Updated on 2022-12-14 GMT+08:00

Synchronizing Kafka Data to ClickHouse

This section describes how to create a Kafka table to automatically synchronize Kafka data to the ClickHouse cluster.

Prerequisites

  • You have created a Kafka cluster. The Kafka client has been installed.
  • A ClickHouse cluster has been created. It is in the same VPC as the Kafka cluster and can communicate with each other.
  • The ClickHouse client has been installed.

Syntax of the Kafka Table

  • Syntax
    CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
    (
        name1 [type1] [DEFAULT|MATERIALIZED|ALIAS expr1],
        name2 [type2] [DEFAULT|MATERIALIZED|ALIAS expr2],
        ...
    ) ENGINE = Kafka()
    SETTINGS
        kafka_broker_list = 'host1:port1,host2:port2',
        kafka_topic_list = 'topic1,topic2,...',
        kafka_group_name = 'group_name',
        kafka_format = 'data_format';
        [kafka_row_delimiter = 'delimiter_symbol',]
        [kafka_schema = '',]
        [kafka_num_consumers = N]
  • Parameter description
    Table 1 Kafka table parameters

    Parameter

    Mandatory

    Description

    kafka_broker_list

    Yes

    A list of Kafka broker instances, separated by comma (,). For example, IP address 1 of the Kafka broker instance:9092,IP address 2 of the Kafka broker instance:9092,IP address 3 of the Kafka broker instance:9092.

    To obtain the IP address of the Kafka broker instance, perform the following steps:

    • For versions earlier than MRS 3.x, click the cluster name to go to the cluster details page and choose Components > Kafka. Click Instances to query the IP addresses of the Kafka instances.
      NOTE:

      If the Components tab is unavailable, complete IAM user synchronization first. (On the Dashboard page, click Synchronize on the right side of IAM User Sync to synchronize IAM users.)

    • For MRS 3.x or later, log in to FusionInsight Manager and choose Cluster > Name of the desired cluster > Services > Kafka. Click Instances to query the IP addresses of the Kafka instances.

    kafka_topic_list

    Yes

    A list of Kafka topics.

    kafka_group_name

    Yes

    A group of Kafka consumers, which can be customized.

    kafka_format

    Yes

    Kafka message format, for example, JSONEachRow, CSV, and XML.

    kafka_row_delimiter

    No

    Delimiter character, which ends a message.

    kafka_schema

    No

    Parameter that must be used if the format requires a schema definition.

    kafka_num_consumers

    No

    Number of consumers in per table. The default value is 1. If the throughput of a consumer is insufficient, more consumers are required. The total number of consumers cannot exceed the number of partitions in a topic because only one consumer can be allocated to each partition.

How to Synchronize Kafka Data to ClickHouse

  1. Switch to the Kafka client installation directory. For details, see Using the Kafka Client.

    1. Log in to the node where the Kafka client is installed as the Kafka client installation user.
    2. Run the following command to go to the client installation directory:

      cd /opt/client

    3. Run the following command to configure environment variables:

      source bigdata_env

    4. If Kerberos authentication is enabled for the current cluster, run the following command to authenticate the current user. If Kerberos authentication is disabled for the current cluster, skip this step.
      1. Run the following command first for an MRS 3.1.0 cluster:

        export CLICKHOUSE_SECURITY_ENABLED=true

      2. kinit Component service user

  2. Run the following command to create a Kafka topic. For details, see Managing Kafka Topics.

    kafka-topics.sh --topic kafkacktest2 --create --zookeeper IP address of the Zookeeper role instance:2181/kafka --partitions 2 --replication-factor 1

    • --topic is the name of the topic to be created, for example, kafkacktest2.
    • --zookeeper is the IP address of the node where the ZooKeeper role instances are located, which can be the IP address of any of the three role instances. You can obtain the IP address of the node by performing the following steps:
      • For versions earlier than MRS 3.x, click the cluster name to go to the cluster details page and choose Components > ZooKeeper > Instances. View the IP addresses of the ZooKeeper role instances.
      • For MRS 3.x or later, log in to FusionInsight Manager. For details, see Accessing FusionInsight Manager (MRS 3.x or Later). Choose Cluster > Name of the desired cluster > Services > ZooKeeper > Instance. View the IP addresses of the ZooKeeper role instances.
    • --partitions and --replication-factor are the topic partitions and topic backup replicas, respectively. The number of the two parameters cannot exceed the number of Kafka role instances.

  3. Log in to the ClickHouse client by referring to Using ClickHouse from Scratch.

    1. Run the following command to go to the client installation directory:

      cd /opt/Bigdata/client

    2. Run the following command to configure environment variables:

      source bigdata_env

    3. If Kerberos authentication is enabled for the current cluster, run the following command to authenticate the current user. The user must have the permission to create ClickHouse tables. Therefore, you need to bind the corresponding role to the user. For details, see ClickHouse User and Permission Management. If Kerberos authentication is disabled for the current cluster, skip this step.

      kinit Component service user

      Example: kinit clickhouseuser

    4. Run the following command to connect to the ClickHouse instance node to which data is to be imported:

      clickhouse client --host IP address of the ClickHouse instance --user Login username --password User password --port Port number of the ClickHouse instance --database Database name --multiline

  4. Create a Kafka table in ClickHouse by referring to Syntax of the Kafka Table. For example, the following table creation statement is used to create a Kafka table whose name is kafka_src_tbl3, topic name is kafkacktest2, and message format is JSONEachRow in the default database.

    create table kafka_src_tbl3 on cluster default_cluster 
    (id UInt32, age UInt32, msg String)  
    ENGINE=Kafka() 
    SETTINGS 
     kafka_broker_list='IP address 1 of the Kafka broker instance:9092,IP address 2 of the Kafka broker instance:9092,IP address 3 of the Kafka broker instance:9092',
     kafka_topic_list='kafkacktest2',
     kafka_group_name='cg12',
     kafka_format='JSONEachRow';

  5. Create a ClickHouse replicated table, for example, the ReplicatedMergeTree table named kafka_dest_tbl3.

    create table kafka_dest_tbl3 on cluster default_cluster 
    ( id UInt32, age UInt32, msg String )
    engine = ReplicatedMergeTree('/clickhouse/tables/{shard}/default/kafka_dest_tbl3', '{replica}')
    partition by age 
    order by id;

  6. Create a materialized view, which converts data in Kafka in the background and saves the data to the created ClickHouse table.

    create materialized view consumer3 on cluster default_cluster to kafka_dest_tbl3 as select * from kafka_src_tbl3;

  7. Perform 1 again to go to the Kafka client installation directory.
  8. Run the following command to send a message to the topic created in 2:

    kafka-console-producer.sh --broker-list IP address 1 of the kafka broker instance:9092,IP address 2 of the kafka broker instance:9092,IP address 3 of the kafka broker instance:9092 --topic kafkacktest2
    >{"id":31, "age":30, "msg":"31 years old"}
    >{"id":32, "age":30, "msg":"31 years old"}
    >{"id":33, "age":30, "msg":"31 years old"}
    >{"id":35, "age":30, "msg":"31 years old"}

  9. Use the ClickHouse client to log in to the ClickHouse instance node in 3 and query the ClickHouse table data, for example, to query the replicated table kafka_dest_tbl3. It shows that the data in the Kafka message has been synchronized to this table.

    select * from kafka_dest_tbl3;