clickhouse集群部署

一、集群部署简介

部署的详情可以看官网
先部署两个server,三个keeper[zookeeper]
clickhouse之前依赖的存储是zookeeper,后来改为了keeper,官网给出了原因
所以这就决定了clickhouse有两种安装方式，依赖于keeper做存储或者依赖于zookeeper做存储

二、zookeeper作为存储

1.zookeeper安装

zookeeper安装可以看之前的文章

2.clickhouse部署

修改配置文件

2.1 修改listen_host

<!-- Same for hosts without support for IPv6: -->
<listen_host>0.0.0.0</listen_host>  <!-- 把这里注释去掉，允许所有地址可以访问 -->

2.2 修改存储路径

<!-- Path to data directory, with trailing slash. --><path>/var/lib/clickhouse/</path>

2.3 添加集群配置

<!--这属于两个分片，每个分片语一个副本的架构-->
<remote_servers><cluster_2S_1R> <!--定义的集群名字--><shard><internal_replication>true</internal_replication><replica><host>chnode1</host><port>9000</port></replica></shard><shard><internal_replication>true</internal_replication><replica><host>chnode2</host><port>9000</port></replica></shard></cluster_2S_1R></remote_servers><!--
注意，上面的写法是放到两个shard里，也可放到一个shard,下面是单一分片两副本的写法，如果放到不同的shard里macros的配置就得不同了
<remote_servers><cluster_2S_1R> <!--定义的集群名字--><shard><internal_replication>true</internal_replication><replica><host>chnode1</host><port>9000</port></replica><replica><host>chnode2</host><port>9000</port></replica></shard></cluster_2S_1R></remote_servers>-->

2.4配置zookeeper

<zookeeper><node><host>example1</host><port>2181</port></node><node><host>example2</host><port>2181</port></node><node><host>example3</host><port>2181</port></node>
</zookeeper>

2.5配置macros

<!--如果是单分片量副本的配置-->
<!-- 配置分片macros变量，在用client创建表的时候会自动带入，第一台ck的配置 --><macros><shard>01</shard><replica>chnode1</replica> <!-- 这里指定当前集群节点的名字或者IP --></macros>
<!-- 配置分片macros变量，在用client创建表的时候会自动带入，第二台ck的配置 --><macros><shard>01</shard><replica>chnode2</replica> <!-- 这里指定当前集群节点的名字或者IP --></macros>

3.启动clickhouse

systemctl start clickhouse-server.service
systemctl enable clickhouse-server.service

4.登录机器并检查集群

4.1登录

# 登录
clickhouse-client
# 查看集群信息
select * from system.clusters

4.2建表

CREATE TABLE t1 ON CLUSTER cluster_2S_1R
(`ts` DateTime,`uid` String,`biz` String
)
ENGINE = ReplicatedMergeTree('/clickhouse/test1/tables/{shard}/t1', '{replica}')
PARTITION BY toYYYYMMDD(ts)
ORDER BY ts
SETTINGS index_granularity = 8192# 出现如下报错
Received exception from server (version 23.6.2):
Code: 159. DB::Exception: Received from localhost:9000. DB::Exception: Watching task /clickhouse/task_queue/ddl/query-0000000004 is executing longer than distributed_ddl_task_timeout (=180) seconds. There are 2 unfinished hosts (0 of them are currently active), they are going to execute the query in background. (TIMEOUT_EXCEEDED)
# 这个报错是某些ck服务异常才出现的报错，我这是因为我配置文件里的remote_server里的host ip写错了，相当于找不到服务了,修改后重启就好了

4.3 测试dml

目前DDL生效，但是插入数据在其他节点不生效
查看节点2的clickhouse日志，其中会有如下报错

2023.08.10 15:49:54.836507 [ 8514 ] {} <Error> test1.t1 (*****-48d4-44ed-9bad-2a03410321a9): auto DB::StorageReplicatedMergeTree::processQueueEntry(ReplicatedMergeTreeQueue::SelectedEntryPtr)::(anonymous class)::operator()(LogEntryPtr &) const: Code: 198. DB::Exception: Not found address of host: bj-ck3. (DNS_ERROR), Stack trace (when copying this message, always include the lines below):

可以看到这里是因为域名无法解析，因为ZooKeeper 里面存储的是hosts域名，不是IP，所以需要配置/etc/hosts

192.168.1.1    bj-ck1
192.168.1.2    bj-ck2
192.168.1.3    bj-ck3

ps: /etc/hosts的配置里，如果配置多个的话，是以第一个为准，其他都类似别名么
比如192.168.1.1配置如下:192.168.1.1 bj-1 bj-2
如果别的机器是以域名访问192.168.1.1，如果别的机器只配置了192.168.1.1 bj-2,其实是解析不到192.168.1.1的

三、keeper作为存储

ClickHouse Keeper 提供数据复制和分布式 DDL 查询执行的协调系统。 ClickHouse Keeper 与 Apache ZooKeeper 兼容。此配置在端口 9181 上启用 ClickHouse Keeper。

注意：
如果出于任何原因更换或重建 Keeper 节点，请勿重复使用现有的 server_id。例如，如果重建了server_id为2的Keeper节点，则将其server_id设置为4或更高。

分片和副本降低了分布式 DDL 的复杂性。配置的值会自动替换到您的 DDL 查询中，从而简化您的 DDL。

1.安装并启动keeper

# 安装clickhouse-keeper
sudo apt-get install -y clickhouse-keeper
# 启用并启动clickhouse-keeper
sudo systemctl enable clickhouse-keeper
sudo systemctl start clickhouse-keeper
sudo systemctl status clickhouse-keeper

2.修改keeper配置文件keeper_config.xml

 <keeper_server><tcp_port>9181</tcp_port><!-- 这里是主要的修改位置，保证集群中每个几点的id是唯一的 --><server_id>1</server_id><log_storage_path>/var/lib/clickhouse/coordination/logs</log_storage_path><snapshot_storage_path>/var/lib/clickhouse/coordination/snapshots</snapshot_storage_path><coordination_settings><operation_timeout_ms>10000</operation_timeout_ms><min_session_timeout_ms>10000</min_session_timeout_ms><session_timeout_ms>100000</session_timeout_ms><raft_logs_level>information</raft_logs_level><!-- All settings listed in https://github.com/ClickHouse/ClickHouse/blob/master/src/Coordination/CoordinationSettings.h --></coordination_settings><!-- enable sanity hostname checks for cluster configuration (e.g. if localhost is used with remote endpoints) --><hostname_checks_enabled>true</hostname_checks_enabled><!-- 这里是第二处需要变更的位置，需要把集群中的keeper配置上 --><raft_configuration><server><id>1</id><!-- Internal port and hostname --><hostname>192.168.1.1</hostname><port>9234</port></server><server><id>2</id><!-- Internal port and hostname --><hostname>192.168.1.2</hostname><port>9234</port></server><server><id>3</id><!-- Internal port and hostname --><hostname>192.168.1.3</hostname><port>9234</port></server><!-- Add more servers here --></raft_configuration></keeper_server><clickhouse><zookeeper><node index="1"><host>chnode1</host><port>9181</port></node><node index="2"><host>chnode2</host><port>9181</port></node><node index="3"><host>chnode3</host><port>9181</port></node></zookeeper></clickhouse>

3.clickhouse的配置

clickhouse的配置与zookeeper作为存储时的配置几乎一致，只需要把zookeeper的配置注释掉即可

ps: 这里还有个小插曲，使用keeper的时候发现dml的数据又一次不同步了，查看clickhouse-server.err.log,发现有如下报错

2023.08.16 11:19:00.782071 [ 8566 ] {} <Error> ConfigReloader: Error updating configuration from '/etc/clickhouse-server/config.xml' config.: Code: 999. Coordination::Exception: Connection loss, path: All connection tries failed while connecting to ZooKeeper