八 pod：pod的调度——将 Pod 指派给节点( 六 ) _生活百科

3.5.1 使用硬策略requiredDuringSchedulingIgnoredDuringExecution创建pod，requiredDuringSchedulingIgnoredDuringExecution参数表示：节点必须包含一个键名为 kubernetes.io/hostname 的标签，并且该标签的取值必须为 k8scloude2 或 k8scloude3 。
你可以使用 operator 字段来为 Kubernetes 设置在解释规则时要使用的逻辑操作符。你可以使用 In、NotIn、Exists、DoesNotExist、Gt 和 Lt 之一作为操作符。NotIn 和 DoesNotExist 可用来实现节点反亲和性行为。你也可以使用节点污点将 Pod 从特定节点上驱逐。
注意：

如果你同时指定了 nodeSelector 和 nodeAffinity，两者必须都要满足，才能将 Pod 调度到候选节点上。
如果你指定了多个与 nodeAffinity 类型关联的 nodeSelectorTerms，只要其中一个 nodeSelectorTerms 满足的话，Pod 就可以被调度到节点上。
如果你指定了多个与同一 nodeSelectorTerms 关联的 matchExpressions，则只有当所有 matchExpressions 都满足时 Pod 才可以被调度到节点上。

[root@k8scloude1 pod]# vim requiredDuringSchedule.yaml#硬策略[root@k8scloude1 pod]# cat requiredDuringSchedule.yamlapiVersion: v1kind: Podmetadata:creationTimestamp: nulllabels:run: pod1name: pod1namespace: podspec:affinity:nodeAffinity:requiredDuringSchedulingIgnoredDuringExecution:nodeSelectorTerms:- matchExpressions:- key: kubernetes.io/hostnameoperator: Invalues:- k8scloude2- k8scloude3containers:- image: nginximagePullPolicy: IfNotPresentname: pod1resources: {}ports:- name: httpcontainerPort: 80protocol: TCPhostPort: 80dnsPolicy: ClusterFirstrestartPolicy: Alwaysstatus: {}[root@k8scloude1 pod]# kubectl apply -f requiredDuringSchedule.yamlpod/pod1 created

可以看到pod运行在k8scloude3节点

[root@k8scloude1 pod]# kubectl get podsNAMEREADYSTATUSRESTARTSAGEpod11/1Running06s[root@k8scloude1 pod]# kubectl get pods -o wideNAMEREADYSTATUSRESTARTSAGEIPNODENOMINATED NODEREADINESS GATESpod11/1Running010s10.244.251.212k8scloude3<none><none>[root@k8scloude1 pod]# kubectl delete pod pod1 --forcewarning: 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 "pod1" force deleted

创建pod，requiredDuringSchedulingIgnoredDuringExecution参数表示：节点必须包含一个键名为 kubernetes.io/hostname 的标签，并且该标签的取值必须为 k8scloude4 或 k8scloude5 。

[root@k8scloude1 pod]# vim requiredDuringSchedule1.yaml [root@k8scloude1 pod]# cat requiredDuringSchedule1.yamlapiVersion: v1kind: Podmetadata:creationTimestamp: nulllabels:run: pod1name: pod1namespace: podspec:affinity:nodeAffinity:requiredDuringSchedulingIgnoredDuringExecution:nodeSelectorTerms:- matchExpressions:- key: kubernetes.io/hostnameoperator: Invalues:- k8scloude4- k8scloude5containers:- image: nginximagePullPolicy: IfNotPresentname: pod1resources: {}ports:- name: httpcontainerPort: 80protocol: TCPhostPort: 80dnsPolicy: ClusterFirstrestartPolicy: Alwaysstatus: {}[root@k8scloude1 pod]# kubectl apply -f requiredDuringSchedule1.yamlpod/pod1 created

由于requiredDuringSchedulingIgnoredDuringExecution是硬策略，k8scloude4 ， k8scloude5不满足条件，所以pod创建失败

[root@k8scloude1 pod]# kubectl get pods -o wideNAMEREADYSTATUSRESTARTSAGEIPNODENOMINATED NODEREADINESS GATESpod10/1Pending07s<none><none><none><none>[root@k8scloude1 pod]# kubectl delete pod pod1 --forcewarning: 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 "pod1" force deleted

3.5.2 使用软策略preferredDuringSchedulingIgnoredDuringExecution给节点打标签

[root@k8scloude1 pod]# kubectl label nodes k8scloude2 xx=72node/k8scloude2 labeled[root@k8scloude1 pod]# kubectl label nodes k8scloude3 xx=59node/k8scloude3 labeled

创建pod，preferredDuringSchedulingIgnoredDuringExecution参数表示：节点最好具有一个键名为 xx 且取值大于 60 的标签。

[root@k8scloude1 pod]# vim preferredDuringSchedule.yaml [root@k8scloude1 pod]# cat preferredDuringSchedule.yamlapiVersion: v1kind: Podmetadata:creationTimestamp: nulllabels:run: pod1name: pod1namespace: podspec:affinity:nodeAffinity:preferredDuringSchedulingIgnoredDuringExecution:- weight: 2preference:matchExpressions:- key: xxoperator: Gtvalues:- "60"containers:- image: nginximagePullPolicy: IfNotPresentname: pod1resources: {}ports:- name: httpcontainerPort: 80protocol: TCPhostPort: 80dnsPolicy: ClusterFirstrestartPolicy: Alwaysstatus: {}[root@k8scloude1 pod]# kubectl apply -f preferredDuringSchedule.yamlpod/pod1 created

可以看到pod运行在k8scloude2，因为k8scloude2标签为 xx=72，72大于60