故障etcd 节点 加入现有集群

故障etcd 节点 加入现有集群

恢复步骤：

1、查看故障节点的etcd状态并关闭服务

      systemctl status etcd  &&  systemctl stop etcd 

2、列出现有的集群节点ID

ETCDCTL_API=3 /opt/etcd/bin/etcdctl --cacert=/opt/etcd/ssl/ca.pem --cert=/opt/etcd/ssl/server.pem --key=/opt/etcd/ssl/server-key.pem --endpoints="https://x.x.x.2:2379,https://x.x.x.3:2379,https://x.x.x.4:2379" member list -w table

3、通过ID 删除故障节点

ETCDCTL_API=3 /opt/etcd/bin/etcdctl --cacert=/opt/etcd/ssl/ca.pem --cert=/opt/etcd/ssl/server.pem --key=/opt/etcd/ssl/server-key.pem --endpoints="https://x.x.x.2:2379,https://x.x.x.3:2379,https://x.x.x.4:2379"

member remove 48ccac483b85c1bd

4、删除故障节点的原数据目录

    rm /var/lib/etcd/default.etcd/*  -rf

5、修改故障ETCD 节点的 配置文件

/opt/etcd/cfg/etcd.conf  修改ETCD_INITIAL_CLUSTER_STATE="new" 为 ETCD_INITIAL_CLUSTER_STATE="existing"

6、把故障节点加入到现有集群

ETCDCTL_API=3 /opt/etcd/bin/etcdctl --cacert=/opt/etcd/ssl/ca.pem --cert=/opt/etcd/ssl/server.pem --key=/opt/etcd/ssl/server-key.pem --endpoints="https://x.x.x.2:2379,https://x.x.x.3:2379,https://x.x.x.4:2379" member add x.x.x.4  --peer-urls=https://x.x.x.4:2380 

7、 查看集群状态，状态为unstarted 需要重新启动etcd

ETCDCTL_API=3 /opt/etcd/bin/etcdctl --cacert=/opt/etcd/ssl/ca.pem --cert=/opt/etcd/ssl/server.pem --key=/opt/etcd/ssl/server-key.pem --endpoints="https://x.x.x.2:2379,https://x.x.x.3:2379,https://x.x.x.4:2379" member list -w table 

8、 重新启动故障节点etcd服务并重启apiserver 服务，再次查看ETCD集群状态就是 started 状态

 systemctl start etcd 

9、至此 数据会从其他2个节点同步到新加入的节点