kubernetes 之Health Check 健康检查
默认的健康检查
这里Pod的restartPolicy设置为OnFailure,默认为Always. [machangwei@mcwk8s-master ~]$ cat mcwHealthcheck.yml #下面这个配置存在问题 apiVersion: v1 kind: Pod metadata: name: mcw-healthcheck labels: test: mcw-healthcheck spec: restartPolicy: OnFailure selector: matchLabels: test: mcw-healthcheck containers: - name: mcw-healthcheck image: busybox args: - /bin/sh - -c - sleep 10;exit 1 [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwH mcwHealthcheck.yml mcwHttpdService.yml mcwHttpd.yml [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwHealthcheck.yml error: error validating "mcwHealthcheck.yml": error validating data: ValidationError(Pod.spec): unknown field "selector" in io.k8s.api.core.v1.PodSpec; if you choose to ignore these errors, turn validation off with --validate=false [machangwei@mcwk8s-master ~]$ #上面的配置存在问题，把选择器删掉，应该是POD类型资源没有选择器这个参数，pod不需要去匹配 [machangwei@mcwk8s-master ~]$ #标签，然后做指定标签做什么操作吧 [machangwei@mcwk8s-master ~]$ vim mcwHealthcheck.yml [machangwei@mcwk8s-master ~]$ cat mcwHealthcheck.yml apiVersion: v1 kind: Pod metadata: name: mcw-healthcheck labels: test: mcw-healthcheck spec: restartPolicy: OnFailure containers: - name: mcw-healthcheck image: busybox args: - /bin/sh - -c - sleep 10;exit 1 [machangwei@mcwk8s-master ~]#上面是正确的配置。restartPolicy为 OnFailure，也就是容器进程返回值非零， [machangwei@mcwk8s-master ~]#则认为容器发生故障需要重启。命令参数的意思是，模拟容器启动10秒后发生故障，退出非0 [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwHealthcheck.yml #部署pod pod/mcw-healthcheck created [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck #查看这个容器，现在珍藏运行 NAME READY STATUS RESTARTS AGE mcw-healthcheck 1/1 Running 0 31s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck #容器退出，发生错误 NAME READY STATUS RESTARTS AGE mcw-healthcheck 0/1 Error 0 45s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck #容器重启1次，又成功运行 NAME READY STATUS RESTARTS AGE mcw-healthcheck 1/1 Running 1 (22s ago) 62s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck #容器再次退出 NAME READY STATUS RESTARTS AGE mcw-healthcheck 0/1 Error 1 (38s ago) 78s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck # NAME READY STATUS RESTARTS AGE mcw-healthcheck 0/1 CrashLoopBackOff 1 (24s ago) 90s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck #再次成功运行 NAME READY STATUS RESTARTS AGE mcw-healthcheck 1/1 Running 2 (41s ago) 107s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck NAME READY STATUS RESTARTS AGE mcw-healthcheck 1/1 Running 3 (44s ago) 2m33s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck #重启过3次 NAME READY STATUS RESTARTS AGE mcw-healthcheck 0/1 Error 3 (53s ago) 2m42s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck #重启过8次 NAME READY STATUS RESTARTS AGE mcw-healthcheck 0/1 CrashLoopBackOff 8 (3m19s ago) 23m [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-healthcheck --show-labels #查看我们设置的标签 NAME READY STATUS RESTARTS AGE LABELS mcw-healthcheck 0/1 CrashLoopBackOff 9 (4m37s ago) 30m test=mcw-healthcheck
Liveness探测，
存活探测吧。
存活探测几次失败后，容器会被杀掉重启。
1、错误的配置文件
[machangwei@mcwk8s-master ~]$ cat mcwLiveness.yml #下面有问题，不知道为啥， apiVersion: v1 kind: Pod metadata: name: mcw-liveness labels: test: mcw-liveness spec: restartPolicy: OnFailure containers: - name: mcw-liveness image: busybox args: - /bin/sh - -c - touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 600 livenessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 periodSenconds: 5 [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwLiveness.yml error: error validating "mcwLiveness.yml": error validating data: ValidationError(Pod.spec.containers[0].livenessProbe): unknown field "periodSenconds" in io.k8s.api.core.v1.Probe; if you choose to ignore these errors, turn validation off with --validate=false [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ vim mcwLiveness.yml [machangwei@mcwk8s-master ~]$ cat mcwLiveness.yml #将存活探测间隔时间改成超时时间，并没有实现想要的效果 apiVersion: v1 kind: Pod metadata: name: mcw-liveness labels: test: mcw-liveness spec: restartPolicy: OnFailure containers: - name: mcw-liveness image: busybox args: - /bin/sh - -c - touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 600 livenessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 timeoutSeconds: 5 [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwLiveness.yml pod/mcw-liveness created [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 32s mcw-liveness 1/1 Running 0 114s [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ vim mcwLiveness.yml [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ cat mcwLiveness.yml #修改sleep 600为120， apiVersion: v1 kind: Pod metadata: name: mcw-liveness labels: test: mcw-liveness spec: restartPolicy: OnFailure containers: - name: mcw-liveness image: busybox args: - /bin/sh - -c - touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 120 livenessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 timeoutSeconds: 5 [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwLiveness.yml #pod重新部署不行，还是有存活探针的重新部署不行呢？ The Pod "mcw-liveness" is invalid: spec: Forbidden: pod updates may not change fields other than `spec.containers[*].image`, `spec.initContainers[*].image`, `spec.activeDeadlineSeconds`, `spec.tolerations` (only additions to existing tolerations) or `spec.terminationGracePeriodSeconds` (allow it to be set to 1 if it was previously negative) core.PodSpec{ Volumes: {{Name: "kube-api-access-cv2xz", VolumeSource: {Projected: &{Sources: {{ServiceAccountToken: &{ExpirationSeconds: 3607, Path: "token"}}, {ConfigMap: &{LocalObjectReference: {Name: "kube-root-ca.crt"}, Items: {{Key: "ca.crt", Path: "ca.crt"}}}}, {DownwardAPI: &{Items: {{Path: "namespace", FieldRef: &{APIVersion: "v1", FieldPath: "metadata.namespace"}}}}}}, DefaultMode: &420}}}}, InitContainers: nil, Containers: []core.Container{ { Name: "mcw-liveness", Image: "busybox", Command: nil, Args: []string{ "/bin/sh", "-c", - "touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 120", + "touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 600", }, WorkingDir: "", Ports: nil, ... // 16 identical fields }, }, EphemeralContainers: nil, RestartPolicy: "OnFailure", ... // 26 identical fields } [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl delete -f mcwLiveness.yml #那就删掉重新部署吧。 pod "mcw-liveness" deleted [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwLiveness.yml pod/mcw-liveness created [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness #没看错啥效果，我应该看日志的，唉，忘记了 NAME READY STATUS RESTARTS AGE mcw-liveness 0/1 ContainerCreating 0 15s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 21s mcw-liveness 1/1 Running 2 (115s ago) 5m15s [machangwei@mcwk8s-master ~]$
2、正确的配置文件
[machangwei@mcwk8s-master ~]$ ls mcwHealthcheck.yml mcwhttpd2quanyml mcwHttpdService.yml mcw.httpd.v16.yml mcw.httpd.v17.yml mcw.httpd.v18.yml mcwHttpd.yml mcwLiveness.yml mm.yml [machangwei@mcwk8s-master ~]$ vim mcwLiveness.yml [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ cat mcwLiveness.yml #重新修改为间隔时间参数，值为5s。这看着和前面也没啥区别，这个是好的，这是为啥？ apiVersion: v1 kind: Pod metadata: name: mcw-liveness labels: test: mcw-liveness spec: restartPolicy: OnFailure containers: - name: mcw-liveness image: busybox args: - /bin/sh - -c - touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 120 livenessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 periodSeconds: 5 [machangwei@mcwk8s-master ~]$ kubectl delete -f mcwLiveness.yml pod "mcw-liveness" deleted [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwLiveness.yml pod/mcw-liveness created [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 23s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness #查看一直显示运行呢 NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 26s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 30s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 31s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 33s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 34s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 36s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 39s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 40s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 42s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 43s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 45s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 47s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 48s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 49s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 51s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 54s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 55s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 57s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 60s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 63s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 65s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 66s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 68s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 70s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 74s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 76s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 78s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 0 81s [machangwei@mcwk8s-master ~]$ kubectl describe pod mcw-liveness #看下它的具体信息 Name: mcw-liveness Namespace: default Priority: 0 Node: mcwk8s-node2/10.0.0.6 Start Time: Sat, 22 Jan 2022 09:47:53 +0800 Labels: test=mcw-liveness Annotations: <none> Status: Running IP: 10.244.2.20 IPs: IP: 10.244.2.20 Containers: mcw-liveness: Container ID: docker://a2c603fc88bd32bd021c05234a719c11a8fe2fb31a1db1b90433bdfd5c1ebecc Image: busybox Image ID: docker-pullable://busybox@sha256:5acba83a746c7608ed544dc1533b87c737a0b0fb730301639a0179f9344b1678 Port: <none> Host Port: <none> Args: /bin/sh -c touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 120 State: Running Started: Sat, 22 Jan 2022 09:48:10 +0800 Ready: True Restart Count: 0 Liveness: exec [cat /tmp/healthy] delay=10s timeout=1s period=5s #success=1 #failure=3 Environment: <none> Mounts: /var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-lhb6w (ro) Conditions: Type Status Initialized True Ready True ContainersReady True PodScheduled True Volumes: kube-api-access-lhb6w: Type: Projected (a volume that contains injected data from multiple sources) TokenExpirationSeconds: 3607 ConfigMapName: kube-root-ca.crt ConfigMapOptional: <nil> DownwardAPI: true QoS Class: BestEffort Node-Selectors: <none> Tolerations: node.kubernetes.io/not-ready:NoExecute op=Exists for 300s node.kubernetes.io/unreachable:NoExecute op=Exists for 300s Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 97s default-scheduler Successfully assigned default/mcw-liveness to mcwk8s-node2 Normal Pulled 81s kubelet Successfully pulled image "busybox" in 15.677393047s Normal Created 81s kubelet Created container mcw-liveness Normal Started 81s kubelet Started container mcw-liveness Warning Unhealthy 37s (x3 over 47s) kubelet Liveness probe failed: cat: can't open '/tmp/healthy': No such file or directory Normal Killing 37s kubelet Container mcw-liveness failed liveness probe, will be restarted Normal Pulling 7s (x2 over 97s) kubelet Pulling image "busybox" [machangwei@mcwk8s-master ~]$ #由上可知，某个时间段存活探测了三次失败x3，然后杀掉容器，重新拉取镜像要重新部署容器了，这拉取镜像是第二次拉取镜像。x2 [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness #下面再看，显示pod已经重启了一次了。 NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 1 (25s ago) 116s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 1 (49s ago) 2m20s [machangwei@mcwk8s-master ~]$ kubectl describe pod mcw-liveness #再次查看pod详情 ....... Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 2m24s default-scheduler Successfully assigned default/mcw-liveness to mcwk8s-node2 Normal Pulled 2m8s kubelet Successfully pulled image "busybox" in 15.677393047s Normal Killing 84s kubelet Container mcw-liveness failed liveness probe, will be restarted Normal Pulling 54s (x2 over 2m24s) kubelet Pulling image "busybox" Normal Created 39s (x2 over 2m8s) kubelet Created container mcw-liveness Normal Pulled 39s kubelet Successfully pulled image "busybox" in 15.73643804s Normal Started 38s (x2 over 2m8s) kubelet Started container mcw-liveness Warning Unhealthy 4s (x4 over 94s) kubelet Liveness probe failed: cat: can't open '/tmp/healthy': No such file or directory [machangwei@mcwk8s-master ~]$ #由上可知，第二轮探测失败开始了。第四次探测失败。重新生成的容器肯定是有文件的。但是后面又被删掉了 [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness NAME READY STATUS RESTARTS AGE mcw-liveness 1/1 Running 1 (115s ago) 3m26s [machangwei@mcwk8s-master ~]$ kubectl describe pod mcw-liveness #过一会再看pod详情 Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 4m20s default-scheduler Successfully assigned default/mcw-liveness to mcwk8s-node2 Normal Pulled 4m4s kubelet Successfully pulled image "busybox" in 15.677393047s Normal Pulled 2m35s kubelet Successfully pulled image "busybox" in 15.73643804s Normal Created 2m35s (x2 over 4m4s) kubelet Created container mcw-liveness Normal Started 2m34s (x2 over 4m4s) kubelet Started container mcw-liveness Warning Unhealthy 110s (x6 over 3m30s) kubelet Liveness probe failed: cat: can't open '/tmp/healthy': No such file or directory Normal Killing 110s (x2 over 3m20s) kubelet Container mcw-liveness failed liveness probe, will be restarted Normal Pulling 80s (x3 over 4m20s) kubelet Pulling image "busybox" Warning Failed 11s kubelet Failed to pull image "busybox": rpc error: code = Unknown desc = Error response from daemon: Get "https://registry-1.docker.io/v2/": dial tcp: lookup registry-1.docker.io on 223.5.5.5:53: read udp 10.0.0.6:45742->223.5.5.5:53: i/o timeout Warning Failed 11s kubelet Error: ErrImagePull Normal BackOff 10s kubelet Back-off pulling image "busybox" Warning Failed 10s kubelet Error: ImagePullBackOff [machangwei@mcwk8s-master ~]$ kubectl describe pod mcw-liveness ......#由上可知拉取镜像三次了，第三次拉取失败了，下面继续拉取第四次，pulling x4， Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 5m31s default-scheduler Successfully assigned default/mcw-liveness to mcwk8s-node2 Normal Pulled 5m15s kubelet Successfully pulled image "busybox" in 15.677393047s Normal Pulled 3m46s kubelet Successfully pulled image "busybox" in 15.73643804s Normal Created 3m46s (x2 over 5m15s) kubelet Created container mcw-liveness Normal Started 3m45s (x2 over 5m15s) kubelet Started container mcw-liveness Warning Unhealthy 3m1s (x6 over 4m41s) kubelet Liveness probe failed: cat: can't open '/tmp/healthy': No such file or directory Normal Killing 3m1s (x2 over 4m31s) kubelet Container mcw-liveness failed liveness probe, will be restarted Warning Failed 82s kubelet Failed to pull image "busybox": rpc error: code = Unknown desc = Error response from daemon: Get "https://registry-1.docker.io/v2/": dial tcp: lookup registry-1.docker.io on 223.5.5.5:53: read udp 10.0.0.6:45742->223.5.5.5:53: i/o timeout Warning Failed 82s kubelet Error: ErrImagePull Normal BackOff 81s kubelet Back-off pulling image "busybox" Warning Failed 81s kubelet Error: ImagePullBackOff Normal Pulling 67s (x4 over 5m31s) kubelet Pulling image "busybox" [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness #查看pod状态，镜像拉取失败的状态 NAME READY STATUS RESTARTS AGE mcw-liveness 0/1 ErrImagePull 1 (2m48s ago) 5m49s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-liveness -o wide #这次拉取成功，容器又重新运行了 NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES mcw-liveness 1/1 Running 2 (3m25s ago) 6m26s 10.244.2.20 mcwk8s-node2 <none> <none> [machangwei@mcwk8s-master ~]$
3、livenss配置文件简介
[machangwei@mcwk8s-master ~]$ cat mcwLiveness.yml apiVersion: v1 kind: Pod metadata: name: mcw-liveness labels: test: mcw-liveness spec: restartPolicy: OnFailure #失败的时候重启 containers: - name: mcw-liveness image: busybox args: - /bin/sh #模拟故障。容器运行后，创建文件，30秒后删除文件，后面这个睡眠时间不清楚是干啥的。 - -c #下面的探测是容器启动10秒后就探测，在30秒删除之前，也就是会有探测成功的，被删除之后才会探测失败 - touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 120 livenessProbe: #添加存活探针的参数。根据执行命令的返回值判断。返回值为0，正常，返回值非0，探测失败 exec: #这里好像默认连续探测三次失败，杀掉容器，重建容器 command: # - cat - /tmp/healthy initialDelaySeconds: 10 #容器启动10秒之后执行存活探测。应该根据应用启动时间来决定这个时间。在应用成功启动的时间之后进行探测 periodSeconds: 5 #指定探测的间隔时间，连续执行三个探测失败，杀掉容器重新建立 [machangwei@mcwk8s-master ~]$ cat mcwLiveness.yml apiVersion: v1 kind: Pod metadata: name: mcw-liveness labels: test: mcw-liveness spec: restartPolicy: OnFailure containers: - name: mcw-liveness image: busybox args: - /bin/sh - -c - touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 120 livenessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 periodSeconds: 5
Readiness探测
删不掉，强制删除 mcw-readiness 0/1 Terminating 0 17m [machangwei@mcwk8s-master ~]$ kubectl delete pod mcw-readiness --grace-period=0 --force warning: Immediate deletion does not wait for confirmation that the running resource has been terminated. The resource may continue to run on the cluster indefinitely. pod "mcw-readiness" force deleted [machangwei@mcwk8s-master ~]$ ls mcwHealthcheck.yml mcwhttpd2quanyml mcwHttpdService.yml mcw.httpd.v16.yml mcw.httpd.v17.yml mcw.httpd.v18.yml mcwHttpd.yml mcwLiveness.yml mcwReadiness.yml mm.yml [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwReadiness.yml pod/mcw-readiness created [machangwei@mcwk8s-master ~]$ cat mcwReadiness.yml #跟liveness配置一样，只是将探针改成readinessProbe apiVersion: v1 kind: Pod metadata: name: mcw-readiness labels: test: mcw-readiness spec: restartPolicy: OnFailure containers: - name: mcw-readiness image: busybox args: - /bin/sh - -c - touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 120 readinessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 periodSeconds: 5 [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-readiness NAME READY STATUS RESTARTS AGE mcw-readiness 0/1 Running 0 24s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-readiness #运行成功 NAME READY STATUS RESTARTS AGE mcw-readiness 1/1 Running 0 32s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-readiness NAME READY STATUS RESTARTS AGE mcw-readiness 1/1 Running 0 44s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-readiness NAME READY STATUS RESTARTS AGE mcw-readiness 1/1 Running 0 49s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-readiness NAME READY STATUS RESTARTS AGE mcw-readiness 1/1 Running 0 54s [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-readiness #连续三次探测失败，设置ready为0，即ready设置为不可用 NAME READY STATUS RESTARTS AGE mcw-readiness 0/1 Running 0 64s [machangwei@mcwk8s-master ~]$ kubectl describe pod mcw-readiness #查看pod详情 ........ Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 90s default-scheduler Successfully assigned default/mcw-readiness to mcwk8s-node2 Normal Pulling 89s kubelet Pulling image "busybox" Normal Pulled 73s kubelet Successfully pulled image "busybox" in 15.706951792s Normal Created 73s kubelet Created container mcw-readiness Normal Started 73s kubelet Started container mcw-readiness Warning Unhealthy 0s (x10 over 40s) kubelet Readiness probe failed: cat: can't open '/tmp/healthy': No such file or directory [machangwei@mcwk8s-master ~]$ #如上，探测到失败，没有那么文件，因为已经被删除掉了，这适合 [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-readiness NAME READY STATUS RESTARTS AGE mcw-readiness 0/1 Running 0 2m35s [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl get pod mcw-readiness NAME READY STATUS RESTARTS AGE mcw-readiness 0/1 Completed 0 24m
Health Check 在 Scale Up中的应用。httpget探测方式
放弃了，回头研究
[machangwei@mcwk8s-master ~]$ cat mcwhttpd2quanyml apiVersion: apps/v1 kind: Deployment metadata: name: mcw-httpd2 namespace: kube-public spec: replicas: 3 selector: matchLabels: run: mcw-httpd2 template: metadata: labels: run: mcw-httpd2 spec: containers: - name: mcw-httpd2 image: httpd ports: - containerPort: 80 --- apiVersion: v1 kind: Service metadata: name: httpd2-svc namespace: kube-public spec: selector: run: mcw-httpd2 ports: - protocol: TCP port: 8080 targetPort: 80 [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ vim mcwRforSaleup.yml [machangwei@mcwk8s-master ~]$ cat mcwRforSaleup.yml apiVersion: apps/v1 kind: Deployment metadata: name: mcw-web spec: replicas: 3 selector: matchLabels: run: mcw-web template: metadata: labels: run: mcw-web spec: containers: - name: mcw-web image: httpd ports: - containerPort: 80 readinessProbe: httpGet: scheme: HTTP path: /healthy port: 8080 initialDelaySeconds: 10 periodSeconds: 5 --- apiVersion: v1 kind: Service metadata: name: web-svc spec: selector: run: mcw-web ports: - protocol: TCP port: 8080 targetPort: 80 [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwRforSaleup.yml deployment.apps/mcw-web created service/web-svc created [machangwei@mcwk8s-master ~]$ kubectl get pod -o wide|grep mcw-web mcw-web-76577f844c-lx9mm 0/1 Running 0 2m56s 10.244.2.24 mcwk8s-node2 <none> <none> mcw-web-76577f844c-pk9sd 0/1 ContainerCreating 0 2s <none> mcwk8s-node1 <none> <none> mcw-web-76577f844c-w554g 0/1 Running 0 11m 10.244.2.23 mcwk8s-node2 <none> <none> [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl describe pod mcw-web-76577f844c-lx9mm ....... Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 5m9s default-scheduler Successfully assigned default/mcw-web-76577f844c-lx9mm to mcwk8s-node2 Normal Pulling 5m8s kubelet Pulling image "httpd" Normal Pulled 4m52s kubelet Successfully pulled image "httpd" in 15.695204115s Normal Created 4m52s kubelet Created container mcw-web Normal Started 4m52s kubelet Started container mcw-web Warning Unhealthy 4s (x59 over 4m39s) kubelet Readiness probe failed: Get "http://10.244.2.24:8080/healthy": dial tcp 10.244.2.24:8080: connect: connection refused [machangwei@mcwk8s-master ~]$
Health Check 在滚动更新中的应用
1、滚动更新失败，不影响业务案例
apiVersion: v1 kind: Pod metadata: name: mcw-readiness labels: test: mcw-readiness spec: restartPolicy: OnFailure containers: - name: mcw-readiness image: busybox args: - /bin/sh - -c - touch /tmp/healthy; sleep 30; rm -rf /tmp/healthy; sleep 120 readinessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 periodSeconds: 5 [machangwei@mcwk8s-master ~]$ kubectl get deployment mcw-app NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 0/2 2 0 26s [machangwei@mcwk8s-master ~]$ vim mcwApp.v1.yml [machangwei@mcwk8s-master ~]$ cat mcwApp.v1.yml apiVersion: apps/v1 kind: Deployment metadata: name: mcw-app spec: replicas: 10 selector: matchLabels: app: mcw-app template: metadata: labels: app: mcw-app spec: containers: - name: mcw-app image: busybox args: - /bin/sh - -c - sleep 20;touch /tmp/healthy; sleep 30000 readinessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 periodSeconds: 5 [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwApp.v1.yml --record Flag --record has been deprecated, --record will be removed in the future deployment.apps/mcw-app configured [machangwei@mcwk8s-master ~]$ kubectl get deployment mcw-app NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 2/10 10 2 59s [machangwei@mcwk8s-master ~]$ kubectl get deployment mcw-app NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 2/10 10 2 67s [machangwei@mcwk8s-master ~]$ kubectl get deployment mcw-app NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 10/10 10 10 37m [machangwei@mcwk8s-master ~]$ kubectl get pod NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 0 37m mcw-app-5d455c9874-5r4xx 1/1 Running 0 38m mcw-app-5d455c9874-7gcnw 1/1 Running 0 37m mcw-app-5d455c9874-87lqk 1/1 Running 0 38m mcw-app-5d455c9874-ctzps 1/1 Running 0 37m mcw-app-5d455c9874-gwpz6 1/1 Running 0 37m mcw-app-5d455c9874-jhvvz 1/1 Running 0 37m mcw-app-5d455c9874-kmxqs 1/1 Running 0 37m mcw-app-5d455c9874-ksljt 1/1 Running 0 37m mcw-app-5d455c9874-skvv6 1/1 Running 0 37m [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ cp mcwApp.v1.yml mcwApp.v2.yml [machangwei@mcwk8s-master ~]$ vim mcwApp.v2.yml [machangwei@mcwk8s-master ~]$ cat mcwApp.v2.yml apiVersion: apps/v1 kind: Deployment metadata: name: mcw-app spec: replicas: 10 selector: matchLabels: app: mcw-app template: metadata: labels: app: mcw-app spec: containers: - name: mcw-app image: busybox args: - /bin/sh - -c - sleep 3000 readinessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 periodSeconds: 5 [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwApp.v2.yml --record Flag --record has been deprecated, --record will be removed in the future deployment.apps/mcw-app configured [machangwei@mcwk8s-master ~]$ kubectl get deployment mcw-app NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 8/10 5 8 39m [machangwei@mcwk8s-master ~]$ kubectl get deployment mcw-app NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 8/10 5 8 39m [machangwei@mcwk8s-master ~]$ kubectl get pod #5个pod是新副本，旧副本从10个减少到8个， NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 0 39m mcw-app-5d455c9874-5r4xx 1/1 Running 0 40m mcw-app-5d455c9874-7gcnw 1/1 Running 0 39m mcw-app-5d455c9874-87lqk 1/1 Running 0 40m mcw-app-5d455c9874-ctzps 1/1 Terminating 0 39m mcw-app-5d455c9874-gwpz6 1/1 Running 0 39m mcw-app-5d455c9874-jhvvz 1/1 Terminating 0 39m mcw-app-5d455c9874-kmxqs 1/1 Running 0 39m mcw-app-5d455c9874-ksljt 1/1 Running 0 39m mcw-app-5d455c9874-skvv6 1/1 Running 0 39m mcw-app-6fc84f4f96-bcqmr 0/1 Running 0 22s mcw-app-6fc84f4f96-bv2zj 0/1 Running 0 22s mcw-app-6fc84f4f96-pf6kd 0/1 ContainerCreating 0 22s mcw-app-6fc84f4f96-qsb8k 0/1 ContainerCreating 0 22s mcw-app-6fc84f4f96-xxzpq 0/1 ContainerCreating 0 22s [machangwei@mcwk8s-master ~]$ kubectl get deployment mcw-app NAME READY UP-TO-DATE AVAILABLE AGE #8个旧+5个新副本 完成更新的副本5，但是没有准备好，因为健康检查失败 mcw-app 8/10 5 8 40m [machangwei@mcwk8s-master ~]$ kubectl get deployment mcw-app NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 8/10 5 8 42m [machangwei@mcwk8s-master ~]$ kubectl get pod NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 0 42m mcw-app-5d455c9874-5r4xx 1/1 Running 0 43m mcw-app-5d455c9874-7gcnw 1/1 Running 0 42m mcw-app-5d455c9874-87lqk 1/1 Running 0 43m mcw-app-5d455c9874-gwpz6 1/1 Running 0 42m mcw-app-5d455c9874-kmxqs 1/1 Running 0 42m mcw-app-5d455c9874-ksljt 1/1 Running 0 42m mcw-app-5d455c9874-skvv6 1/1 Running 0 42m mcw-app-6fc84f4f96-bcqmr 0/1 Running 0 3m17s mcw-app-6fc84f4f96-bv2zj 0/1 Running 0 3m17s mcw-app-6fc84f4f96-pf6kd 0/1 Running 0 3m17s mcw-app-6fc84f4f96-qsb8k 0/1 Running 0 3m17s mcw-app-6fc84f4f96-xxzpq 0/1 Running 0 3m17s [machangwei@mcwk8s-master ~]$ kubectl get deployment mcw-app -o wide NAME READY UP-TO-DATE AVAILABLE AGE CONTAINERS IMAGES SELECTOR mcw-app 8/10 5 8 43m mcw-app busybox app=mcw-app [machangwei@mcwk8s-master ~]$
2、滚动升级失败回退
[machangwei@mcwk8s-master ~]$ kubectl get deployment NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 8/10 5 8 47h [machangwei@mcwk8s-master ~]$ kubectl describe deployment mcw-app Conditions: Type Status Reason ---- ------ ------ Available True MinimumReplicasAvailable Progressing False ProgressDeadlineExceeded OldReplicaSets: mcw-app-5d455c9874 (8/8 replicas created) NewReplicaSet: mcw-app-6fc84f4f96 (5/5 replicas created) Events: <none> [machangwei@mcwk8s-master ~]$ kubectl rollout history deployment mcw-app deployment.apps/mcw-app REVISION CHANGE-CAUSE 1 kubectl apply --filename=mcwApp.v1.yml --record=true 2 kubectl apply --filename=mcwApp.v2.yml --record=true [machangwei@mcwk8s-master ~]$ kubectl rollout undo deployment mcw-app --to-revision=1 #回退到版本1 deployment.apps/mcw-app rolled back [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl get deployment NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 8/10 10 8 47h [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl get deployment NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 8/10 10 8 47h [machangwei@mcwk8s-master ~]$ kubectl get deployment NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 10/10 10 10 47h [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl get pod NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 2 (42m ago) 47h mcw-app-5d455c9874-5r4xx 1/1 Running 2 (44m ago) 47h mcw-app-5d455c9874-7gcnw 1/1 Running 2 (42m ago) 47h mcw-app-5d455c9874-87lqk 1/1 Running 2 (44m ago) 47h mcw-app-5d455c9874-gwpz6 1/1 Running 2 (43m ago) 47h mcw-app-5d455c9874-kmxqs 1/1 Running 2 (43m ago) 47h mcw-app-5d455c9874-ksljt 1/1 Running 2 (42m ago) 47h mcw-app-5d455c9874-mzgds 1/1 Running 0 55s mcw-app-5d455c9874-skvv6 1/1 Running 2 (42m ago) 47h mcw-app-5d455c9874-z5qn2 1/1 Running 0 56s [machangwei@mcwk8s-master ~]$
3、滚动升级定制maxSurge和maxUnavailable
需要添加：
strategy: rollingUpdate: maxSurge: 35% maxUnavailable: 35%
maxSurge： 滚动更新中副本总数超过目标数的上限。上面案例是默认25%，所以副本总数最大值为10+10*25%=1+2.5=13 向上取整，所以滚动更新时最大副本数是13
maxUnavailable：不可用副本占目标数(可设置为整数)， 目标值10-roundDown（10*25%）=10-roundDown（2.5）=10-2=8 向下取整，所以滚动更新时最大不可用副本数是2
[machangwei@mcwk8s-master ~]$ cp mcwApp.v2.yml mcwApp.v3.yml [machangwei@mcwk8s-master ~]$ vim mcwApp.v3.yml [machangwei@mcwk8s-master ~]$ cat mcwApp.v3.yml apiVersion: apps/v1 kind: Deployment metadata: name: mcw-app spec: strategy: rollingUpdate: maxSurge: 35% maxUnavailable: 35% replicas: 10 selector: matchLabels: app: mcw-app template: metadata: labels: app: mcw-app spec: containers: - name: mcw-app image: busybox args: - /bin/sh - -c - sleep 3000 readinessProbe: exec: command: - cat - /tmp/healthy initialDelaySeconds: 10 periodSeconds: 5 [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl apply -f mcwApp.v3.yml deployment.apps/mcw-app configured [machangwei@mcwk8s-master ~]$ kubectl get deployment NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 7/10 7 7 47h [machangwei@mcwk8s-master ~]$ kubectl get pod NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 2 (57m ago) 47h mcw-app-5d455c9874-5r4xx 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-7gcnw 1/1 Running 2 (57m ago) 47h mcw-app-5d455c9874-87lqk 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-gwpz6 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-kmxqs 1/1 Running 2 (57m ago) 47h mcw-app-5d455c9874-ksljt 1/1 Running 2 (57m ago) 47h mcw-app-5d455c9874-mzgds 1/1 Terminating 0 15m mcw-app-5d455c9874-skvv6 1/1 Terminating 2 (57m ago) 47h mcw-app-5d455c9874-z5qn2 1/1 Terminating 0 15m mcw-app-6fc84f4f96-68qxf 0/1 ContainerCreating 0 19s mcw-app-6fc84f4f96-9fsgv 0/1 ContainerCreating 0 19s mcw-app-6fc84f4f96-fnvq9 0/1 ContainerCreating 0 19s mcw-app-6fc84f4f96-m9ts7 0/1 ContainerCreating 0 19s mcw-app-6fc84f4f96-pg5hb 0/1 ContainerCreating 0 19s mcw-app-6fc84f4f96-wsdq8 0/1 ContainerCreating 0 19s mcw-app-6fc84f4f96-z8rnm 0/1 ContainerCreating 0 19s [machangwei@mcwk8s-master ~]$ kubectl get deployment NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 7/10 7 7 47h [machangwei@mcwk8s-master ~]$ kubectl get pod NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-5r4xx 1/1 Running 2 (59m ago) 47h mcw-app-5d455c9874-7gcnw 1/1 Running 2 (57m ago) 47h mcw-app-5d455c9874-87lqk 1/1 Running 2 (59m ago) 47h mcw-app-5d455c9874-gwpz6 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-kmxqs 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-ksljt 1/1 Running 2 (57m ago) 47h mcw-app-6fc84f4f96-68qxf 0/1 ContainerCreating 0 44s mcw-app-6fc84f4f96-9fsgv 0/1 Running 0 44s mcw-app-6fc84f4f96-fnvq9 0/1 Running 0 44s mcw-app-6fc84f4f96-m9ts7 0/1 ContainerCreating 0 44s mcw-app-6fc84f4f96-pg5hb 0/1 ContainerCreating 0 44s mcw-app-6fc84f4f96-wsdq8 0/1 Running 0 44s mcw-app-6fc84f4f96-z8rnm 0/1 Running 0 44s [machangwei@mcwk8s-master ~]$ kubectl get pod NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-5r4xx 1/1 Running 2 (59m ago) 47h mcw-app-5d455c9874-7gcnw 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-87lqk 1/1 Running 2 (59m ago) 47h mcw-app-5d455c9874-gwpz6 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-kmxqs 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-ksljt 1/1 Running 2 (57m ago) 47h mcw-app-6fc84f4f96-68qxf 0/1 ContainerCreating 0 56s mcw-app-6fc84f4f96-9fsgv 0/1 Running 0 56s mcw-app-6fc84f4f96-fnvq9 0/1 Running 0 56s mcw-app-6fc84f4f96-m9ts7 0/1 ContainerCreating 0 56s mcw-app-6fc84f4f96-pg5hb 0/1 Running 0 56s mcw-app-6fc84f4f96-wsdq8 0/1 Running 0 56s mcw-app-6fc84f4f96-z8rnm 0/1 Running 0 56s [machangwei@mcwk8s-master ~]$ kubectl get pod NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-5r4xx 1/1 Running 2 (59m ago) 47h mcw-app-5d455c9874-7gcnw 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-87lqk 1/1 Running 2 (59m ago) 47h mcw-app-5d455c9874-gwpz6 1/1 Running 2 (59m ago) 47h mcw-app-5d455c9874-kmxqs 1/1 Running 2 (58m ago) 47h mcw-app-5d455c9874-ksljt 1/1 Running 2 (58m ago) 47h mcw-app-6fc84f4f96-68qxf 0/1 Running 0 72s mcw-app-6fc84f4f96-9fsgv 0/1 Running 0 72s mcw-app-6fc84f4f96-fnvq9 0/1 Running 0 72s mcw-app-6fc84f4f96-m9ts7 0/1 Running 0 72s mcw-app-6fc84f4f96-pg5hb 0/1 Running 0 72s mcw-app-6fc84f4f96-wsdq8 0/1 Running 0 72s mcw-app-6fc84f4f96-z8rnm 0/1 Running 0 72s [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl describe deployment mcw-app Name: mcw-app Namespace: default CreationTimestamp: Wed, 26 Jan 2022 00:51:50 +0800 Labels: <none> Annotations: deployment.kubernetes.io/revision: 4 Selector: app=mcw-app Replicas: 10 desired | 7 updated | 14 total | 7 available | 7 unavailable StrategyType: RollingUpdate MinReadySeconds: 0 RollingUpdateStrategy: 35% max unavailable, 35% max surge Pod Template: Labels: app=mcw-app Containers: mcw-app: Image: busybox Port: <none> Host Port: <none> Args: /bin/sh -c sleep 3000 Readiness: exec [cat /tmp/healthy] delay=10s timeout=1s period=5s #success=1 #failure=3 Environment: <none> Mounts: <none> Volumes: <none> Conditions: Type Status Reason ---- ------ ------ Available True MinimumReplicasAvailable Progressing True ReplicaSetUpdated OldReplicaSets: mcw-app-5d455c9874 (7/7 replicas created) NewReplicaSet: mcw-app-6fc84f4f96 (7/7 replicas created) Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal ScalingReplicaSet 17m deployment-controller Scaled down replica set mcw-app-6fc84f4f96 to 0 Normal ScalingReplicaSet 17m deployment-controller Scaled up replica set mcw-app-5d455c9874 to 10 Normal ScalingReplicaSet 98s deployment-controller Scaled up replica set mcw-app-6fc84f4f96 to 4 Normal ScalingReplicaSet 98s deployment-controller Scaled down replica set mcw-app-5d455c9874 to 7 Normal ScalingReplicaSet 98s deployment-controller Scaled up replica set mcw-app-6fc84f4f96 to 7 [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl rollout history deployment mcw-app deployment.apps/mcw-app REVISION CHANGE-CAUSE 3 kubectl apply --filename=mcwApp.v1.yml --record=true 4 kubectl apply --filename=mcwApp.v2.yml --record=true [machangwei@mcwk8s-master ~]$ mv mcwApp.v1.yml mcwApp.v1.ymlbak #将这个文件移走，照样可以回退，不受影响，但是v3部署没有设置记录参数，就没有记录下来 [machangwei@mcwk8s-master ~]$ kubectl rollout undo deployment mcw-app --to-revision=3 #回退 deployment.apps/mcw-app rolled back [machangwei@mcwk8s-master ~]$ kubectl get deployment #修改两个最大的参数，两个数量都已经发生了改变，现在已经执行了回退了 NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 7/10 10 7 47h [machangwei@mcwk8s-master ~]$ kubectl get pod NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 2 (61m ago) 47h mcw-app-5d455c9874-5r4xx 1/1 Running 2 (62m ago) 47h mcw-app-5d455c9874-78hxb 0/1 ContainerCreating 0 25s mcw-app-5d455c9874-7gcnw 1/1 Running 2 (61m ago) 47h mcw-app-5d455c9874-87lqk 1/1 Running 2 (62m ago) 47h mcw-app-5d455c9874-dtk7n 0/1 Running 0 25s mcw-app-5d455c9874-gwpz6 1/1 Running 2 (61m ago) 47h mcw-app-5d455c9874-kmxqs 1/1 Running 2 (61m ago) 47h mcw-app-5d455c9874-ksljt 1/1 Running 2 (61m ago) 47h mcw-app-5d455c9874-npsbq 0/1 Running 0 25s mcw-app-6fc84f4f96-68qxf 0/1 Terminating 0 4m3s mcw-app-6fc84f4f96-9fsgv 0/1 Terminating 0 4m3s mcw-app-6fc84f4f96-fnvq9 0/1 Terminating 0 4m3s mcw-app-6fc84f4f96-m9ts7 0/1 Terminating 0 4m3s mcw-app-6fc84f4f96-pg5hb 0/1 Terminating 0 4m3s mcw-app-6fc84f4f96-wsdq8 0/1 Terminating 0 4m3s mcw-app-6fc84f4f96-z8rnm 0/1 Terminating 0 4m3s [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ [machangwei@mcwk8s-master ~]$ kubectl get deployment NAME READY UP-TO-DATE AVAILABLE AGE mcw-app 10/10 10 10 47h [machangwei@mcwk8s-master ~]$ kubectl get pod NAME READY STATUS RESTARTS AGE mcw-app-5d455c9874-2g7zl 1/1 Running 2 (62m ago) 47h mcw-app-5d455c9874-5r4xx 1/1 Running 2 (63m ago) 47h mcw-app-5d455c9874-78hxb 1/1 Running 0 62s mcw-app-5d455c9874-7gcnw 1/1 Running 2 (61m ago) 47h mcw-app-5d455c9874-87lqk 1/1 Running 2 (63m ago) 47h mcw-app-5d455c9874-dtk7n 1/1 Running 0 62s mcw-app-5d455c9874-gwpz6 1/1 Running 2 (62m ago) 47h mcw-app-5d455c9874-kmxqs 1/1 Running 2 (62m ago) 47h mcw-app-5d455c9874-ksljt 1/1 Running 2 (61m ago) 47h mcw-app-5d455c9874-npsbq 1/1 Running 0 62s [machangwei@mcwk8s-master ~]$
