文章目录
- 探针的使用
- 容器探针启动实验1-启动探针的使用-startupprobe
- Liveness Probes 和 Readiness Probes
-
探针的使用
kubectl edit deploy -n kube-system corednslivenessprobe 的使用 livenessProbe:failureThreshold: 5httpGet:path: /healthport: 8080scheme: HTTPinitialDelaySeconds: 60periodSeconds: 10successThreshold: 1timeoutSeconds: 5readinessProbereadinessProbe:failureThreshold: 3httpGet:path: /readyport: 8181scheme: HTTPperiodSeconds: 10successThreshold: 1timeoutSeconds: 1kubectl edit po nginx kubectl describe po nginx-daemon
最下面有容器启动时候相关日志
容器探针启动实验1-启动探针的使用-startupprobe
Events:Type Reason Age From Message---- ------ ---- ---- -------Normal Scheduled 20s default-scheduler Successfully assigned default/my-pod1 to ha2.example.localNormal Pulled 20s kubelet Container image "nginx:1.7.9" already present on machineNormal Created 20s kubelet Created container nginxNormal Started 20s kubelet Started container nginxWarning Unhealthy 4s (x2 over 14s) kubelet Startup probe failed: HTTP probe failed with statuscode: 404
[root@kubeadm-master1 test]# kubectl get pod
NAME READY STATUS RESTARTS AGE
client 1/1 Running 0 48d
my-pod 1/1 Running 0 7d23h
my-pod1 0/1 Running 0 37s
net-test1 1/1 Running 133 55d
net-test2 1/1 Running 13 55d
nginx-deployment-67dfd6c8f9-5s6nz 1/1 Running 1 55d
tomcat-deployment-6c44f58b47-4pz6d 1/1 Running 1 55d
[root@kubeadm-master1 test]# kubectl get pod
NAME READY STATUS RESTARTS AGE
client 1/1 Running 0 48d
my-pod 1/1 Running 0 7d23h
my-pod1 0/1 Running 0 42s
net-test1 1/1 Running 133 55d
net-test2 1/1 Running 13 55d
nginx-deployment-67dfd6c8f9-5s6nz 1/1 Running 1 55d
tomcat-deployment-6c44f58b47-4pz6d 1/1 Running 1 55d
[root@kubeadm-master1 test]# cat nginx-po.yaml
apiVersion: v1
kind: Pod
metadata:name: my-pod1labels:type: apptest: "1.0.0"namespace: default
spec:containers:- name: nginximage: nginx:1.7.9imagePullPolicy: IfNotPresentcommand:- nginx- -g- 'daemon off;'workingDir: /usr/share/nginx/htmlports:- name: httpcontainerPort: 80protocol: TCPenv:- name: JVM_OPTSvalue: '-Xms128m -Xmx128m'resources:requests:cpu: 100mmemory: 128Milimits:cpu: 200mmemory: 256MistartupProbe:httpGet:path: /api/pathport: 80failureThreshold: 3periodSeconds: 10successThreshold: 1timeoutSeconds: 5restartPolicy: OnFailure
Liveness Probes 和 Readiness Probes
用于检查容器是否还在运行。如果 liveness 探针失败，Kubernetes 将杀死容器，并根据其重启策略来处理。
用于检查容器是否已经准备好接收流量。如果 readiness 探针失败，Kubernetes 将不会将流量路由到该容器。定义了一个 startupProbe，它在容器启动后通过 HTTP GET 请求检查 /api/path 端点。现在我们将添加 livenessProbe 和 readinessProbe。一个 livenessProbe 可以如下定义：livenessProbe:httpGet:path: /api/healthport: 80initialDelaySeconds: 30periodSeconds: 10
这个 livenessProbe 会在容器启动后的30秒开始工作，每10秒检查一次 /api/health 端点。一个 readinessProbe 可以如下定义：readinessProbe:httpGet:path: /api/readyport: 80initialDelaySeconds: 5periodSeconds: 5
这个 readinessProbe 会在容器启动后的5秒开始工作，每5秒检查一次 /api/ready 端点。
演示
[root@kubeadm-master1 test]# cat liveness.yml
apiVersion: v1
kind: Pod
metadata:name: my-pod1labels:type: apptest: "1.0.0"namespace: default
spec:containers:- name: nginximage: nginx:1.7.9imagePullPolicy: IfNotPresentcommand:- nginx- -g- 'daemon off;'workingDir: /usr/share/nginx/htmlports:- name: httpcontainerPort: 80protocol: TCPenv:- name: JVM_OPTSvalue: '-Xms128m -Xmx128m'resources:requests:cpu: 100mmemory: 128Milimits:cpu: 200mmemory: 256MistartupProbe:httpGet:path: /api/pathport: 80failureThreshold: 3periodSeconds: 10successThreshold: 1timeoutSeconds: 5livenessProbe:httpGet:path: /api/healthport: 80initialDelaySeconds: 30periodSeconds: 10readinessProbe:httpGet:path: /api/readyport: 80initialDelaySeconds: 5periodSeconds: 5restartPolicy: OnFailureState: RunningStarted: Thu, 15 Feb 2024 15:15:19 +0800Ready: FalseRestart Count: 0Limits:cpu: 200mmemory: 256MiRequests:cpu: 100mmemory: 128MiLiveness: http-get http://:80/api/health delay=30s timeout=1s period=10s #success=1 #failure=3Readiness: http-get http://:80/api/ready delay=5s timeout=1s period=5s #success=1 #failure=3Startup: http-get http://:80/api/path delay=0s timeout=5s period=10s #success=1 #failure=3Environment:JVM_OPTS: -Xms128m -Xmx128mMounts:/var/run/secrets/kubernetes.io/serviceaccount from default-token-75cq9 (ro)
Conditions:Type StatusInitialized TrueReady FalseContainersReady FalsePodScheduled True
Volumes:default-token-75cq9:Type: Secret (a volume populated by a Secret)SecretName: default-token-75cq9Optional: false
QoS Class: Burstable
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute op=Exists for 300snode.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events:Type Reason Age From Message---- ------ ---- ---- -------Normal Scheduled 19s default-scheduler Successfully assigned default/my-pod1 to ha2.example.localNormal Pulled 19s kubelet Container image "nginx:1.7.9" already present on machineNormal Created 19s kubelet Created container nginxNormal Started 19s kubelet Started container nginxWarning Unhealthy 2s (x2 over 12s) kubelet Startup probe failed: HTTP probe failed with statuscode: 404
若存在started.html 则进行
