Kafka容错性测试

青旅半醒 2023-06-14 04:10 16阅读 0赞

**1. 查看`topic`信息**(“副本”是为该分区复制日志的节点列表，无论它们是引导者还是当前处于活动状态，“ isr”是“同步”副本的集合。这是副本列表的子集，当前仍处于活动状态并追随领导者。)

[hadoop@hadoop000 tmp]$ kafka-topics.sh --describe --zookeeper hadoop000:2181 --topic my-replicated-topic
    Topic:my-replicated-topic	PartitionCount:1	ReplicationFactor:3	Configs:
    	Topic: my-replicated-topic	Partition: 0	Leader: 3	Replicas: 3,1,2	Isr: 1,2,3

**2. 查看三个broker的配置**

[hadoop@hadoop000 tmp]$ jps -m
    3761 ConsoleProducer --broker-list hadoop000:9093,hadoop000:9094,hadoop000:9095 --topic my-replicated-topic
    4178 Jps -m
    3717 ConsoleConsumer --zookeeper hadoop000:2181 --from-beginning --topic my-replicated-topic
    3637 ConsoleProducer --broker-list hadoop000:9092 --topic my-replicated-topic
    3175 QuorumPeerMain /home/hadoop/app/zookeeper-3.4.5-cdh5.7.0/bin/../conf/zoo.cfg
    3788 ConsoleConsumer --zookeeper hadoop000:2181 --topic my-replicated-topic
    3310 Kafka /home/hadoop/app/kafka_2.11-0.9.0.0/config/server-2.properties
    3230 Kafka /home/hadoop/app/kafka_2.11-0.9.0.0/config/server-1.properties
    3406 Kafka /home/hadoop/app/kafka_2.11-0.9.0.0/config/server-3.properties

**3. 杀死broker2模拟容错**

[hadoop@hadoop000 tmp]$ kill -9 3310
    
    #领导权已切换到关注者之一，并且节点2不再位于同步副本集中：
    [hadoop@hadoop000 tmp]$ jps -m
    4208 Jps -m
    3761 ConsoleProducer --broker-list hadoop000:9093,hadoop000:9094,hadoop000:9095 --topic my-replicated-topic
    3717 ConsoleConsumer --zookeeper hadoop000:2181 --from-beginning --topic my-replicated-topic
    3637 ConsoleProducer --broker-list hadoop000:9092 --topic my-replicated-topic
    3175 QuorumPeerMain /home/hadoop/app/zookeeper-3.4.5-cdh5.7.0/bin/../conf/zoo.cfg
    3788 ConsoleConsumer --zookeeper hadoop000:2181 --topic my-replicated-topic
    3230 Kafka /home/hadoop/app/kafka_2.11-0.9.0.0/config/server-1.properties
    3406 Kafka /home/hadoop/app/kafka_2.11-0.9.0.0/config/server-3.properties
    
    [hadoop@hadoop000 tmp]$ kafka-topics.sh --describe --zookeeper hadoop000:2181 --topic my-replicated-topic
    Topic:my-replicated-topic	PartitionCount:1	ReplicationFactor:3	Configs:
    	Topic: my-replicated-topic	Partition: 0	Leader: 3	Replicas: 3,1,2	Isr: 1,3

**4. 但是，即使最初进行写操作的领导者已经下线，消息仍然可以使用**

[hadoop@hadoop000 ~]$ kafka-console-consumer.sh --zookeeper hadoop000:2181 --from-beginning --topic my-replicated-topic
    hello
    
    [hadoop@hadoop000 ~]$ kafka-console-producer.sh --broker-list hadoop000:9093,hadoop000:9094,hadoop000:9095 --topic my-replicated-topic
    hello

**至此说明 `Kafka`的容错性是完全有保障的。**