k8s污点Taints与容忍详解Toleration

污点状态：

• NoSchedule：如果 Node 上带有污点 effect 为 NoSchedule，而 Node 上不带相应容忍，Kubernetes 就不会调度 Pod 到这台 Node 上。

• PreferNoShedule：如果 Node 上带有污点 effect 为 PreferNoShedule，这时候 Kubernetes 会努力不要调度这个 Pod 到这个 Node 上。

• NoExecute：如果 Node 上带有污点 effect 为 NoExecute，这个已经在 Node 上运行的 Pod 会从 Node 上驱逐掉。没有运行在 Node 的 Pod 不能被调度到这个 Node 上。

污点值：

• 污点 value 的值可以为 NoSchedule、PreferNoSchedule 或 NoExecute

污点属性：

• 污点是k8s集群的pod中的一种属性

• 污点属性分为以上三种

污点组成：

• key、value 及一个 effect 三个元素

<key>=<value>:<effect>

1、设置单污点及单容忍度

kubectl taint nodes master1 node-role.kubernetes.io/master=:NoSchedule     

kubectl taint node node1 key1=value1:NoSchedule       # 设置value值

kubectl taint node master1 key2=:PreferNoSchedule     # 不设置value值

2、设置多污点及多容忍度

kubectl taint nodes node1 key1=value1:NoSchedule  

kubectl taint nodes node1 key1=value1:NoExecute  

kubectl taint nodes node1 key2=value2:NoSchedule

3、查看pod中的污点状态

[root@master1 ~]# kubectl describe nodes master1
Name:               master1
Roles:              master
Labels:             beta.kubernetes.io/arch=amd64
                    beta.kubernetes.io/os=linux
                    kubernetes.io/arch=amd64
                    kubernetes.io/hostname=master1
                    kubernetes.io/os=linux
                    node-role.kubernetes.io/master=
Annotations:        flannel.alpha.coreos.com/backend-data: {"VNI":1,"VtepMAC":"36:51:e1:31:e5:9e"}
                    flannel.alpha.coreos.com/backend-type: vxlan
                    flannel.alpha.coreos.com/kube-subnet-manager: true
                    flannel.alpha.coreos.com/public-ip: 192.168.200.3
                    kubeadm.alpha.kubernetes.io/cri-socket: /var/run/dockershim.sock
                    node.alpha.kubernetes.io/ttl: 0
                    volumes.kubernetes.io/controller-managed-attach-detach: true
CreationTimestamp:  Wed, 13 Jan 2021 06:04:10 -0500
Taints:             node-role.kubernetes.io/master:NoSchedule              # 污点状态及容忍度
Unschedulable:      false
Lease:
  HolderIdentity:  master1
  AcquireTime:     <unset>
  RenewTime:       Thu, 14 Jan 2021 01:14:07 -0500
Conditions:
  Type                 Status  LastHeartbeatTime                 LastTransitionTime                Reason                       Message
  ----                 ------  -----------------                 ------------------                ------                       -------
  NetworkUnavailable   False   Wed, 13 Jan 2021 06:12:43 -0500   Wed, 13 Jan 2021 06:12:43 -0500   FlannelIsUp                  Flannel is running on this node
  MemoryPressure       False   Thu, 14 Jan 2021 01:11:17 -0500   Wed, 13 Jan 2021 06:50:32 -0500   KubeletHasSufficientMemory   kubelet has sufficient memory available
  DiskPressure         False   Thu, 14 Jan 2021 01:11:17 -0500   Wed, 13 Jan 2021 06:50:32 -0500   KubeletHasNoDiskPressure     kubelet has no disk pressure
  PIDPressure          False   Thu, 14 Jan 2021 01:11:17 -0500   Wed, 13 Jan 2021 06:50:32 -0500   KubeletHasSufficientPID      kubelet has sufficient PID available
  Ready                True    Thu, 14 Jan 2021 01:11:17 -0500   Wed, 13 Jan 2021 06:50:32 -0500   KubeletReady                 kubelet is posting ready status
Addresses:
  InternalIP:  192.168.200.3
  Hostname:    master1
Capacity:
  cpu:                4
  ephemeral-storage:  17394Mi
  hugepages-1Gi:      0
  hugepages-2Mi:      0
  memory:             2897500Ki
  pods:               110
Allocatable:
  cpu:                4
  ephemeral-storage:  16415037823
  hugepages-1Gi:      0
  hugepages-2Mi:      0
  memory:             2795100Ki
  pods:               110
System Info:
  Machine ID:                 feb4edfea2404d3c8ad028ca4593bb32
  System UUID:                C6F44D56-0F24-6114-23E7-8DF6CD4E4CFE
  Boot ID:                    afcc0ef6-d767-4b97-9a7b-9b2500757f2e
  Kernel Version:             3.10.0-862.el7.x86_64
  OS Image:                   CentOS Linux 7 (Core)
  Operating System:           linux
  Architecture:               amd64
  Container Runtime Version:  docker://19.3.0
  Kubelet Version:            v1.18.2
  Kube-Proxy Version:         v1.18.2
PodCIDR:                      10.244.0.0/24
PodCIDRs:                     10.244.0.0/24
Non-terminated Pods:          (6 in total)
  Namespace                   Name                               CPU Requests  CPU Limits  Memory Requests  Memory Limits  AGE
  ---------                   ----                               ------------  ----------  ---------------  -------------  ---
  kube-system                 etcd-master1                       0 (0%)        0 (0%)      0 (0%)           0 (0%)         19h
  kube-system                 kube-apiserver-master1             250m (6%)     0 (0%)      0 (0%)           0 (0%)         19h
  kube-system                 kube-controller-manager-master1    200m (5%)     0 (0%)      0 (0%)           0 (0%)         19h
  kube-system                 kube-flannel-ds-wzf7w              100m (2%)     100m (2%)   50Mi (1%)        50Mi (1%)      19h
  kube-system                 kube-proxy-7h5sb                   0 (0%)        0 (0%)      0 (0%)           0 (0%)         19h
  kube-system                 kube-scheduler-master1             100m (2%)     0 (0%)      0 (0%)           0 (0%)         19h
Allocated resources:
  (Total limits may be over 100 percent, i.e., overcommitted.)
  Resource           Requests    Limits
  --------           --------    ------
  cpu                650m (16%)  100m (2%)
  memory             50Mi (1%)   50Mi (1%)
  ephemeral-storage  0 (0%)      0 (0%)
  hugepages-1Gi      0 (0%)      0 (0%)
  hugepages-2Mi      0 (0%)      0 (0%)
Events:              <none>

4、过滤出有几台节点存在污和容忍度是什么

[root@master1 ~]# kubectl describe node master1 | grep Taints
Taints:             node-role.kubernetes.io/master:NoSchedule

[root@master1 ~]# kubectl describe node master2 | grep Taints
Taints:             node-role.kubernetes.io/master:NoSchedule

[root@master1 ~]# kubectl describe node master3 | grep Taints
Taints:             node-role.kubernetes.io/master:NoSchedule

5、有无污点返回的结果

Taints:             node-role.kubernetes.io/master:NoSchedule     # 有污点

Taints:             <none>                                        # 没污点

6、删除污点使其pod能够调度和使用

kubectl taint node master1 node-role.kubernetes.io/master:NoSchedule-

kubectl taint nodes master1 key:NoSchedule-