k8s etcd备份与恢复

1. 停止api-server（k8s的所有master节点）

   # 所有master节点执行：
   # 停api-server
   
   mkdir -p tpm_api_conf
   mv  /etc/kubernetes/manifests/kube-apiserver.yaml /root/tpm_api_conf/
   

2. etcd备份（集群中某一节点）

   # etcd备份
   
   ETCDCTL_API=3;/usr/local/bin/etcdctl --endpoints='https://192.168.1.30:2379' --cacert="/etc/ssl/etcd/ssl/ca.pem" --cert="/etc/ssl/etcd/ssl/admin-ks-master01.pem" --key="/etc/ssl/etcd/ssl/admin-ks-master01-key.pem" snapshot save  snapshot_20230928.db 
   

3. 查看备份数据状态

   # 查看备份数据状态
   
   ETCDCTL_API=3;/usr/local/bin/etcdctl --write-out=table snapshot status snapshot_20230928.db
   

4. 停止etcd服务并备份数据目录（etcd所有节点）

   # 停etcd:
   
   systemctl stop etcd
   mv /var/lib/etcd/ /root/etcd_bak
   

5. 各个etcd节点恢复数据

   # 节点30:
   
   ETCDCTL_API=3;/usr/local/bin/etcdctl snapshot restore /root/snapshot_20230928.db \
    --name etcd-ks-master01  \
    --cert="/etc/ssl/etcd/ssl/admin-ks-master01.pem" \
    --key="/etc/ssl/etcd/ssl/admin-ks-master01-key.pem"  \
    --cacert="/etc/ssl/etcd/ssl/ca.pem"   \
    --endpoints="https://127.0.0.1:2379" \
    --initial-advertise-peer-urls="https://192.168.1.30:2380"  \
    --initial-cluster="etcd-ks-master01=https://192.168.1.30:2380,etcd-ks-master02=https://192.168.1.31:2380,etcd-ks-master03=https://192.168.1.32:2380" \
    --data-dir=/var/lib/etcd
    
   # 节点31:
   
   ETCDCTL_API=3;/usr/local/bin/etcdctl snapshot restore /root/snapshot_20230928.db \
    --name etcd-ks-master02  \
    --cert="/etc/ssl/etcd/ssl/admin-ks-master02.pem" \
    --key="/etc/ssl/etcd/ssl/admin-ks-master02-key.pem"  \
    --cacert="/etc/ssl/etcd/ssl/ca.pem"   \
    --endpoints="https://127.0.0.1:2379" \
    --initial-advertise-peer-urls="https://192.168.1.31:2380"  \
    --initial-cluster="etcd-ks-master01=https://192.168.1.30:2380,etcd-ks-master02=https://192.168.1.31:2380,etcd-ks-master03=https://192.168.1.32:2380" \
    --data-dir=/var/lib/etcd
    
   # 节点32:
   
   ETCDCTL_API=3;/usr/local/bin/etcdctl snapshot restore /root/snapshot_20230928.db \
    --name etcd-ks-master03  \
    --cert="/etc/ssl/etcd/ssl/admin-ks-master03.pem" \
    --key="/etc/ssl/etcd/ssl/admin-ks-master03-key.pem"  \
    --cacert="/etc/ssl/etcd/ssl/ca.pem"   \
    --endpoints="https://127.0.0.1:2379" \
    --initial-advertise-peer-urls="https://192.168.1.32:2380"  \
    --initial-cluster="etcd-ks-master01=https://192.168.1.30:2380,etcd-ks-master02=https://192.168.1.31:2380,etcd-ks-master03=https://192.168.1.32:2380" \
    --data-dir=/var/lib/etcd
   

6. etcd节点修改数据目录权限并启动etcd

   # 所有节点执行：
   # 修改属组
   chown -R etcd:root /var/lib/etcd/
   
   # 启动etcd
   systemctl start etcd
   

7. 启动api-server（所有k8s master节点）

   # etcd启动完成后，恢复api配置
   mv  /root/tpm_api_conf/kube-apiserver.yaml   /etc/kubernetes/manifests/kube-apiserver.yaml