K8s之Etcd的备份与恢复

ETCD简介

ETCD用于共享和配置服务发现的分布式，一致性的KV存储系统。 ETCD是CoreOS公司发起的一个开源项目，授权协议为Apache。

ETCD 存储 k8s 所有数据信息

ETCD 是k8s集群极为重要的一块服务，存储了集群所有的数据信息。同理，如果发生灾难或者 etcd 的数据丢失，都会影响集群数据的恢复。

ETCD使用场景

ETCD 有很多使用场景，包括但不限于：

配置管理
服务注册于发现
选主
应用调度
分布式队列
分布式锁

首先在master节点安装etcd

[root@master ~]# yum install etcd -y

查看etcd版本

[root@master ~]# etcdctl -version
etcdctl version: 3.3.11
API version: 2

对 etcd 的访问相当于集群中的 root 权限，因此理想情况下只有 API 服务器才能访问它。考虑到数据的敏感性，建议只向需要访问 etcd 集群的节点授予权限。要使用安全客户端通信对 etcd 进行配置，请指定参数 --key-file=k8sclient.key和--cert-file=k8sclient.cert ,并使用 HTTPS 作为 URL 模式。使用安全通信的客户端命令的示例：

[root@master ~]# ETCDCTL_API=3 etcdctl --endpoints 192.168.248.128:2379 \  
> --cert=/etc/kubernetes/pki/etcd/server.crt \
> --key=/etc/kubernetes/pki/etcd/server.key \
> --cacert=/etc/kubernetes/pki/etcd/ca.crt \
> member list
6b96994e8b1eabe5, started, master, https://192.168.248.128:2380, https://192.168.248.128:2379

注：etcd最新的API版本是v3，与v2相比，v3更高效更清晰。k8s默认使用的etcd V3版本API，ectdctl默认使用V2版本API。要使用v3，设置环境变量export ETCDCTL_API=3临时更改为V3或者vim /etc/profile后在里面添加export ETCDCTL_API=3，然后执行source /etc/profile则永久更改为V3。

2379和2380为etcd在IANA 的注册端口【为默认端口】

2379:为客户端提供通讯
2380:为服务器间提供通讯

单节点etcd备份恢复示例：

先在主机上运行一个deployment，里面起5个pod：

[root@master ~]# kubectl get pod
NAME                               READY   STATUS    RESTARTS   AGE
deployment-nginx-8c459867c-4fw9q   1/1     Running   1          4d19h
deployment-nginx-8c459867c-ccbkz   1/1     Running   1          4d19h
deployment-nginx-8c459867c-cksqd   1/1     Running   1          4d19h
deployment-nginx-8c459867c-mrdz4   1/1     Running   1          4d19h
deployment-nginx-8c459867c-zt8n8   1/1     Running   1          4d19h

创建一个目录，存放etcd数据，然后以快照方式保存现在etcd的数据。

[root@master ~]# mkdir -p etcd/backup/
[root@master ~]#  ETCDCTL_API=3 etcdctl --endpoints 192.168.248.128:2379 \
>  --cert=/etc/kubernetes/pki/etcd/server.crt \
>  --key=/etc/kubernetes/pki/etcd/server.key \
>  --cacert=/etc/kubernetes/pki/etcd/ca.crt \
> snapshot save ~/etcd/backup/snap.db
Snapshot saved at /root/etcd/backup/snap.db
[root@master ~]# cd etcd/backup/
[root@master backup]# ls
snap.db

现在把上面的pod删除后再恢复etcd数据，验证是否能恢复pod。

[root@master ~]# kubectl delete -f deployment2.yaml 
deployment.apps "deployment-nginx" deleted
[root@master ~]# kubectl get pods
No resources found in default namespace   #此时已经没有pod在运行了

恢复etcd，首先停止kube-apiserver和etcd，防止再有数据写入etcd。由于kube-apiserver和etcd属于静态pod，是由kubelet创建，所以需要将/etc/kubernetes/manifests/下的yaml文件移除，让其不可用。

[root@master ~]# mv /etc/kubernetes/manifests/ /etc/kubernetes/manifests.bak
[root@master ~]# mv /var/lib/etcd/ /var/lib/etcd.bak

[root@master ~]# kubectl get pod -A 
The connection to the server 192.168.248.128:6443 was refused - did you specify the right host or port?

此时kubectl以不可用，现在恢复etcd数据

[root@master lib]# ETCDCTL_API=3 etcdctl snapshot restore /root/etcd/backup/snap.db --data-dir=/var/lib/etcd
2022-08-11 22:09:39.297931 I | mvcc: restore compact to 306417
2022-08-11 22:09:39.304403 I | etcdserver/membership: added member 8e9e05c52164694d [http://localhost:2380] to cluster cdf818194e3a8c32

将/etc/kubernetes/manifests/下的yaml文件还原，静态pod会自动重建kube-apiserver和etcd

[root@master lib]# mv /etc/kubernetes/manifests.bak/ /etc/kubernetes/manifests

[root@master lib]# kubectl get pod
NAME                               READY   STATUS    RESTARTS   AGE
deployment-nginx-8c459867c-4fw9q   1/1     Running   1          5d
deployment-nginx-8c459867c-ccbkz   1/1     Running   1          5d
deployment-nginx-8c459867c-cksqd   1/1     Running   1          5d
deployment-nginx-8c459867c-mrdz4   1/1     Running   1          5d
deployment-nginx-8c459867c-zt8n8   1/1     Running   1          5d

此时etcd数据恢复后，原来的pod又重新回来了。

etcd自动备份脚本。

[root@master etcd]# cat etcd_backup.sh 
#!/bin/bash
CACERT="/etc/kubernetes/pki/etcd/ca.crt "
CERT="/etc/kubernetes/pki/etcd/server.crt"
EKY="/etc/kubernetes/pki/etcd/server.key"
ENDPOINTS="192.168.248.128:2379"

ETCDCTL_API=3 etcdctl \
--cacert="${CACERT}" \
--cert="${CERT}" \
--key="${EKY}" \
--endpoints=${ENDPOINTS} \
snapshot save /data/etcd_backup_dir/etcd-snapshot-`date +%Y%m%d`.db

# 备份保留30天
find /data/etcd_backup_dir/ -name *.db -mtime +30 -exec rm -f {} \;

多master节点集群恢复etcd

只需要备份一个master节点数据就可以了，当我们master节点数据备份后打包传到其他节点上

mkdir /var/lib/etcd_backup                          #若没有，创建etcd_backup目录；若有，则不用创建
cp $BACKUP_FILE_NAME.tar /var/lib/etcd_backup       #拷贝tar包至etcd_backup目录
cd /var/lib/etcd_backup                             #进入/var/lib/etcd_backup目录
tar -xzvf $BACKUP_FILE_NAME.tar                     #解压备份的tar包

停止所有节点上的kube-apiserver和etcd

systemctl stop kube-apiserver
systemctl stop etcd

移除所有etcd服务实例的数据目录

mv /var/lib/etcd/default.etcd /var/lib/etcd/default.etcd_bak

执行恢复命令，所有etcd节点依次执行

etcdctl snapshot restore /var/lib/etcd_backup/$BACKUP_FILE_NAME/etcd_snapshot.db \  #备份文件
--cacert=$ETCD_TRUSTED_CA_FILE  \                                                   #ca.crt路径
--cert=$ETCD_CERT_FILE  \                                                           #server.crt路径
--key=$ETCD_KEY_FILE \                                                              #server.key路径
--name $ETCD_NAME \                                                                 #主机名
--initial-cluster $ETCD_INITIAL_CLUSTER  \                                          #描述集群节点信息 
--initial-advertise-peer-urls $ETCD_INITIAL_ADVERTISE_PEER_URLS \                   #etcd主机IP
--data-dir=/var/lib/etcd/default.etcd

备注 ：vim /etc/etcd/etcd.conf，查看配置信息，以及所需字段：ETCD_TRUSTED_CA_FILE、ETCD_CERT_FILE、ETCD_KEY_FILE、ETCD_NAME、ETCD_INITIAL_CLUSTER、ETCD_INITIAL_ADVERTISE_PEER_URLS

例：

# k8s-master1 机器上操作
$ ETCDCTL_API=3 etcdctl snapshot restore /tmp/backup/etcd/etcd-snapshot-20210610.db \
  --name k8s-m1 \
  --initial-cluster "k8s-m1=https://172.16.2.91:2380，k8s-m2=https://172.16.2.92:2380，k8s-m3=https://172.16.2.93:2380" \
  --initial-cluster-token etcd-cluster \
  --initial-advertise-peer-urls https://172.16.2.91:2380 \
  --data-dir=/var/lib/etcd/default.etcd
  
# k8s-master2 机器上操作
$ ETCDCTL_API=3 etcdctl snapshot restore /tmp/backup/etcd/etcd-snapshot-20210610.db \
  --name k8s-m2 \
  --initial-cluster "k8s-m1=https://172.16.2.91:2380，k8s-m2=https://172.16.2.92:2380，k8s-m3=https://172.16.2.93:2380"  \
  --initial-cluster-token etcd-cluster \
  --initial-advertise-peer-urls https://172.16.2.92:2380 \
  --data-dir=/var/lib/etcd/default.etcd
  
# k8s-master3 机器上操作
$ ETCDCTL_API=3 etcdctl snapshot restore /tmp/backup/etcd/etcd-snapshot-20210610.db \
  --name k8s-m3 \
  --initial-cluster "k8s-m1=https://172.16.2.91:2380，k8s-m2=https://172.16.2.92:2380，k8s-m3=https://172.16.2.93:2380"  \
  --initial-cluster-token etcd-cluster \
  --initial-advertise-peer-urls https://172.16.2.93:2380 \
  --data-dir=/var/lib/etcd/default.etcd

修改etcd数据目录权限

chown -R etcd:etcd /var/lib/etcd/default.etcd
chmod -R 700 /var/lib/etcd/default.etcd

启动集群所有的etcd实例

systemctl start etcd

检查所有etcd健康状态

ETCDCTL_API=3 etcdctl --cacert=/etc/kubernetes/pki/etcd/ca.crt \
--cert=/etc/kubernetes/pki/etcd/server.crt \
--key=/etc/kubernetes/pki/etcd/server.key  \
--endpoints="https://172.16.2.91:2379,https://172.16.2.92:2379,https://172.16.2.93:2379" \
 endpoint health -w table

到每台 Master 启动 kube-apiserver

 systemctl start kube-apiserver
systemctl status kube-apiserver

检查 Kubernetes 集群是否恢复正常

[root@master ~]# kubectl get cs
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME                 STATUS      MESSAGE                                                                                       ERROR
controller-manager   Unhealthy   Get "http://127.0.0.1:10252/healthz": dial tcp 127.0.0.1:10252: connect: connection refused   
scheduler            Unhealthy   Get "http://127.0.0.1:10251/healthz": dial tcp 127.0.0.1:10251: connect: connection refused   
etcd-0               Healthy     {"health":"true"}

如果出现上述情况，需要将/etc/kubernetes/manifests/kube-scheduler.yaml和/etc/kubernetes/manifests/kube-controller-manager.yaml中的 - --port=0注释掉重启kubelet就恢复正常了。

apiVersion: v1
kind: Pod
metadata:
  creationTimestamp: null
  labels:
    component: kube-scheduler
    tier: control-plane
  name: kube-scheduler
  namespace: kube-system
spec:
  containers:
  - command:
    - kube-scheduler
    - --authentication-kubeconfig=/etc/kubernetes/scheduler.conf
    - --authorization-kubeconfig=/etc/kubernetes/scheduler.conf
    - --bind-address=127.0.0.1
    - --kubeconfig=/etc/kubernetes/scheduler.conf
    - --leader-elect=true
#    - --port=0
    image: registry.aliyuncs.com/google_containers/kube-scheduler:v1.21.0
    imagePullPolicy: IfNotPresent
    livenessProbe:
      failureThreshold: 8
      httpGet:
        host: 127.0.0.1
        path: /healthz
        port: 10259
        scheme: HTTPS
      initialDelaySeconds: 10
      periodSeconds: 10
      timeoutSeconds: 15
    name: kube-scheduler
    resources:
      requests:
        cpu: 100m
    startupProbe:
      failureThreshold: 24
      httpGet:
        host: 127.0.0.1
        path: /healthz
        port: 10259
        scheme: HTTPS
      initialDelaySeconds: 10
      periodSeconds: 10
      timeoutSeconds: 15
    volumeMounts:
    - mountPath: /etc/kubernetes/scheduler.conf
      name: kubeconfig
      readOnly: true
  hostNetwork: true
  priorityClassName: system-node-critical
  volumes:
  - hostPath:
      path: /etc/kubernetes/scheduler.conf
      type: FileOrCreate
    name: kubeconfig
status: {}

/etc/kubernetes/manifests/kube-scheduler.yaml

apiVersion: v1
kind: Pod
metadata:
  creationTimestamp: null
  labels:
    component: kube-controller-manager
    tier: control-plane
  name: kube-controller-manager
  namespace: kube-system
spec:
  containers:
  - command:
    - kube-controller-manager
    - --allocate-node-cidrs=true
    - --authentication-kubeconfig=/etc/kubernetes/controller-manager.conf
    - --authorization-kubeconfig=/etc/kubernetes/controller-manager.conf
    - --bind-address=127.0.0.1
    - --client-ca-file=/etc/kubernetes/pki/ca.crt
    - --cluster-cidr=10.244.0.0/16
    - --cluster-name=kubernetes
    - --cluster-signing-cert-file=/etc/kubernetes/pki/ca.crt
    - --cluster-signing-key-file=/etc/kubernetes/pki/ca.key
    - --controllers=*,bootstrapsigner,tokencleaner
    - --kubeconfig=/etc/kubernetes/controller-manager.conf
    - --leader-elect=true
#    - --port=0
    - --requestheader-client-ca-file=/etc/kubernetes/pki/front-proxy-ca.crt
    - --root-ca-file=/etc/kubernetes/pki/ca.crt
    - --service-account-private-key-file=/etc/kubernetes/pki/sa.key
    - --service-cluster-ip-range=10.96.0.0/12
    - --use-service-account-credentials=true
    image: registry.aliyuncs.com/google_containers/kube-controller-manager:v1.21.0
    imagePullPolicy: IfNotPresent
    livenessProbe:
      failureThreshold: 8
      httpGet:
        host: 127.0.0.1
        path: /healthz
        port: 10257
        scheme: HTTPS
      initialDelaySeconds: 10
      periodSeconds: 10
      timeoutSeconds: 15
    name: kube-controller-manager
    resources:
      requests:
        cpu: 200m
    startupProbe:
      failureThreshold: 24
      httpGet:
        host: 127.0.0.1
        path: /healthz
        port: 10257
        scheme: HTTPS
      initialDelaySeconds: 10
      periodSeconds: 10
      timeoutSeconds: 15
    volumeMounts:
    - mountPath: /etc/ssl/certs
      name: ca-certs
      readOnly: true
    - mountPath: /etc/pki
      name: etc-pki
      readOnly: true
    - mountPath: /usr/libexec/kubernetes/kubelet-plugins/volume/exec
      name: flexvolume-dir
    - mountPath: /etc/kubernetes/pki
      name: k8s-certs
      readOnly: true
    - mountPath: /etc/kubernetes/controller-manager.conf
      name: kubeconfig
      readOnly: true
  hostNetwork: true
  priorityClassName: system-node-critical
  volumes:
  - hostPath:
      path: /etc/ssl/certs
      type: DirectoryOrCreate
    name: ca-certs
  - hostPath:
      path: /etc/pki
      type: DirectoryOrCreate
    name: etc-pki
  - hostPath:
      path: /usr/libexec/kubernetes/kubelet-plugins/volume/exec
      type: DirectoryOrCreate
    name: flexvolume-dir
  - hostPath:
      path: /etc/kubernetes/pki
      type: DirectoryOrCreate
    name: k8s-certs
  - hostPath:
      path: /etc/kubernetes/controller-manager.conf
      type: FileOrCreate
    name: kubeconfig
status: {}

/etc/kubernetes/manifests/kube-controller-manager.yaml

[root@master ~]# 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"}