准备三台机器，192.168.20.7，192.168.20.8，192.168.20.10用于搭建clickhouse集群。本次搭建的集群，为三副本的，即一份数据会在三台机器上分别存储，搭建集群只是为了容灾。
1. 在192.168.20.7上操作
在clickhouse config.d目录下新建cluster.xml，内容如下
<clickhouse>
<keeper_server>
<tcp_port>9181</tcp_port>
<server_id>1</server_id>
<log_storage_path>/data/clickhouse/coordination/log</log_storage_path> <snapshot_storage_path>/data/clickhouse/coordination/snapshots</snapshot_storage_path>
<coordination_settings>
<operation_timeout_ms>10000</operation_timeout_ms>
<session_timeout_ms>30000</session_timeout_ms>
<raft_logs_level>trace</raft_logs_level>
</coordination_settings>
<raft_configuration>
<server>
<id>1</id>
<hostname>192.168.20.7</hostname>
<port>9234</port>
</server>
<server>
<id>2</id>
<hostname>192.168.20.8</hostname>
<port>9234</port>
</server>
<server>
<id>3</id>
<hostname>192.168.20.10</hostname>
<port>9234</port>
</server>
</raft_configuration>
</keeper_server>
<remote_servers>
<default>
<shard>
<internal_replication>1</internal_replication>
<replica>
<host>192.168.20.7</host>
<port>8124</port>
</replica>
<replica>
<host>192.168.20.8</host>
<port>8124</port>
</replica>
<replica>
<host>192.168.20.10</host>
<port>8124</port>
</replica>
</shard>
</default>
</remote_servers>
<macros>
<shard>1</shard>
<replica>192.168.20.7</replica>
</macros>
<interserver_http_host>192.168.20.7</interserver_http_host>
</clickhouse>
2. 在192.168.20.8上操作
<clickhouse>
<keeper_server>
<tcp_port>9181</tcp_port>
<server_id>2</server_id>
<log_storage_path>/data/clickhouse/coordination/log</log_storage_path> <snapshot_storage_path>/data/clickhouse/coordination/snapshots</snapshot_storage_path>
<coordination_settings>
<operation_timeout_ms>10000</operation_timeout_ms>
<session_timeout_ms>30000</session_timeout_ms>
<raft_logs_level>trace</raft_logs_level>
</coordination_settings>
<raft_configuration>
<server>
<id>1</id>
<hostname>192.168.20.7</hostname>
<port>9234</port>
</server>
<server>
<id>2</id>
<hostname>192.168.20.8</hostname>
<port>9234</port>
</server>
<server>
<id>3</id>
<hostname>192.168.20.10</hostname>
<port>9234</port>
</server>
</raft_configuration>
</keeper_server>
<remote_servers>
<default>
<shard>
<internal_replication>1</internal_replication>
<replica>
<host>192.168.20.7</host>
<port>8124</port>
</replica>
<replica>
<host>192.168.20.8</host>
<port>8124</port>
</replica>
<replica>
<host>192.168.20.10</host>
<port>8124</port>
</replica>
</shard>
</default>
</remote_servers>
<macros>
<shard>1</shard>
<replica>192.168.20.8</replica>
</macros>
<interserver_http_host>192.168.20.8</interserver_http_host>
</clickhouse>
修改了server_id，macros.replica、interserver_http_host三个配置项
3. 在192.168.20.10上操作
<clickhouse>
<keeper_server>
<tcp_port>9181</tcp_port>
<server_id>3</server_id>
<log_storage_path>/data/clickhouse/coordination/log</log_storage_path>
<snapshot_storage_path>/data/clickhouse/coordination/snapshots</snapshot_storage_path>
<coordination_settings>
<operation_timeout_ms>10000</operation_timeout_ms>
<session_timeout_ms>30000</session_timeout_ms>
<raft_logs_level>trace</raft_logs_level>
</coordination_settings>
<raft_configuration>
<server>
<id>1</id>
<hostname>192.168.20.7</hostname>
<port>9234</port>
</server>
<server>
<id>2</id>
<hostname>192.168.20.8</hostname>
<port>9234</port>
</server>
<server>
<id>3</id>
<hostname>192.168.20.10</hostname>
<port>9234</port>
</server>
</raft_configuration>
</keeper_server>
<remote_servers>
<default>
<shard>
<internal_replication>1</internal_replication>
<replica>
<host>192.168.20.7</host>
<port>8124</port>
</replica>
<replica>
<host>192.168.20.8</host>
<port>8124</port>
</replica>
<replica>
<host>192.168.20.10</host>
<port>8124</port>
</replica>
</shard>
</default>
</remote_servers>
<macros>
<shard>1</shard>
<replica>192.168.20.10</replica>
</macros>
<interserver_http_host>192.168.20.10</interserver_http_host>
</clickhouse>
修改了server_id，macros.replica、interserver_http_host三个配置项
4. 启动
在三台机器上分别执行systemctl start clickhouse-server，启动clickhouse服务
启动成功之后，登录clickhouse执行语句，select * from system.clusters;可以看到三个节点，代表集群建立成功。
5.建表
clickhouse集群，需要在每个节点上分别建表，表结构如下所示
CREATE TABLE test.test1
(
`id` Int64
)
ENGINE = ReplicatedMergeTree('/clickhouse/tables/{shard}/test/test1', '{replica}')
ORDER BY id
SETTINGS index_granularity = 8192;
表建好之后，在其中一台节点执行insert后，在其它节点都可以看到该条数据。