以下操作可以在mini01、mini02、mini03任意一台操作即可。
1. kafka通过网页管理
参考博文：kafka集群管理工具kafka-manager部署安装
2. 创建topic
1 # 参数说明 --replication-factor 2 表示有2个副本2 # --partitions 4 表示有4个分区3 [yun@mini01 ~]$ kafka-topics.sh --create --zookeeper mini01:2181 --replication-factor 2 --partitions 4 --topic test4 Created topic "test".5 [yun@mini01 ~]$ kafka-topics.sh --create --zookeeper mini01:2181 --replication-factor 3 --partitions 4 --topic zhang6 Created topic "zhang".7 [yun@mini01 ~]$ kafka-topics.sh --list --zookeeper mini01:2181 # 再次查看8 zhang9 test2.1. 各主机信息查看
mini01
1 [yun@mini01 logs]$ pwd2 /app/kafka/logs3 [yun@mini01 logs]$ ll 4 total 1605 ………………6 drwxrwxr-x 2 yun yun 141 Sep 15 18:53 test-17 drwxrwxr-x 2 yun yun 141 Sep 15 18:53 test-28 drwxrwxr-x 2 yun yun 141 Sep 15 18:53 test-3mini02
1 [yun@mini02 logs]$ pwd2 /app/kafka/logs3 [yun@mini02 logs]$ ll4 total 2605 ………………6 drwxrwxr-x 2 yun yun 141 Sep 15 18:53 test-07 drwxrwxr-x 2 yun yun 141 Sep 15 18:53 test-2mini03
1 [yun@mini03 logs]$ pwd2 /app/kafka/logs3 [yun@mini03 logs]$ ll4 total 1325 ………………6 drwxrwxr-x 2 yun yun 141 Sep 15 18:53 test-07 drwxrwxr-x 2 yun yun 141 Sep 15 18:53 test-18 drwxrwxr-x 2 yun yun 141 Sep 15 18:53 test-33. 修改topic
3.1. 增加分区数
注意：分区数不能减少
Kafka目前不支持减少主题的分区数量。
1 [yun@mini01 ~]$ kafka-topics.sh --list --zookeeper mini01:2181 2 __consumer_offsets 3 test 4 test01 5 test02 6 test03 7 test04 8 zhang 9 [yun@mini01 ~]$ kafka-topics.sh --describe --zookeeper mini01:2181 --topic test0110 Topic:test01 PartitionCount:5 ReplicationFactor:1 Configs:11 Topic: test01 Partition: 0 Leader: 0 Replicas: 0 Isr: 012 Topic: test01 Partition: 1 Leader: 1 Replicas: 1 Isr: 113 Topic: test01 Partition: 2 Leader: 2 Replicas: 2 Isr: 214 Topic: test01 Partition: 3 Leader: 0 Replicas: 0 Isr: 015 Topic: test01 Partition: 4 Leader: 1 Replicas: 1 Isr: 116 [yun@mini01 ~]$ kafka-topics.sh --alter --zookeeper mini01:2181 --partitions 2 --topic test01 # 失败，分区数不能减少 17 WARNING: If partitions are increased for a topic that has a key, the partition logic or ordering of the messages will be affected18 Error while executing topic command : The number of partitions for a topic can only be increased. Topic test01 currently has 5 partitions, 2 would not be an increase.19 [2018-09-16 09:12:40,034] ERROR org.apache.kafka.common.errors.InvalidPartitionsException: The number of partitions for a topic can only be increased. Topic test01 currently has 5 partitions, 2 would not be an increase.20 (kafka.admin.TopicCommand$)21 [yun@mini01 ~]$ kafka-topics.sh --alter --zookeeper mini01:2181 --partitions 7 --topic test01 # 增加分区数 22 WARNING: If partitions are increased for a topic that has a key, the partition logic or ordering of the messages will be affected23 Adding partitions succeeded!24 [yun@mini01 ~]$ kafka-topics.sh --describe --zookeeper mini01:2181 --topic test0125 Topic:test01 PartitionCount:7 ReplicationFactor:1 Configs:26 Topic: test01 Partition: 0 Leader: 0 Replicas: 0 Isr: 027 Topic: test01 Partition: 1 Leader: 1 Replicas: 1 Isr: 128 Topic: test01 Partition: 2 Leader: 2 Replicas: 2 Isr: 229 Topic: test01 Partition: 3 Leader: 0 Replicas: 0 Isr: 030 Topic: test01 Partition: 4 Leader: 1 Replicas: 1 Isr: 131 Topic: test01 Partition: 5 Leader: 2 Replicas: 2 Isr: 232 Topic: test01 Partition: 6 Leader: 0 Replicas: 0 Isr: 04. 删除topic
1 # server.properties中设置delete.topic.enable=true 【当前版本默认就是true】否则只是标记删除或者直接重启2 [yun@mini01 ~]$ kafka-topics.sh --delete --zookeeper mini01:2181 --topic test3 Topic test is marked for deletion.4 Note: This will have no impact if delete.topic.enable is not set to true.5 [yun@mini01 ~]$ kafka-topics.sh --list --zookeeper mini01:2181 # 再次查看 只有 zhang，则表示真的删除了6 zhang5. 查看所有topic
1 [yun@mini01 ~]$ kafka-topics.sh --list --zookeeper mini01:2181 2 __consumer_offsets3 test4 zhang6. 查看某个Topic的详情
1 [yun@mini01 ~]$ kafka-topics.sh --describe --zookeeper mini01:2181 --topic zhang 2 Topic:zhang PartitionCount:4 ReplicationFactor:3 Configs:3 Topic: zhang Partition: 0 Leader: 1 Replicas: 1,2,0 Isr: 1,2,04 Topic: zhang Partition: 1 Leader: 2 Replicas: 2,0,1 Isr: 2,0,15 Topic: zhang Partition: 2 Leader: 0 Replicas: 0,1,2 Isr: 0,1,26 Topic: zhang Partition: 3 Leader: 1 Replicas: 1,0,2 Isr: 1,0,27. 通过shell命令生产消息
7.1. 输入单条数据
1 [yun@mini01 ~]$ kafka-console-producer.sh --broker-list mini01:9092 --topic zhang 2 >111 3 >222 4 >333 5 >444 6 >555 7 >666 8 >777 9 >88810 >9997.2. 批量导入数据
1 [yun@mini01 zhangliang]$ kafka-console-producer.sh --broker-list mini01:9092 --topic liang < 001.info8. 通过shell命令消费消息
1 # --from-beginning 从最开始读取 2 # kafka-console-consumer.sh --zookeeper mini01:2181 --from-beginning --topic zhang # 老版本 3 [yun@mini01 ~]$ kafka-console-consumer.sh --bootstrap-server mini01:9092 --from-beginning --topic zhang 4 111 5 555 6 999 7 333 8 777 9 44410 88811 22212 6669. 消费组消费
9.1. 创建topic
1 [yun@mini01 ~]$ kafka-topics.sh --create --zookeeper mini01:2181 --replication-factor 1 --partitions 4 --topic order 2 Created topic "order". 3 [yun@mini01 ~]$ kafka-topics.sh --list --zookeeper mini01:2181 # 查看所有topic列表 4 __consumer_offsets 5 order 6 test 7 zhang 8 [yun@mini01 ~]$ kafka-topics.sh --describe --zookeeper mini01:2181 --topic order # 查看topic详情 9 Topic:order PartitionCount:4 ReplicationFactor:1 Configs:10 Topic: order Partition: 0 Leader: 0 Replicas: 0 Isr: 011 Topic: order Partition: 1 Leader: 1 Replicas: 1 Isr: 112 Topic: order Partition: 2 Leader: 2 Replicas: 2 Isr: 213 Topic: order Partition: 3 Leader: 0 Replicas: 0 Isr: 09.2. 生产消息
1 [yun@mini01 ~]$ kafka-console-producer.sh --broker-list mini01:9092 --topic order2 >1113 >2224 >3335 >4446 >5559.3. 消费组消费消息
mini02机器上运行
1 # --group 指定组2 [yun@mini02 ~]$ kafka-console-consumer.sh --bootstrap-server mini01:9092 --topic order --group order-groupmini03机器上运行
1 # --group 指定组2 [yun@mini03 ~]$ kafka-console-consumer.sh --bootstrap-server mini01:9092 --topic order --group order-group3 4 # 新开一个窗口执行5 [yun@mini03 ~]$ kafka-console-consumer.sh --bootstrap-server mini01:9092 --topic order --group order-group表示order-group消费组有3个消费者，消费topic order的信息。
9.4. 消费组消费位置信息查看
1 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --describe --group order-group # 查看消费情况 2 3 TOPIC PARTITION CURRENT-OFFSET LOG-END-OFFSET LAG CONSUMER-ID HOST CLIENT-ID 4 order 0 4 4 0 consumer-1-2e9805db-e021-4595-8c62-92f8691fbf20 /172.16.1.13 consumer-1 5 order 1 5 5 0 consumer-1-2e9805db-e021-4595-8c62-92f8691fbf20 /172.16.1.13 consumer-1 6 order 2 5 5 0 consumer-1-9e65dcfb-246f-4043-aaf7-3ee83532237f /172.16.1.13 consumer-1 7 order 3 4 4 0 consumer-1-ee17939d-1ffe-42c7-8261-b19be8acea43 /172.16.1.12 consumer-1 8 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --describe --group order-group --members --verbose 9 10 CONSUMER-ID HOST CLIENT-ID #PARTITIONS ASSIGNMENT11 consumer-1-9e65dcfb-246f-4043-aaf7-3ee83532237f /172.16.1.13 consumer-1 1 order(2)12 consumer-1-2e9805db-e021-4595-8c62-92f8691fbf20 /172.16.1.13 consumer-1 2 order(0,1)13 consumer-1-ee17939d-1ffe-42c7-8261-b19be8acea43 /172.16.1.12 consumer-1 1 order(3)10. 管理消费组
10.1. 查看所有消费组
1 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --list2 console-consumer-267273 console-consumer-929844 console-consumer-607555 console-consumer-116616 console-consumer-317137 console-consumer-202448 console-consumer-6573310.2. 查看消费组消费情况【消费位置】
1 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --describe --group console-consumer-26727 2 Consumer group 'console-consumer-26727' has no active members. 3 4 TOPIC PARTITION CURRENT-OFFSET LOG-END-OFFSET LAG CONSUMER-ID HOST CLIENT-ID 5 zhang 3 11 11 0 - - - 6 zhang 0 9 9 0 - - - 7 zhang 2 8 8 0 - - - 8 zhang 1 11 11 0 - - - 9 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --describe --group console-consumer-65733 10 11 TOPIC PARTITION CURRENT-OFFSET LOG-END-OFFSET LAG CONSUMER-ID HOST CLIENT-ID12 zhang 0 11 11 0 consumer-1-17c812f0-116b-42a9-88d8-90d1a85949e1 /172.16.1.13 consumer-113 zhang 1 12 12 0 consumer-1-17c812f0-116b-42a9-88d8-90d1a85949e1 /172.16.1.13 consumer-114 zhang 2 10 10 0 consumer-1-17c812f0-116b-42a9-88d8-90d1a85949e1 /172.16.1.13 consumer-115 zhang 3 12 12 0 consumer-1-17c812f0-116b-42a9-88d8-90d1a85949e1 /172.16.1.13 consumer-1--members
1 # --members 此选项提供使用者组中所有活动成员的列表。2 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --describe --group console-consumer-65733 --members 3 4 CONSUMER-ID HOST CLIENT-ID #PARTITIONS 5 consumer-1-17c812f0-116b-42a9-88d8-90d1a85949e1 /172.16.1.13 consumer-1 4 --verbose
1 # --verbose 这个选项还提供了分配给每个成员的分区。2 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --describe --group console-consumer-65733 --members --verbose3 4 CONSUMER-ID HOST CLIENT-ID #PARTITIONS ASSIGNMENT5 consumer-1-17c812f0-116b-42a9-88d8-90d1a85949e1 /172.16.1.13 consumer-1 4 zhang(0,1,2,3)--state
1 # --state 这个选项提供了有用的组级信息。2 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --describe --group console-consumer-65733 --state3 4 COORDINATOR (ID) ASSIGNMENT-STRATEGY STATE #MEMBERS5 mini01:9092 (0) range Stable 110.3. 删除一个或多个用户组
1 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --list 2 console-consumer-3826 3 console-consumer-92984 4 console-consumer-60755 5 console-consumer-11661 6 console-consumer-31713 7 console-consumer-20244 8 console-consumer-65733 9 # 删除一个或多个组10 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --delete --group console-consumer-11661 --group console-consumer-3171311 Deletion of requested consumer groups ('console-consumer-31713', 'console-consumer-11661') was successful.12 [yun@mini01 ~]$ kafka-consumer-groups.sh --bootstrap-server mini01:9092 --list13 console-consumer-382614 console-consumer-9298415 console-consumer-6075516 console-consumer-2024417 console-consumer-6573311. 增加副本因子
1 [yun@mini01 kafka_20180916]$ kafka-topics.sh --create --zookeeper mini01:2181 --replication-factor 1 --partitions 4 --topic order2 Created topic "order".3 [yun@mini01 kafka_20180916]$ kafka-topics.sh --describe --zookeeper mini01:2181 --topic order4 Topic:order PartitionCount:4 ReplicationFactor:1 Configs:5 Topic: order Partition: 0 Leader: 0 Replicas: 0 Isr: 06 Topic: order Partition: 1 Leader: 1 Replicas: 1 Isr: 17 Topic: order Partition: 2 Leader: 2 Replicas: 2 Isr: 28 Topic: order Partition: 3 Leader: 0 Replicas: 0 Isr: 0要求：topic order 的副本数由1变为2， 之前每个分区在哪台机器上在上面已给出。
说明：part 0分布在群集0,1； part 1分布在集群1,2；part 2 分布在集群2,0；part 3分布在集群0,1。
11.1. 创建一个重新调整计划文件
1 [yun@mini01 kafka_20180916]$ cat increase-replication-factor.json 2 { 3 "version":1, 4 "partitions":[ 5 {"topic": "order","partition": 0,"replicas": [0,1]}, 6 {"topic": "order","partition": 1,"replicas": [1,2]}, 7 {"topic": "order","partition": 2,"replicas": [2,0]}, 8 {"topic": "order","partition": 3,"replicas": [0,1]} 9 ]10 }11.2. 语句执行
1 [yun@mini01 kafka_20180916]$ kafka-reassign-partitions.sh --zookeeper mini01:2181 --reassignment-json-file increase-replication-factor.json --execute 2 Current partition replica assignment3 4 {"version":1,"partitions":[{"topic":"order","partition":2,"replicas":[2],"log_dirs":["any"]},{"topic":"order","partition":1,"replicas":[1],"log_dirs":["any"]},{"topic":"order","partition":3,"replicas":[0],"log_dirs":["any"]},{"topic":"order","partition":0,"replicas":[0],"log_dirs":["any"]}]}5 6 Save this to use as the --reassignment-json-file option during rollback7 Successfully started reassignment of partitions.11.3. 查看是否执行成功
1 [yun@mini01 kafka_20180916]$ kafka-reassign-partitions.sh --zookeeper mini01:2181 --reassignment-json-file increase-replication-factor.json --verify2 Status of partition reassignment: 3 Reassignment of partition order-0 completed successfully4 Reassignment of partition order-1 completed successfully5 Reassignment of partition order-2 completed successfully6 Reassignment of partition order-3 completed successfully11.4. 再次查看该topic详情
1 [yun@mini01 kafka_20180916]$ kafka-topics.sh --describe --zookeeper mini01:2181 --topic order # 由下可见分配成功2 Topic:order PartitionCount:4 ReplicationFactor:2 Configs:3 Topic: order Partition: 0 Leader: 0 Replicas: 0,1 Isr: 0,14 Topic: order Partition: 1 Leader: 1 Replicas: 1,2 Isr: 1,25 Topic: order Partition: 2 Leader: 2 Replicas: 2,0 Isr: 2,06 Topic: order Partition: 3 Leader: 0 Replicas: 0,1 Isr: 0,112. 创建partitions时在broker的分配机制
1 kafka-topics.sh --create --zookeeper mini01:2181 --replication-factor 1 --partitions 5 --topic test012 kafka-topics.sh --create --zookeeper mini01:2181 --replication-factor 1 --partitions 11 --topic test02注意在各机器上partitions的分布
1 mini01 2 test01-0 3 test01-3 4 test02-2 5 test02-5 6 test02-8 7 8 mini02 9 test01-110 test01-411 test02-012 test02-313 test02-614 test02-915 16 mini0317 test01-218 test02-119 test02-1020 test02-421 test02-7版权声明：本文为weixin_30583711原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接和本声明。