K8S容器探针
容器探针
探针是由 kubelet对容器执行的按期诊断。要执行诊断， kubelet 调用由容器实现的 Handler 。有三种类型的处理程序：html
- ExecAction ：在容器内执行指定命令。若是命令退出时返回码为 0 则认为诊断成功。
- TCPSocketAction ：对指定端口上的容器的 IP 地址进行 TCP 检查。若是端口打开，则诊断被认为是成功的。
- HTTPGetAction ：对指定的端口和路径上的容器的 IP 地址执行 HTTP Get 请求。若是响应的状态码大于等于 200 且小于 400 ，则诊断被认为是成功的
每次探测都将得到如下三种结果之一：node
- 成功：容器经过了诊断。
- 失败：容器未经过诊断。
- 未知：诊断失败，所以不会采起任何行动
探针的方式
- livenessProbe ：指示容器是否正在运行。若是存活探测失败，则 kubelet 会杀死容器，而且容器将受到其 重启策略 的影响。若是容器不提供存活探针，则默认状态为 Success
- readinessProbe ：指示容器是否准备好服务请求。若是就绪探测失败，端点控制器将从与 Pod 匹配的全部 Service 的端点中删除该 Pod 的 IP 地址。初始延迟以前的就绪状态默认为 Failure 。若是容器不提供就绪探针，则默认状态为 Success。
测试
检测探针 - 就绪检测linux
read.yamlnginx
[root@k8s-master mnt]# cat read.yaml
apiVersion: v1
kind: Pod
metadata:
name: readiness-httpget-pod
namespace: default
spec:
containers:
- name: readiness-httpget-container
image: wangyanglinux/myapp:v1
imagePullPolicy: IfNotPresent
readinessProbe:
httpGet:
port: 80
path: /index1.html
initialDelaySeconds: 1
periodSeconds: 3
[root@k8s-master mnt]#
[root@k8s-master mnt]# vim read.yaml
[root@k8s-master mnt]# kubectl create -f read.yaml
pod/readiness-httpget-pod created
[root@k8s-master mnt]# kubectl get pod
NAME READY STATUS RESTARTS AGE
myapp-pod 1/1 Running 0 70m
readiness-httpget-pod 0/1 Running 0 17s
[root@k8s-master mnt]# kubectl describe pod readiness-httpget-pod
Name: readiness-httpget-pod
Namespace: default
Priority: 0
Node: k8s-node01/192.168.180.133
Start Time: Wed, 18 Dec 2019 23:12:59 +0800
Labels: <none>
Annotations: <none>
Status: Running
IP: 10.244.2.10
IPs:
IP: 10.244.2.10
Containers:
readiness-httpget-container:
Container ID: docker://566ff6cdcf44daaba316b796fb8bf6f9563ddd44000c9ae9f572fd0a6719684c
Image: wangyanglinux/myapp:v1
Image ID: docker-pullable://wangyanglinux/myapp@sha256:9c3dc30b5219788b2b8a4b065f548b922a34479577befb54b03330999d30d513
Port: <none>
Host Port: <none>
State: Running
Started: Wed, 18 Dec 2019 23:13:01 +0800
Ready: False
Restart Count: 0
Readiness: http-get http://:80/index1.html delay=1s timeout=1s period=3s #success=1 #failure=3
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from default-token-gx2h8 (ro)
Conditions:
Type Status
Initialized True
Ready False
ContainersReady False
PodScheduled True
Volumes:
default-token-gx2h8:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-gx2h8
Optional: false
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled <unknown> default-scheduler Successfully assigned default/readiness-httpget-pod to k8s-node01
Normal Pulled 66s kubelet, k8s-node01 Container image "wangyanglinux/myapp:v1" already present on machine
Normal Created 66s kubelet, k8s-node01 Created container readiness-httpget-container
Normal Started 66s kubelet, k8s-node01 Started container readiness-httpget-container
Warning Unhealthy 0s (x22 over 63s) kubelet, k8s-node01 Readiness probe failed: HTTP probe failed with statuscode: 404
[root@k8s-master mnt]# kubectl exec readiness-httpget-pod -it /bin/sh
/ # ls
bin dev etc home lib media mnt proc root run sbin srv sys tmp usr var
/ # cd /usr/share/nginx
/usr/share/nginx # ls -l
total 0
drwxr-xr-x 1 root root 24 Feb 25 2018 html
/usr/share/nginx # cd html/
/usr/share/nginx/html # ls -l
total 8
-rw-r--r-- 1 root root 537 Jan 10 2018 50x.html
-rw-r--r-- 1 root root 65 Mar 2 2018 index.html
/usr/share/nginx/html # cat index.html
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>
/usr/share/nginx/html # echo "123" >> index1.html
/usr/share/nginx/html # exit
[root@k8s-master mnt]# kubectl get pod
NAME READY STATUS RESTARTS AGE
myapp-pod 1/1 Running 1 73m
readiness-httpget-pod 1/1 Running 0 3m41s
说明：因为index1.html不存在，会致使他重启，手动建立后就正常了。docker
检测探针 - 存活检测vim
[root@k8s-master mnt]# cat live-exec.yaml
apiVersion: v1
kind: Pod
metadata:
name: liveness-exec-pod
namespace: default
spec:
containers:
- name: liveness-exec-container
image: busybox
imagePullPolicy: IfNotPresent
command: ["/bin/sh","-c","touch /tmp/live ; sleep 60; rm -rf /tmp/live; sleep 3600"]
livenessProbe:
exec:
command: ["test","-e","/tmp/live"]
initialDelaySeconds: 1
periodSeconds: 3
[root@k8s-master mnt]#
[root@k8s-master mnt]# vim live-exec.yaml [root@k8s-master mnt]# kubectl create -f live-exec.yaml pod/liveness-exec-pod created [root@k8s-master mnt]# kubectl get pod -w NAME READY STATUS RESTARTS AGE liveness-exec-pod 1/1 Running 0 25s myapp-pod 1/1 Running 1 81m readiness-httpget-pod 1/1 Running 0 11m liveness-exec-pod 1/1 Running 1 101s liveness-exec-pod 1/1 Running 2 3m19s ^Z [1]+ 已中止 kubectl get pod -w
说明：因为/tmp/live不存在，会一直重启api
[root@k8s-master mnt]# cat live-http.yaml
apiVersion: v1
kind: Pod
metadata:
name: liveness-httpget-pod
namespace: default
spec:
containers:
- name: liveness-httpget-container
image: wangyanglinux/myapp:v1
imagePullPolicy: IfNotPresent
ports:
- name: http
containerPort: 80
livenessProbe:
httpGet:
port: http
path: /index.html
initialDelaySeconds: 1
periodSeconds: 3
timeoutSeconds: 10
[root@k8s-master mnt]#
[root@k8s-master mnt]# kubectl create -f live-http.yaml pod/liveness-httpget-pod created [root@k8s-master mnt]# kubectl get pod NAME READY STATUS RESTARTS AGE liveness-httpget-pod 1/1 Running 0 14s myapp-pod 1/1 Running 1 90m readiness-httpget-pod 1/1 Running 0 20m [root@k8s-master mnt]# kubectl get pod -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES liveness-httpget-pod 1/1 Running 0 26s 10.244.2.12 k8s-node01 <none> <none> myapp-pod 1/1 Running 1 90m 10.244.1.9 k8s-node02 <none> <none> readiness-httpget-pod 1/1 Running 0 20m 10.244.2.10 k8s-node01 <none> <none> [root@k8s-master mnt]# curl 10.244.2.12 Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a> [root@k8s-master mnt]# curl 10.244.2.12/index.html Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a> [root@k8s-master mnt]# kubectl exec liveness-httpget-pod -it -- /bin/sh / # cd /usr/share/nginx/html/ /usr/share/nginx/html # ls -l total 8 -rw-r--r-- 1 root root 537 Jan 10 2018 50x.html -rw-r--r-- 1 root root 65 Mar 2 2018 index.html /usr/share/nginx/html # rm -rf index.html /usr/share/nginx/html # exit [root@k8s-master mnt]# kubectl get pod NAME READY STATUS RESTARTS AGE liveness-httpget-pod 1/1 Running 0 2m24s myapp-pod 1/1 Running 1 92m readiness-httpget-pod 1/1 Running 0 22m [root@k8s-master mnt]# kubectl get pod NAME READY STATUS RESTARTS AGE liveness-httpget-pod 1/1 Running 1 2m41s myapp-pod 1/1 Running 1 92m readiness-httpget-pod 1/1 Running 0 22m
说明：删除Html,会发现Pod开始重启了。bash
[root@k8s-master mnt]# cat live-tcp.yaml
apiVersion: v1
kind: Pod
metadata:
name: probe-tcp
spec:
containers:
- name: nginx
image: wangyanglinux/myapp:v1
livenessProbe:
initialDelaySeconds: 5
timeoutSeconds: 1
tcpSocket:
port: 8080
periodSeconds: 3
[root@k8s-master mnt]#
[root@k8s-master mnt]# vim live-tcp.yaml [root@k8s-master mnt]# kubectl create -f live-tcp.yaml pod/probe-tcp created [root@k8s-master mnt]# kubectl get pod NAME READY STATUS RESTARTS AGE liveness-httpget-pod 1/1 Running 1 9m24s myapp-pod 1/1 Running 1 99m probe-tcp 1/1 Running 0 5s readiness-httpget-pod 1/1 Running 0 29m [root@k8s-master mnt]# kubectl get pod NAME READY STATUS RESTARTS AGE liveness-httpget-pod 1/1 Running 1 9m37s myapp-pod 1/1 Running 1 99m probe-tcp 1/1 Running 1 18s readiness-httpget-pod 1/1 Running 0 29m [root@k8s-master mnt]# kubectl get pod NAME READY STATUS RESTARTS AGE liveness-httpget-pod 1/1 Running 1 9m41s myapp-pod 1/1 Running 1 99m probe-tcp 1/1 Running 1 22s readiness-httpget-pod 1/1 Running 0 29m [root@k8s-master mnt]# kubectl get pod NAME READY STATUS RESTARTS AGE liveness-httpget-pod 1/1 Running 1 9m43s myapp-pod 1/1 Running 1 99m probe-tcp 1/1 Running 1 24s readiness-httpget-pod 1/1 Running 0 29m [root@k8s-master mnt]# kubectl get pod NAME READY STATUS RESTARTS AGE liveness-httpget-pod 1/1 Running 1 9m44s myapp-pod 1/1 Running 1 99m probe-tcp 1/1 Running 1 25s readiness-httpget-pod 1/1 Running 0 29m [root@k8s-master mnt]# kubectl get pod NAME READY STATUS RESTARTS AGE liveness-httpget-pod 1/1 Running 1 10m myapp-pod 1/1 Running 1 100m probe-tcp 1/1 Running 3 47s readiness-httpget-pod 1/1 Running 0 29m [root@k8s-master mnt]# kubectl delete -f live-tcp.yaml pod "probe-tcp" deleted [root@k8s-master mnt]#
说明：删除Html,会发现Pod开始重启了。网络
Pod hook
Pod hook （钩子）是由 Kubernetes 管理的 kubelet 发起的，当容器中的进程启动前或者容器中的进
程终止以前运行，这是包含在容器的生命周期之中。能够同时为 Pod 中的全部容器都配置 hook
Hook 的类型包括两种：
exec ：执行一段命令
HTTP ：发送 HTTP 请求app
重启策略
PodSpec 中有一个 restartPolicy 字段，可能的值为 Always 、 OnFailure 和 Never 。默认为
Always 。 restartPolicy 适用于 Pod 中的全部容器。 restartPolicy 仅指经过同一节点上的
kubelet 从新启动容器。失败的容器由 kubelet 以五分钟为上限的指数退避延迟（ 10 秒， 20 秒， 40
秒 ... ）从新启动，并在成功执行十分钟后重置。如 Pod 文档 中所述，一旦绑定到一个节点， Pod 将
永远不会从新绑定到另外一个节点。
Pod phase
Pod 的 status 字段是一个 PodStatus 对象， PodStatus 中有一个 phase 字段。
Pod 的相位（ phase ）是 Pod 在其生命周期中的简单宏观概述。该阶段并非对容器或 Pod 的综合汇总，也不是为了作为综合状态机
Pod 相位的数量和含义是严格指定的。除了本文档中列举的状态外，不该该再假定 Pod 有其余的phase 值
几种常见的值
- 挂起（ Pending ）： Pod 已被 Kubernetes 系统接受，但有一个或者多个容器镜像还没有建立。等待时间包括调度 Pod 的时间和经过网络下载镜像的时间，这可能须要花点时间
- 运行中（ Running ）：该 Pod 已经绑定到了一个节点上， Pod 中全部的容器都已被建立。至少有一个容器正在运行，或者正处于启动或重启状态
- 成功（ Succeeded ）： Pod 中的全部容器都被成功终止，而且不会再重启
- 失败（ Failed ）： Pod 中的全部容器都已终止了，而且至少有一个容器是由于失败终止。也就是说，容器以非 0 状态退出或者被系统终止
- 未知（ Unknown ）：由于某些缘由没法取得 Pod 的状态，一般是由于与 Pod 所在主机通讯失败
[root@k8s-master mnt]# vim post.yaml
[root@k8s-master mnt]# kubectl create -f post.yaml
pod/lifecycle-demo created
[root@k8s-master mnt]# kubectl get pod
NAME READY STATUS RESTARTS AGE
lifecycle-demo 1/1 Running 0 9s
liveness-httpget-pod 1/1 Running 1 40m
myapp-pod 1/1 Running 1 130m
readiness-httpget-pod 1/1 Running 0 60m
[root@k8s-master mnt]# kubectl exec lifecycle-demo -it -- /bin/bash
OCI runtime exec failed: exec failed: container_linux.go:346: starting container process caused "exec: \"/bin/bash\": stat /bin/bash: no such file or directory": unknown
command terminated with exit code 126
[root@k8s-master mnt]# kubectl exec lifecycle-demo -it -- /bin/sh
/ # cd /usr/share/message
/bin/sh: cd: can't cd to /usr/share/message
/ # cat /usr/share/message
Hello from the postStart handler
/ # exit
[root@k8s-master mnt]# cat post.yaml
apiVersion: v1
kind: Pod
metadata:
name: lifecycle-demo
spec:
containers:
- name: lifecycle-demo-container
image: wangyanglinux/myapp:v1
lifecycle:
postStart:
exec:
command: ["/bin/sh", "-c", "echo Hello from the postStart handler > /usr/share/message"]
preStop:
exec:
command: ["/bin/sh", "-c", "echo Hello from the poststop handler > /usr/share/message"]
[root@k8s-master mnt]#
原文连接：http://dwz.date/MMr
相关文章
- 1. k8s - Container probe 容器探针
- 2. k8s的容器监测探针
- 3. kubernetes容器探针检测
- 4. Kubernetes集群——（k8s）pod的生命周期+init容器+探针
- 5. kubernetes(k8s) 之pod的生命周期（init容器和探针)
- 6. k8s实践(五)：容器探针(liveness and readiness probe)
- 7. kubernetes(k8s)—Pod生命周期介绍，init容器，探针
- 8. k8s 存活探针LivenessProbe
- 9. 容器探针(liveness and readiness probe)
- 10. k8s探针实现grpc健康检查
- 更多相关文章...
- • Docker 容器连接 - Docker教程
- • Docker 容器使用 - Docker教程
- • Docker容器实战(七) - 容器眼光下的文件系统
- • Docker容器实战(六) - 容器的隔离与限制
相关标签/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
最新文章
- 1. 跳槽面试的几个实用小技巧，不妨看看！
- 2. Mac实用技巧 |如何使用Mac系统中自带的预览工具将图片变成黑白色?
- 3. Mac实用技巧 |如何使用Mac系统中自带的预览工具将图片变成黑白色?
- 4. 如何使用Mac系统中自带的预览工具将图片变成黑白色?
- 5. Mac OS非兼容Windows软件运行解决方案——“以VMware & Microsoft Access为例“
- 6. 封装 pyinstaller -F -i b.ico excel.py
- 7. 数据库作业三ER图待完善
- 8. nvm安装使用低版本node.js(非命令安装)
- 9. 如何快速转换图片格式
- 10. 将表格内容分条转换为若干文档
欢迎关注本站公众号,获取更多信息
相关文章
- 1. k8s - Container probe 容器探针
- 2. k8s的容器监测探针
- 3. kubernetes容器探针检测
- 4. Kubernetes集群——（k8s）pod的生命周期+init容器+探针
- 5. kubernetes(k8s) 之pod的生命周期（init容器和探针)
- 6. k8s实践(五)：容器探针(liveness and readiness probe)
- 7. kubernetes(k8s)—Pod生命周期介绍，init容器，探针
- 8. k8s 存活探针LivenessProbe
- 9. 容器探针(liveness and readiness probe)
- 10. k8s探针实现grpc健康检查