "statefulsets/status",
# 問題:job類型的資源在哪里查看 batch/extension?
# 查看deployment名稱和selflink
$ curl -s 127.0.0.1:8080/apis/apps/v1/deployments | jq .items | grep -E "(\bname\b|\bnamespace\b|selfLink)"
"name": "test-go-web",
"namespace": "default",
"selfLink": "/apis/apps/v1/namespaces/default/deployments/test-go-web",
# 查看某個deployment詳情
$ kubectl get deployments -o wide
NAME DESIRED CURRENT UP-TO-DATE AVAILABLE AGE CONTAINERS IMAGES SELECTOR
test-go-web 5 5 5 5 1h test-go-web xxbandy123/go-web run=test-go-web
$ curl -s 127.0.0.1:8080/apis/apps/v1/namespaces/default/deployments/test-go-web | jq .spec
{
"replicas": 5,
"selector": {
"matchLabels": {
"run": "test-go-web"
}
},
"template": {
"metadata": {
"creationTimestamp": null,
"labels": {
"run": "test-go-web"
}
},
"spec": {
"containers": [
{
"name": "test-go-web",
"image": "xxbandy123/go-web",
"ports": [
{
"containerPort": 9090,
"protocol": "TCP"
}
],
"resources": {},
"terminationMessagePath": "/dev/termination-log",
"terminationMessagePolicy": "File",
"imagePullPolicy": "Always"
}
],
"restartPolicy": "Always",
"terminationGracePeriodSeconds": 30,
"dnsPolicy": "ClusterFirst",
"securityContext": {},
"schedulerName": "default-scheduler"
}
},
"strategy": {
"type": "RollingUpdate",
"rollingUpdate": {
"maxUnavailable": 1,
"maxSurge": 1
}
},
"revisionHistoryLimit": 10,
"progressDeadlineSeconds": 600
}
# 查看service資源下的實例
# curl -s 127.0.0.1:8080/api/v1/services/ | jq ".items" | jq values[1] |grep -E "(\bname\b|selfLink|\bnamespace\b)"
"name": "test-go-web",
"namespace": "default",
"selfLink": "/api/v1/namespaces/default/services/test-go-web",
# 查看某個service詳情 【具體的資源類型會被劃分到namespaces下】
$ curl -s 127.0.0.1:8080/api/v1/namespaces/default/services/test-go-web | jq -r '.spec'
{
"ports": [
{
"protocol": "TCP",
"port": 9090,
"targetPort": 9090,
"nodePort": 30315
}
],
"selector": {
"run": "test-go-web"
},
"clusterIP": "10.254.141.49",
"type": "NodePort",
"sessionAffinity": "None",
"externalTrafficPolicy": "Cluster"
}
# 查看configmap
$ curl -s 127.0.0.1:8080/api/v1/configmaps | jq .items
# 查看secrets
$ curl -s 127.0.0.1:8080/api/v1/secrets | jq .items | grep "\"\bname\b"
附錄2:參考資料:
- pod 操作 https://blog.51cto.com/3241766/2420421
- k8s EN-doc https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.13
- k8s CN-doc https://www.kubernetes.org.cn/
- volumes http://www.itdecent.cn/p/d351c967298e
- Headless service + statefulSet