集群安装

前期linux配置

在安装文件详解了

安装包准备

-rw-rw-r-- 1 bigdata bigdata     22419 Mar  9 11:51 clickhouse-client-24.1.2.5.x86_64.rpm
-rw-rw-r-- 1 bigdata bigdata 275571533 Feb  2 19:47 clickhouse-common-static-24.1.2.5.x86_64.rpm
-rw-rw-r-- 1 bigdata bigdata     51555 Mar  9 11:51 clickhouse-server-24.1.2.5.x86_64.rpm

安装

集群规划

主机	规划设置主机名	角色
192.168.61.202	hadoop1	zk、clickhouse
192.168.61.203	hadoop2	zk、clickhouse
192.168.61.204	hadoop3	zk、clickhouse

安装包安装

安装包分发到所有的机器（hadoop1,hadoop2,hadoop3）

./xsync.sh /home/bigdata/package/clickhouse

sudo rpm -ivh *.rpm

#出现Enter password for default user:直接回车就是空密码，后期配置文件配置

#安装的时候如果要输入密码直接回车就不需要输入密码

#clickhouse查看安装情况
sudo rpm -qa | grep clickhouse

[bigdata@hadoop1 clickhouse]$ sudo rpm -qa | grep clickhouse
clickhouse-server-24.1.2.5-1.x86_64
clickhouse-client-24.1.2.5-1.x86_64
clickhouse-common-static-24.1.2.5-1.x86_64

修改配置文件

进入配置文件目录(切换root配置文件)

cd /etc/clickhouse-server/

创建数据保存和日志保存目录

mkdir /home/bigdata/package/clickhouse/log
mkdir /home/bigdata/package/clickhouse/data
mkdir /home/bigdata/package/clickhouse/tmp
mkdir /home/bigdata/package/clickhouse/caches
chmod 777 -R /home/bigdata/package/clickhouse/log
chmod 777 -R /home/bigdata/package/clickhouse/data
chmod 777 -R /home/bigdata/package/clickhouse/tmp
chmod 777 -R /home/bigdata/package/clickhouse/caches
chown root:root -R /home/bigdata/package/clickhouse/log
chown root:root -R /home/bigdata/package/clickhouse/data
chown root:root -R /home/bigdata/package/clickhouse/tmp
chown root:root -R /home/bigdata/package/clickhouse/caches

config.xml

touch /etc/clickhouse-server/metrika.xml

vim /etc/clickhouse-server/config.xml

数据日志和数据文件保存位置

 <log>/home/bigdata/package/clickhouse/log/clickhouse-server.log</log>
 <errorlog>/home/bigdata/package/clickhouse/data/clickhouse-server.err.log</errorlog>
 <!-- Cache path for custom (created from SQL) cached disks -->
 <custom_cached_disks_base_directory>/home/bigdata/package/clickhouse/caches/</custom_cached_disks_base_directory>
 <!-- Path to data directory, with trailing slash. -->
 <path>/home/bigdata/package/clickhouse/data/</path>
 <!-- Path to temporary data for processing hard queries. -->
 <tmp_path>/home/bigdata/package/clickhouse/tmp/</tmp_path>
 <!-- Same for hosts without support for IPv6: -->
 <listen_host>0.0.0.0</listen_host>
 <!-- 下面这个要手动添加 -->
 <include_from>/etc/clickhouse-server/metrika.xml</include_from>
 <remote_servers incl="clickhouse_remote_servers" optional="true" />
 <zookeeper incl="zookeeper-servers" optional="true" />

metrika.xml

vim /etc/clickhouse-server/metrika.xml

<yandex>
    <!-- /etc/clickhouse-server/config.xml 中配置的remote_servers的incl属性值，需要在config.xml中指定，默认为remote_servers-->
    <clickhouse_remote_servers>
        <clickhouse_cluster><!-- 自定义的集群名 -->
            <!-- 数据分片1  -->
            <shard>
                 <internal_replication>true</internal_replication>
                 <!-- 分片1的副本，只设置一个，就是它本身  -->
                <replica>
                    <host>hadoop1</host>
                    <port>9000</port>
                </replica>
            </shard>
            <!-- 数据分片2  -->
            <shard>
                 <internal_replication>true</internal_replication>
                <replica>
                    <host>hadoop2</host>
                    <port>9000</port>
                </replica>
            </shard>
            <!-- 数据分片3  -->
            <shard>
                 <internal_replication>true</internal_replication>
                <replica>
                    <host>hadoop3</host>
                    <port>9000</port>
                </replica>
            </shard>
        </clickhouse_cluster><!-- 自定义的集群名 --> 
    </clickhouse_remote_servers>
    
    <!-- 本节点副本名称 -->
    <macros>
        <replica>hadoop1</replica>
    </macros>
    
    <!-- 监听网络-->
    <networks>
        <ip>::/0</ip>
    </networks>
    
    <!-- /etc/clickhouse-server/config.xml 中配置的zookeeper的incl属性值，需要在config.xml中指定，默认为zookeeper-->
    <!--zookeeper相关配置-->
    <zookeeper-servers>
        <!-- index是连接zookeeper的顺序，zookeeper中节点的index顺序需要与主机名或者ip匹配，不一定一一对应，可以去查看一下你部署的zookeeper主机映射，保证节点映射匹配-->
        <node index="1">
            <host>hadoop1</host>
            <port>2181</port>
        </node>
        <node index="2">
            <host>hadoop2</host>
            <port>2181</port>
        </node>
        <node index="3">
            <host>hadoop3</host>
            <port>2181</port>
        </node>
    </zookeeper-servers>
    
    <!-- 数据压缩算法  -->
    <clickhouse_compression>
        <case>
            <min_part_size>10000000000</min_part_size>
            <min_part_size_ratio>0.01</min_part_size_ratio>
            <method>lz4</method>
        </case>
    </clickhouse_compression>

</yandex>

分发配置文件

sudo ./xsync.sh /etc/clickhouse-server

修改其他机器的metrika.xml

vim /etc/clickhouse-server/metrika.xml

 <!-- 本节点副本名称 -->
    <macros>
        <replica>hadoop2</replica>
    </macros>

集群启动脚本

启动的时候打印日志

下面启动脚本启动不了时可以用这个启动。

clickhouse-server --config-file=/etc/clickhouse-server/config.xml

nohup clickhouse-server --config-file=/etc/clickhouse-server/config.xml &

集群启动脚本

报错检查

journalctl -u clickhouse-server

#!/bin/bash
 
case $1 in
"start"){
    for i in hadoop1 hadoop2 hadoop3
    do
        echo " --------启动 $i clickhouse-------"
        ssh $i "sudo systemctl start clickhouse-server"
    done
};;
"status"){
    for i in hadoop1 hadoop2 hadoop3
    do
        echo " --------启动 $i clickhouse-------"
        ssh $i "systemctl status clickhouse-server"
    done
};;
"stop"){
    for i in hadoop1 hadoop2 hadoop3
    do
        echo " --------停止 $i clickhouse-------"
        ssh $i "sudo systemctl stop clickhouse-server"
    done
};;
esac

客户端连接脚本

clickhouse-client  -m -u default

clickhouse-client -h ip地址 -d default -m -u default --password 密码明文

查看集群启动情况

SELECT * FROM system.clusters

web界面

http://hadoop1:8123/play