集群与高可用

# 节点 1
etcd --name etcd1 \
  --initial-advertise-peer-urls http://10.0.0.1:2380 \
  --listen-peer-urls http://0.0.0.0:2380 \
  --listen-client-urls http://0.0.0.0:2379 \
  --advertise-client-urls http://10.0.0.1:2379 \
  --initial-cluster-token etcd-cluster-1 \
  --initial-cluster etcd1=http://10.0.0.1:2380,etcd2=http://10.0.0.2:2380,etcd3=http://10.0.0.3:2380 \
  --initial-cluster-state new \
  --data-dir /var/lib/etcd

# 节点 2
etcd --name etcd2 \
  --initial-advertise-peer-urls http://10.0.0.2:2380 \
  --listen-peer-urls http://0.0.0.0:2380 \
  --listen-client-urls http://0.0.0.0:2379 \
  --advertise-client-urls http://10.0.0.2:2379 \
  --initial-cluster-token etcd-cluster-1 \
  --initial-cluster etcd1=http://10.0.0.1:2380,etcd2=http://10.0.0.2:2380,etcd3=http://10.0.0.3:2380 \
  --initial-cluster-state new \
  --data-dir /var/lib/etcd

# 节点 3
etcd --name etcd3 \
  --initial-advertise-peer-urls http://10.0.0.3:2380 \
  --listen-peer-urls http://0.0.0.0:2380 \
  --listen-client-urls http://0.0.0.0:2379 \
  --advertise-client-urls http://10.0.0.3:2379 \
  --initial-cluster-token etcd-cluster-1 \
  --initial-cluster etcd1=http://10.0.0.1:2380,etcd2=http://10.0.0.2:2380,etcd3=http://10.0.0.3:2380 \
  --initial-cluster-state new \
  --data-dir /var/lib/etcd

# 使用已有的 etcd 集群作为 discovery
etcd --discovery-srv example.com \
  --initial-cluster-state new \
  ...

# 使用 public discovery 服务（仅用于开发测试）
etcd --discovery https://discovery.etcd.io/new?size=3 \
  --initial-cluster-state new \
  ...

cluster := clientv3.NewCluster(cli)

// 列出所有成员
resp, err := cluster.MemberList(ctx)
for _, m := range resp.Members {
    fmt.Printf("ID=%x Name=%s PeerURLs=%v ClientURLs=%v IsLearner=%v\n",
        m.ID, m.Name, m.PeerURLs, m.ClientURLs, m.IsLearner)
}

// 添加成员
resp2, err := cluster.MemberAdd(ctx, []string{"http://10.0.0.4:2380"})

// 添加 Learner（不参与投票，用于热备份或跨地域复制）
resp3, err := cluster.MemberAddAsLearner(ctx, []string{"http://10.0.0.5:2380"})

// Learner 提升为 Voter
_, err = cluster.MemberPromote(ctx, learnerID)

// 移除成员
_, err = cluster.MemberRemove(ctx, memberID)

// 更新成员的 PeerURLs
_, err = cluster.MemberUpdate(ctx, memberID, []string{"http://new-host:2380"})

etcdctl member list                    # 列出成员
etcdctl member add etcd4 --peer-urls=http://10.0.0.4:2380
etcdctl member remove <member-id>
etcdctl member update <id> --peer-urls=http://new:2380
etcdctl member promote <learner-id>

maintenance := clientv3.NewMaintenance(cli)

// 创建快照并保存到文件
reader, err := maintenance.Snapshot(ctx)
if err != nil {
    log.Fatal(err)
}
defer reader.Close()

f, _ := os.Create("backup.db")
defer f.Close()
io.Copy(f, reader)

# 命令行
etcdctl snapshot save backup.db

# 1. 从快照恢复数据目录
etcdctl snapshot restore backup.db \
  --name etcd-restored \
  --data-dir /var/lib/etcd-restored \
  --initial-cluster etcd-restored=http://localhost:2380 \
  --initial-advertise-peer-urls http://localhost:2380

# 2. 以恢复后的数据目录启动
etcd --data-dir /var/lib/etcd-restored ...

# 查看 DB 大小和剩余空间
etcdctl endpoint status --write-out=table

# 压缩历史版本（清理旧 revision）
etcdctl compact <latest-revision>

# 碎片整理（释放空间回操作系统）
etcdctl defrag

# 集群级别碎片整理
etcdctl defrag --cluster

# etcd 存储空间超过 --quota-backend-bytes 的 95% 发出 NOSPACE 告警
etcdctl alarm list
# memberID:xxx alarm:NOSPACE

# 此时写入会被拒绝 → 必须先解除告警
etcdctl compact <rev>
etcdctl defrag
etcdctl alarm disarm

maintenance := clientv3.NewMaintenance(cli)

// 获取节点状态
status, _ := maintenance.Status(ctx, "localhost:2379")
fmt.Printf("Version=%s, DB Size=%d MB, Leader=%x, IsLearner=%v\n",
    status.Version, status.DbSize/1024/1024, status.Leader, status.IsLearner)

// 哈希一致性检查
hashResp, _ := maintenance.HashKV(ctx, "localhost:2379", 0)
fmt.Printf("Hash=%d, CompactRev=%d\n", hashResp.Hash, hashResp.CompactRevision)

// 碎片整理
_, err := maintenance.Defragment(ctx, "localhost:2379")

// 强制切换 Leader
_, err := maintenance.MoveLeader(ctx, targetMemberID)

// 告警列表
alarms, _ := maintenance.AlarmList(ctx)
for _, a := range alarms.Alarms {
    if a.Alarm == pb.AlarmType_NOSPACE {
        log.Println("警告：存储空间不足！")
    }
}