如何管理Pod?
daemonset的控制器会监听kuberntes的daemonset对象、pod对象、node对象,
这些被监听的对象之变动,就会触发syncLoop循环让kubernetes集群朝着daemonset对象描述的状态进行演进。
Deployment 部署的副本 Pod 会分布在各个 Node 上,每个 Node 都可能运行好几个副本。
DaemonSet 的不同之处在于:每个 Node 上最多只能运行一个副本。
[root@master ~]# kubectl explain ds
KIND: DaemonSet
VERSION: apps/v1
DESCRIPTION:
DaemonSet represents the configuration of a daemon set.
FIELDS:
apiVersion <string> # 当前资源使用的api版本,跟VERSION: apps/v1保持一致
kind <string> # 资源类型,跟KIND: DaemonSet保持一致
metadata <Object> # 元数据,定义DaemonSet名字的
spec <Object> # 定义容器的
status <Object> # 状态信息,不能改
[root@master ~]# kubectl explain ds.spec
KIND: DaemonSet
VERSION: apps/v1
RESOURCE: spec <Object>
DESCRIPTION:
The desired behavior of this daemon set. More info:
https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#spec-and-status
DaemonSetSpec is the specification of a daemon set.
FIELDS:
minReadySeconds <integer> # 当新的pod启动几秒种后,再kill掉旧的pod。
revisionHistoryLimit <integer> # 历史版本
selector <Object> -required- # 用于匹配pod的标签选择器
template <Object> -required- # 定义Pod的模板,基于这个模板定义的所有pod是一样的
updateStrategy <Object> # daemonset的升级策略
对于template而言,其内部定义的就是pod,pod模板是一个独立的对象
[root@master ~]# kubectl explain ds.spec.template
KIND: DaemonSet
VERSION: apps/v1
RESOURCE: template <Object>
DESCRIPTION:
......
FIELDS:
metadata <Object>
spec <Object>
把fluentd-2-5-1.tar.gz上传到工作节点node01、node02上,解压.
特别注意: 因为我们安装的k8s版本是1.25,那就需要按照文档步骤ctr -n=k8s.io images import导出镜像,如果k8s版本是1.24之前的,可以用docker load -i解压,视频里用的docker load -i,现在我们课程安装更新到了k8s1.25,所以导出镜像需要用ctr -n=k8s.io images import
[root@master 13]# scp fluentd_2_5_1.tar.gz root@10.32.1.147:/root/package
fluentd_2_5_1.tar.gz 100% 142MB 28.0MB/s 00:05
[root@master 13]# scp fluentd_2_5_1.tar.gz root@10.32.1.148:/root/package
fluentd_2_5_1.tar.gz 100% 142MB 52.1MB/s 00:02
[root@master 13]#
[root@node01 package]# ctr -n=k8s.io images import fluentd_2_5_1.tar.gz
unpacking docker.io/xianchao/fluentd:v2.5.1 (sha256:3baf02875b273fe6f675710d0f25ea9374b94d6704c334c75e854f8465a777a9)...done
[root@node02 package]# ctr -n=k8s.io images import fluentd_2_5_1.tar.gz
unpacking docker.io/xianchao/fluentd:v2.5.1 (sha256:3baf02875b273fe6f675710d0f25ea9374b94d6704c334c75e854f8465a777a9)...done
[root@master 13]# cat daemonset.yaml
apiVersion: apps/v1
kind: DaemonSet
metadata:
labels:
k8s-app: fluentd-logging
name: fluentd-elasticsearch
namespace: kube-system
spec:
selector:
matchLabels:
name: fluentd-elasticsearch
template:
metadata:
name: fluentd
labels:
name: fluentd-elasticsearch
spec:
tolerations:
- key: node-role.kubernetes.io/master
effect: NoSchedule
containers:
- name: fluentd-elasticsearch
image: xianchao/fluentd:v2.5.1
resources:
limits:
memory: 500Mi
requests:
cpu: 100m
memory: 200Mi
volumeMounts:
- name: varlog
mountPath: /var/log
- name: varlibdockercontainers
mountPath: /var/lib/docker/containers
readOnly: true
terminationGracePeriodSeconds: 30
volumes:
- name: varlog
hostPath:
path: /var/log
- name: varlibdockercontainers
hostPath:
path: /var/lib/docker/containers
[root@master 13]# kubectl apply -f daemonset.yaml
daemonset.apps/fluentd-elasticsearch created
[root@master 13]# kubectl get ds -n kube-system
NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
calico-node 3 3 2 3 2 kubernetes.io/os=linux 24d
fluentd-elasticsearch 2 2 2 2 2 <none> 30s
kube-proxy 3 3 3 3 3 kubernetes.io/os=linux 24d
[root@master 13]# kubectl get pods -n kube-system -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
fluentd-elasticsearch-27xqk 1/1 Running 0 100s 10.244.140.113 node02 <none> <none>
fluentd-elasticsearch-zzzlr 1/1 Running 0 100s 10.244.196.163 node01 <none> <none>
通过上面可以看到在k8s的三个节点均创建了fluentd这个pod。pod的名字是由控制器的名字-随机数组成的。
[root@master 13]# cat daemonset.yaml
apiVersion: apps/v1 # DaemonSet使用的api版本
kind: DaemonSet # 资源类型
metadata:
labels:
k8s-app: fluentd-logging # 资源具有的标签
name: fluentd-elasticsearch # 资源的名字
namespace: kube-system # 资源所在的名称空间
spec:
selector: # 标签选择器
matchLabels:
name: fluentd-elasticsearch
template:
metadata:
name: fluentd
labels: # 基于这回模板定义的pod具有的标签
name: fluentd-elasticsearch
spec:
tolerations: # 定义容忍度
- key: node-role.kubernetes.io/master
effect: NoSchedule
containers: # 定义容器
- name: fluentd-elasticsearch
image: xianchao/fluentd:v2.5.1
resources: # 资源配额
limits:
memory: 500Mi
requests:
cpu: 100m
memory: 200Mi
volumeMounts:
- name: varlog
mountPath: /var/log # 将varlog卷挂载到容器中指定挂载点/var/log
- name: varlibdockercontainers
mountPath: /var/lib/docker/containers # 将varlibdockercontainers卷容器中指定挂载点/var/lib/docker/containers
readOnly: true # 挂载目录是只读权限
terminationGracePeriodSeconds: 30 # 优雅的关闭服务
volumes:
- name: varlog
hostPath:
path: /var/log # 基于本地目录创建一个卷
- name: varlibdockercontainers
hostPath:
path: /var/lib/docker/containers # 基于本地目录创建一个卷
[root@master 13]# kubectl explain ds.spec.updateStrategy
KIND: DaemonSet
VERSION: apps/v1
RESOURCE: updateStrategy <Object>
DESCRIPTION:
An update strategy to replace existing DaemonSet pods with new pods.
DaemonSetUpdateStrategy is a struct used to control the update strategy for
a DaemonSet.
FIELDS:
rollingUpdate <Object>
Rolling update config params. Present only if type = "RollingUpdate".
type <string>
Type of daemon set update. Can be "RollingUpdate" or "OnDelete". Default is
RollingUpdate.
Possible enum values:
- `"OnDelete"` Replace the old daemons only when it's killed
- `"RollingUpdate"` Replace the old daemons by new ones using rolling
update i.e replace them on each node one after the other.
[root@master 13]# kubectl explain ds.spec.updateStrategy.rollingUpdate
KIND: DaemonSet
VERSION: apps/v1
RESOURCE: rollingUpdate <Object>
DESCRIPTION:
Rolling update config params. Present only if type = "RollingUpdate".
Spec to control the desired behavior of daemon set rolling update.
FIELDS:
maxSurge <string>
maxUnavailable <string>
# rollingUpdate更新策略只支持maxUnavailabe,先删除再更新
# 因为我们不支持一个节点运行两个pod,因此需要先删除一个,再更新一个。
kubectl set image daemonsets fluentd-elasticsearch fluentd-elasticsearch=ikubernetes/filebeat:5.6.6-alpine -n kube-system
这个镜像启动pod会有问题,主要是演示daemonset如何在命令行更新pod