0
点赞
收藏
分享

微信扫一扫

kubernetes二进制方式:etcd备份和恢复

杰森wang 2022-02-09 阅读 52
数据环境变量nginx
加载失败

Etcd是Kubernetes集群中的一个十分重要的组件，用于保存集群所有的网络配置和对象的状态信息。

整个kubernetes系统中一共有两个服务需要用到etcd用来协同和存储配置，分别是：


  • 网络插件flannel、对于其它网络插件也需要用到etcd存储网络的配置信息
  • kubernetes本身，包括各种对象的状态和元信息配置


设置环境变量

当操作kubernetes时，需设置环境变量，ETCDCTL_API=3

export ETCDCTL_API=3


# 或者在`/etc/profile`文件中添加环境变量
vi /etc/profile
...
export ETCDCTL_API=3
...
source /etc/profile


# 或者在命令执行前加 ETCDCTL_API=3
ETCDCTL_API=3 etcdctl --endpoints=$ENDPOINTS member list
#查看版本
#etcdctl  version -w table
etcdctl version: 3.4.9
API version: 3.


设置别名，避免每次得设置证书


$vim /etc/profile
    alias etcdctl="ETCDCTL_API=3 /data/local/etcd/bin/etcdctl --cacert=/data/local/etcd/ssl/ca.pem --cert=/data/local/etcd/ssl/server.pem --key=/data/local/etcd/ssl/server-key.pem --endpoints="https://192.168.253.227:2379,https://192.168.
    253.228:2379,https://192.168.253.229:2379""
    #alias etcdctl="ETCDCTL_API=3 /data/local/etcd/bin/etcdctl --cacert=/data/local/etcd/ssl/ca.pem --cert=/data/local/etcd/ssl/server.pem --key=/data/local/etcd/ssl/server-key.pem --endpoints="https://192.168.253.227:2379""
$etcdctl member list -w table
+------------------+---------+--------+------------------------------+------------------------------+------------+
|        ID        | STATUS  |  NAME  |          PEER ADDRS          |         CLIENT ADDRS         | IS LEARNER |
+------------------+---------+--------+------------------------------+------------------------------+------------+
|  c0142b9a19898be | started | etcd-1 | https://192.168.253.227:2380 | https://192.168.253.227:2379 |      false |
| 78d50c713efbd616 | started | etcd-2 | https://192.168.253.228:2380 | https://192.168.253.228:2379 |      false |
| bcb81ad99e84629a | started | etcd-3 | https://192.168.253.229:2380 | https://192.168.253.229:2379 |      false |
+------------------+---------+--------+------------------------------+------------------------------+------------+


查看pod



etcdctl get /registry/pods --prefix --keys-only

  • –keys-only 默认为true,只显示key,如果设置为false,会显示key的所有值.
  • –prefix 默认为true可以看到所有的子目录.


查看service


etcdctl get /registry/service --prefix --keys-only

备份


  2. 备份命令只能写etcd集群中的单个节点，不能写多个节点。
  3.  指定在/etc/profile别名中endpoints="https://192.168.253.227:2379" 一个etcd节点即可

etcdctl snapshot save   /data/backup/etcd/snapshot.db
{"level":"info","ts":1637575884.2848496,"caller":"snapshot/v3_snapshot.go:119","msg":"created temporary db file","path":"/data/backup/etcd/snapshot.db.part"}
{"level":"info","ts":"2021-11-22T18:11:24.297+0800","caller":"clientv3/maintenance.go:200","msg":"opened snapshot stream; downloading"}
{"level":"info","ts":1637575884.2974415,"caller":"snapshot/v3_snapshot.go:127","msg":"fetching snapshot","endpoint":"https://192.168.253.227:2379"}
{"level":"info","ts":"2021-11-22T18:11:24.340+0800","caller":"clientv3/maintenance.go:208","msg":"completed snapshot read; closing"}
{"level":"info","ts":1637575884.3451302,"caller":"snapshot/v3_snapshot.go:142","msg":"fetched snapshot","endpoint":"https://192.168.253.227:2379","size":"3.2 MB","took":0.060068191}
{"level":"info","ts":1637575884.34525,"caller":"snapshot/v3_snapshot.go:152","msg":"saved","path":"/data/backup/etcd/snapshot.db"}
Snapshot saved at /data/backup/etcd/snapshot.db


删除pod

kubectl  delete pod nginx
pod "nginx" deleted

恢复准备

停止所有Master 上 kube-apiserver 服务

systemctl stop kube-apiserver
# 确认 kube-apiserver 服务是否停止 
ps -ef | grep kube-apiserver


停止集群中 ETCD 服务

systemctl stop etcd

移除所有 ETCD 存储目录下数据（最好直接备份走从命名）


mv /var/lib/etcd/default.etcd /var/lib/etcd/default.etcd.bak

单td恢复备份

这边我已经在集群配置好/etc/profile，所以不用配置节点

etcdctl snapshot restore /data/backup/etcd/snapshot.db --data-dir=/var/lib/etcd/default.etcd


多etcd恢复备份

注：


  • 2380是提供集群其他节点通信的的，2379是提供HTTP API服务的，别改错了
  • 备份是一样的，都是etcd集群恢复也是在某一台etcd备份文件。


拷贝 ETCD 备份快照


scp /data/backup/etcd/snapshot.db 192.168.253.228:/data/backup/etcd/
scp /data/backup/etcd/snapshot.db 192.168.253.229:/data/backup/etcd/


test-k8s-master-227操作

etcdctl snapshot restore /data/backup/etcd/snapshot.db \
  --name etcd-1 \
  --initial-cluster "etcd-1=https://192.168.253.227:2380,etcd-2=https://192.168.253.228:2380,etcd-3=https://192.168.253.229:2380" \
  --initial-cluster-token etcd-cluster \
  --initial-advertise-peer-urls https://192.168.253.227:2380 \
  --data-dir=/var/lib/etcd/default.etcd


test-k8s-node-228操作

etcdctl snapshot restore /data/backup/etcd/snapshot.db \
  --name etcd-2 \
  --initial-cluster "etcd-1=https://192.168.253.227:2380,etcd-2=https://192.168.253.228:2380,etcd-3=https://192.168.253.229:2380" \
  --initial-cluster-token etcd-cluster \
  --initial-advertise-peer-urls https://192.168.253.228:2380 \
  --data-dir=/var/lib/etcd/default.etcd


test-k8s-node-229操作

etcdctl snapshot restore /data/backup/etcd/snapshot.db \
  --name etcd-3 \
  --initial-cluster "etcd-1=https://192.168.253.227:2380,etcd-2=https://192.168.253.228:2380,etcd-3=https://192.168.253.229:2380" \
  --initial-cluster-token etcd-cluster \
  --initial-advertise-peer-urls https://192.168.253.229:2380 \
  --data-dir=/var/lib/etcd/default.etcd


说明:

name 节点名称
data-dir 节点数据存储目录
initial-cluster 集群中所有节点的信息
initial-advertise-peer-urls 该节点同伴监听地址，这个值会告诉集群中其他节点
nitial-cluster-token 创建集群的 token，这个值每个集群保持唯一


上面三台 ETCD 都恢复完成后，依次登陆三台机器启动 ETCD

systemctl start etcd


三台 ETCD 启动完成，检查 ETCD 集群状态


ETCDCTL_API=3 etcdctl --cacert=/data/local/etcd/ssl/ca.pem --cert=/data/local/etcd/ssl/server.pem --key=/data/local/etcd/ssl/server-key.pem --endpoints=https://192.168.253.227:2379,https://192.168.253.228:2379,https://192.168.253.229:2379 endpoint health
#查看etcd成员列表
etcdctl member list -w table
+------------------+---------+--------+------------------------------+------------------------------+------------+
|        ID        | STATUS  |  NAME  |          PEER ADDRS          |         CLIENT ADDRS         | IS LEARNER |
+------------------+---------+--------+------------------------------+------------------------------+------------+
|  c0142b9a19898be | started | etcd-1 | https://192.168.253.227:2380 | https://192.168.253.227:2379 |      false |
| 78d50c713efbd616 | started | etcd-2 | https://192.168.253.228:2380 | https://192.168.253.228:2379 |      false |
| bcb81ad99e84629a | started | etcd-3 | https://192.168.253.229:2380 | https://192.168.253.229:2379 |      false |
+------------------+---------+--------+------------------------------+------------------------------+------------+


依次启动apiserver

systemctl start kube-apiserver
kubectl  get cs
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME                 STATUS    MESSAGE             ERROR
scheduler            Healthy   ok                  
controller-manager   Healthy   ok                  
etcd-0               Healthy   {"health":"true"}   
etcd-2               Healthy   {"health":"true"}   
etcd-1               Healthy   {"health":"true"}


查看pod是否恢复



kubectl  get po
NAME    READY   STATUS    RESTARTS   AGE
nginx   1/1     Running   0          18h


注意


  • flannel操作etcd使用的是v2的API，而kubernetes操作etcd使用的v3的API。
  • 若使用 v3 备份数据时存在 v2 的数据则不影响恢复。
  • 若使用 v2 备份数据时存在 v3 的数据则恢复失败。
  • 备份的节点最好不要单一，如果备份的节点，出现有没备份的情况，等恢复的时候，就悲剧了，最好备份同时在两个节点以上，然后给备份加个监控。
  • 备份ETCD集群时，只需要备份一个ETCD就行，恢复时，拿同一份备份数据恢复。
  • 恢复顺序:停止kube-apiserver --> 停止ETCD --> 恢复数据 --> 启动ETCD --> 启动kube-apiserver



举报

相关推荐

kubernetes 部署：二进制方式 部署 kubernetes 集群

Kubernetes 容器编排K8sDoker

kubernetes 1.20 二进制方式高可用部署

kuberneteslinuxjsonvim

【Kubernetes】Kubernetes之二进制部署

linuxarm

MySQL 二进制日志备份

mysqlgit备份文件虚拟化云计算

Kubernetes从二进制安装

kubernetes

MySQL xtrabackup全量备份+增量备份+二进制日志恢复实战

mysql增量备份数据

升级二进制kubernetes集群

gitlinuxgithub编程语言

kubernetes二进制部署 1.19

Docker初始化二进制文件kubernetes云计算

二进制方式安装Docker

dockerlinux环境变量Docker云计算

二进制方式安装mysql

数据库linux服务器
0 条评论
加载失败