【Kubernetes】创建Pod并分配到指定节点
一、编辑yaml文件###
[root@K8s-Master Tools]# cat hello-world-pod.yaml
apiVersion: v1
kind: Pod
metadata:
name: hello-world
spec:
nodeSelector:
type: node1
containers:
- name: hello
image: "ubuntu:14.04"
command: [ "/bin/bash", "-ce", "tail -f /dev/null" ]
二、创建并查看运行状态###
[root@K8s-Master Tools]# kubectl create -f hello-world-pod.yaml
pod/hello-world created
[root@K8s-Master Tools]# kubectl get pod hello-world
NAME READY STATUS RESTARTS AGE
hello-world 1/1 Running 0 8m2s
[root@K8s-Master Tools]# kubectl get pods --all-namespaces
NAMESPACE NAME READY STATUS RESTARTS AGE
default hello-world 1/1 Running 0 3s
kube-system coredns-bccdc95cf-plst5 1/1 Running 13 36d
kube-system coredns-bccdc95cf-vkqzr 1/1 Running 13 36d
kube-system etcd-k8s-master 1/1 Running 15 36d
kube-system kube-apiserver-k8s-master 1/1 Running 15 36d
kube-system kube-controller-manager-k8s-master 1/1 Running 15 36d
kube-system kube-flannel-ds-amd64-4g6qc 1/1 Running 6 36d
kube-system kube-flannel-ds-amd64-ccj8v 1/1 Running 14 36d
kube-system kube-flannel-ds-amd64-fhrfj 1/1 Running 8 36d
kube-system kube-proxy-jp6mp 1/1 Running 13 36d
kube-system kube-proxy-rkbcx 1/1 Running 6 36d
kube-system kube-proxy-rmkqm 1/1 Running 8 36d
kube-system kube-scheduler-k8s-master 1/1 Running 15 36d
kube-system kubernetes-dashboard-6db4897b74-lqfl6 1/1 Running 1 24h
注意点：将pod分配到指定的节点
将pod调度到指定节点有2种方式：
方式一：强制约束Pod调度到指定Node节点上
Pod.spec.nodeName
方式二：通过label-selector机制选择节点
Pod.spec.nodeSelector
以下使用第二种方式实现pod调度到指定节点。
①为Node添加label
kubectl label node k8s-node1 type=node1
kubectl get nodes --show-labels
[root@K8s-Master Tools]# kubectl get nodes --show-labels
NAME STATUS ROLES AGE VERSION LABELS
k8s-master Ready master 37d v1.15.3 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=k8s-master,kubernetes.io/os=linux,node-role.kubernetes.io/master=,type=master
k8s-node1 Ready <none> 37d v1.15.3 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=k8s-node1,kubernetes.io/os=linux,type=node1
k8s-node2 Ready <none> 37d v1.15.3 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=k8s-node2,kubernetes.io/os=linux,type=node2
②将pod分配到带有指定label的node
Pod操作：
删除Pod
kubectl delete pod hello
更新Pod
kubctl replace hello-world.yaml
三、可能遇到的问题###
Error：Back-off restarting failed container
[root@K8s-Master Tools]# kubectl describe pod hello-world
Name: hello-world
Namespace: default
Priority: 0
Node: k8s-node1/10.0.2.15
Start Time: Fri, 18 Oct 2019 17:16:13 +1100
Labels: <none>
Annotations: <none>
Status: Running
IP: 192.168.1.15
Containers:
hello:
Container ID: docker://ae85d78070dc02a19fed619c6270732cd98bfc1d824262b2935bc2362d7b70e5
Image: ubuntu:14.04
Image ID: docker-pullable://ubuntu@sha256:2f7c79927b346e436cc14c92bd4e5bd778c3bd7037f35bc639ac1589a7acfa90
Port: <none>
Host Port: <none>
Command:
/bin/echo
Hello
world
State: Waiting
Reason: CrashLoopBackOff
Last State: Terminated
Reason: Completed
Exit Code: 0
Started: Fri, 18 Oct 2019 17:18:10 +1100
Finished: Fri, 18 Oct 2019 17:18:10 +1100
Ready: False
Restart Count: 4
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from default-token-hctt8 (ro)
Conditions:
Type Status
Initialized True
Ready False
ContainersReady False
PodScheduled True
Volumes:
default-token-hctt8:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-hctt8
Optional: false
QoS Class: BestEffort
Node-Selectors: type=node1
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 2m31s default-scheduler Successfully assigned default/hello-world to k8s-node1
Normal Pulling 2m30s kubelet, k8s-node1 Pulling image "ubuntu:14.04"
Normal Pulled 2m10s kubelet, k8s-node1 Successfully pulled image "ubuntu:14.04"
Normal Created 34s (x5 over 2m10s) kubelet, k8s-node1 Created container hello
Normal Started 34s (x5 over 2m9s) kubelet, k8s-node1 Started container hello
Normal Pulled 34s (x4 over 2m8s) kubelet, k8s-node1 Container image "ubuntu:14.04" already present on machine
Warning BackOff 34s (x9 over 2m7s) kubelet, k8s-node1 Back-off restarting failed container
解决方法：
As per Describe Pod command listing, your Container inside the Pod has been already completed with exit code 0, which states about successful completion without any errors/problems, but the life cycle for the Pod was very short. To keep Pod running continuously you must specify a task that will never finish.
apiVersion: v1
kind: Pod
metadata:
name: ubuntu
spec:
containers:
- name: ubuntu
image: ubuntu
command: [ "/bin/bash", "-ce", "tail -f /dev/null" ]
