图文精华

About云-梭伦科技»专题 › 技术学习(版主发帖区) › 大数据学习 › Kafka › 实操演练› kafka2.9.2的分布式集群安装和demo(java api)测试
分享

kafka2.9.2的分布式集群安装和demo(java api)测试

pig2 2014-8-26 02:29:00 发表于 实操演练 [显示全部楼层] 只看大图 回帖奖励 阅读模式 关闭右栏 11 152370
About云VIP帮助找工作和提升
本帖最后由 pig2 于 2014-8-26 02:35 编辑
问题导读
1、什么是kafka?
2、kafka的官方网站在哪里？
3、在哪里下载？需要哪些组件的支持？
4、如何安装？






　　一、什么是kafka?
　　kafka是LinkedIn开发并开源的一个分布式MQ系统，现在是Apache的一个孵化项目。在它的主页描述kafka为一个高吞吐量的分布式（能将消息分散到不同的节点上）MQ。Kafka仅仅由7000行Scala编写，据了解，Kafka每秒可以生产约25万消息（50 MB），每秒处理55万消息（110 MB）。
　　kafka目前支持多种客户端语言：java，python，c++，php等等。
　　kafka集群的简要图解如下，producer写入消息，consumer读取消息：
145455f6bg6fggektmo4ko.png


　　kafka设计目标
  • 高吞吐量是其核心设计之一。
  • 数据磁盘持久化：消息不在内存中cache，直接写入到磁盘，充分利用磁盘的顺序读写性能。
  • zero-copy：减少IO操作步骤。
  • 支持数据批量发送和拉取。
  • 支持数据压缩。
  • Topic划分为多个partition，提高并行处理能力。


　　kafka名词解释和工作方式：
  • Producer ：消息生产者，就是向kafka broker发消息的客户端。
  • Consumer ：消息消费者，向kafka broker取消息的客户端
  • Topic ：可以理解为一个队列。
  • Consumer Group （CG）：这是kafka用来实现一个topic消息的广播（发给所有的consumer）和单播（发给任意一个consumer）的手段。一个topic可以有多个CG。topic的消息会复制（不是真的复制，是概念上的）到所有的CG，但每个CG只会把消息发给该CG中的一个consumer。如果需要实现广播，只要每个consumer有一个独立的CG就可以了。要实现单播只要所有的consumer在同一个CG。用CG还可以将consumer进行自由的分组而不需要多次发送消息到不同的topic。
  • Broker ：一台kafka服务器就是一个broker。一个集群由多个broker组成。一个broker可以容纳多个topic。
  • Partition：为了实现扩展性，一个非常大的topic可以分布到多个broker（即服务器）上，一个topic可以分为多个partition，每个partition是一个有序的队列。partition中的每条消息都会被分配一个有序的id（offset）。kafka只保证按一个partition中的顺序将消息发给consumer，不保证一个topic的整体（多个partition间）的顺序。
  • Offset：kafka的存储文件都是按照offset.kafka来命名，用offset做名字的好处是方便查找。例如你想找位于2049的位置，只要找到2048.kafka的文件即可。当然the first offset就是00000000000.kafka


　　kafak系统扩展性：
  • kafka使用zookeeper来实现动态的集群扩展，不需要更改客户端（producer和consumer）的配置。broker会在zookeeper注册并保持相关的元数据（topic，partition信息等）更新。
  • 而客户端会在zookeeper上注册相关的watcher。一旦zookeeper发生变化，客户端能及时感知并作出相应调整。这样就保证了添加或去除broker时，各broker间仍能自动实现负载均衡。


　　kafak和zookeeper的关系：
  • Producer端使用zookeeper用来"发现"broker列表,以及和Topic下每个partition leader建立socket连接并发送消息.
  • Broker端使用zookeeper用来注册broker信息,已经监测partition leader存活性.
  • Consumer端使用zookeeper用来注册consumer信息,其中包括consumer消费的partition列表等,同时也用来发现broker列表,并和partition leader建立socket连接,并获取消息.



　　二、kafka的官方网站在哪里？
　　http://kafka.apache.org/

　　三、在哪里下载？需要哪些组件的支持？
　　kafka2.9.2在下面的地址可以下载：
　　https://www.apache.org/dyn/closer.cgi?path=/kafka/0.8.1.1/kafka_2.9.2-0.8.1.1.tgz

　四、如何安装？
　　1、解压kafka_2.9.2-0.8.1.1.tgz，本文中解压到/home/hadoop目录下

  1. root@m1:/home/hadoop/kafka_2.9.2-0.8.1.1# pwd

  2. /home/hadoop/kafka_2.9.2-0.8.1.1
复制代码

2、修改server.properties配置文件。这里使用zookeeper的部分，见下方第123行：


root@m1:/home/hadoop/kafka_2.9.2-0.8.1.1# cat config/server.properties
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements.  See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License.  You may obtain a copy of the License at
#
#    http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# see kafka.server.KafkaConfig for additional details and defaults

############################# Server Basics #############################

# The id of the broker. This must be set to a unique integer for each broker.
#整数，建议根据ip区分,这里我是使用zookeeper中的id来设置
broker.id=1

############################# Socket Server Settings #############################

# The port the socket server listens on
#broker用于接收producer消息的端口
port=9092
#port=44444

# Hostname the broker will bind to. If not set, the server will bind to all interfaces
#broker的hostname
host.name=m1

# Hostname the broker will advertise to producers and consumers. If not set, it uses the
# value for "host.name" if configured.  Otherwise, it will use the value returned from
# java.net.InetAddress.getCanonicalHostName().
#这个是配置PRODUCER/CONSUMER连上来的时候使用的地址
advertised.host.name=m1

# The port to publish to ZooKeeper for clients to use. If this is not set,
# it will publish the same port that the broker binds to.
#advertised.port=<port accessible by clients>

# The number of threads handling network requests
num.network.threads=2
  
# The number of threads doing disk I/O
num.io.threads=8

# The send buffer (SO_SNDBUF) used by the socket server
socket.send.buffer.bytes=1048576

# The receive buffer (SO_RCVBUF) used by the socket server
socket.receive.buffer.bytes=1048576

# The maximum size of a request that the socket server will accept (protection against OOM)
socket.request.max.bytes=104857600


############################# Log Basics #############################

# A comma seperated list of directories under which to store log files
#kafka存放消息文件的路径
log.dirs=/home/hadoop/kafka_2.9.2-0.8.1.1/kafka-logs

# The default number of log partitions per topic. More partitions allow greater
# parallelism for consumption, but this will also result in more files across
# the brokers.
#topic的默认分区数
num.partitions=2

############################# Log Flush Policy #############################

# Messages are immediately written to the filesystem but by default we only fsync() to sync
# the OS cache lazily. The following configurations control the flush of data to disk.
# There are a few important trade-offs here:
#    1. Durability: Unflushed data may be lost if you are not using replication.
#    2. Latency: Very large flush intervals may lead to latency spikes when the flush does occur as there will be a lot of data to flush.
#    3. Throughput: The flush is generally the most expensive operation, and a small flush interval may lead to exceessive seeks.
# The settings below allow one to configure the flush policy to flush data after a period of time or
# every N messages (or both). This can be done globally and overridden on a per-topic basis.

# The number of messages to accept before forcing a flush of data to disk
#log.flush.interval.messages=10000

# The maximum amount of time a message can sit in a log before we force a flush
#log.flush.interval.ms=1000

############################# Log Retention Policy #############################

# The following configurations control the disposal of log segments. The policy can
# be set to delete segments after a period of time, or after a given size has accumulated.
# A segment will be deleted whenever *either* of these criteria are met. Deletion always happens
# from the end of the log.

# The minimum age of a log file to be eligible for deletion
#kafka接收日志的存储目录(目前我们保存7天数据log.retention.hours=168)
log.retention.hours=168

# A size-based retention policy for logs. Segments are pruned from the log as long as the remaining
# segments don't drop below log.retention.bytes.
#log.retention.bytes=1073741824

# The maximum size of a log segment file. When this size is reached a new log segment will be created.
log.segment.bytes=536870912

# The interval at which log segments are checked to see if they can be deleted according
# to the retention policies
log.retention.check.interval.ms=60000

# By default the log cleaner is disabled and the log retention policy will default to just delete segments after their retention expires.
# If log.cleaner.enable=true is set the cleaner will be enabled and individual logs can then be marked for log compaction.
log.cleaner.enable=false

############################# Zookeeper #############################

# Zookeeper connection string (see zookeeper docs for details).
# This is a comma separated host:port pairs, each corresponding to a zk
# server. e.g. "127.0.0.1:3000,127.0.0.1:3001,127.0.0.1:3002".
# You can also append an optional chroot string to the urls to specify the
# root directory for all kafka znodes.
zookeeper.connect=m1:2181,m2:2181,s1:2181,s2:2181

# Timeout in ms for connecting to zookeeper
zookeeper.connection.timeout.ms=1000000

3、启动zookeeper和kafka
　　　　1)zookeeper的启动
　　　　启动后可以用以下命令在每台机器上查看状态：
  1. root@m1:/home/hadoop# /home/hadoop/zookeeper-3.4.5/bin/zkServer.sh status

  2. JMX enabled by default

  3. Using config: /home/hadoop/zookeeper-3.4.5/bin/../conf/zoo.cfg

  4. Mode: leader
复制代码

　　　　2)在m1,m2,s1,s2的机器上启动kafka，在这之前请先将m1上的kafka复制到另外三台机器上，复制后，记得更改server.properties配置文件中的host名称为当前所在机器。以下代码是在m1上执行后的效果：
  1. root@m1:/home/hadoop# /home/hadoop/kafka_2.9.2-0.8.1.1/bin/kafka-server-start.sh /home/hadoop/kafka_2.9.2-0.8.1.1/config/server.properties &

  2. [1] 31823

  3. root@m1:/home/hadoop# [2014-08-05 10:03:11,210] INFO Verifying properties (kafka.utils.VerifiableProperties)

  4. [2014-08-05 10:03:11,261] INFO Property advertised.host.name is overridden to m1 (kafka.utils.VerifiableProperties)

  5. [2014-08-05 10:03:11,261] INFO Property broker.id is overridden to 1 (kafka.utils.VerifiableProperties)

  6. [2014-08-05 10:03:11,264] INFO Property host.name is overridden to m1 (kafka.utils.VerifiableProperties)

  7. [2014-08-05 10:03:11,264] INFO Property log.cleaner.enable is overridden to false (kafka.utils.VerifiableProperties)

  8. [2014-08-05 10:03:11,264] INFO Property log.dirs is overridden to /home/hadoop/kafka_2.9.2-0.8.1.1/kafka-logs (kafka.utils.VerifiableProperties)

  9. [2014-08-05 10:03:11,265] INFO Property log.retention.check.interval.ms is overridden to 60000 (kafka.utils.VerifiableProperties)

  10. [2014-08-05 10:03:11,265] INFO Property log.retention.hours is overridden to 168 (kafka.utils.VerifiableProperties)

  11. [2014-08-05 10:03:11,265] INFO Property log.segment.bytes is overridden to 536870912 (kafka.utils.VerifiableProperties)

  12. [2014-08-05 10:03:11,265] INFO Property num.io.threads is overridden to 8 (kafka.utils.VerifiableProperties)

  13. [2014-08-05 10:03:11,266] INFO Property num.network.threads is overridden to 2 (kafka.utils.VerifiableProperties)

  14. [2014-08-05 10:03:11,266] INFO Property num.partitions is overridden to 2 (kafka.utils.VerifiableProperties)

  15. [2014-08-05 10:03:11,267] INFO Property port is overridden to 9092 (kafka.utils.VerifiableProperties)

  16. [2014-08-05 10:03:11,267] INFO Property socket.receive.buffer.bytes is overridden to 1048576 (kafka.utils.VerifiableProperties)

  17. [2014-08-05 10:03:11,268] INFO Property socket.request.max.bytes is overridden to 104857600 (kafka.utils.VerifiableProperties)

  18. [2014-08-05 10:03:11,268] INFO Property socket.send.buffer.bytes is overridden to 1048576 (kafka.utils.VerifiableProperties)

  19. [2014-08-05 10:03:11,268] INFO Property zookeeper.connect is overridden to m1:2181,m2:2181,s1:2181,s2:2181 (kafka.utils.VerifiableProperties)

  20. [2014-08-05 10:03:11,269] INFO Property zookeeper.connection.timeout.ms is overridden to 1000000 (kafka.utils.VerifiableProperties)

  21. [2014-08-05 10:03:11,302] INFO [Kafka Server 1], starting (kafka.server.KafkaServer)

  22. [2014-08-05 10:03:11,303] INFO [Kafka Server 1], Connecting to zookeeper on m1:2181,m2:2181,s1:2181,s2:2181 (kafka.server.KafkaServer)

  23. [2014-08-05 10:03:11,335] INFO Starting ZkClient event thread. (org.I0Itec.zkclient.ZkEventThread)

  24. [2014-08-05 10:03:11,348] INFO Client environment:zookeeper.version=3.3.3-1203054, built on 11/17/2011 05:47 GMT (org.apache.zookeeper.ZooKeeper)

  25. [2014-08-05 10:03:11,348] INFO Client environment:host.name=m1 (org.apache.zookeeper.ZooKeeper)

  26. [2014-08-05 10:03:11,349] INFO Client environment:java.version=1.7.0_65 (org.apache.zookeeper.ZooKeeper)

  27. [2014-08-05 10:03:11,349] INFO Client environment:java.vendor=Oracle Corporation (org.apache.zookeeper.ZooKeeper)

  28. [2014-08-05 10:03:11,349] INFO Client environment:java.home=/usr/lib/jvm/java-7-oracle/jre (org.apache.zookeeper.ZooKeeper)

  29. [2014-08-05 10:03:11,349] INFO Client environment:java.class.path=.:/usr/lib/jvm/java-7-oracle/lib/tools.jar:/usr/lib/jvm/java-7-oracle/lib/dt.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../core/build/dependant-libs-2.8.0/*.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../perf/build/libs//kafka-perf_2.8.0*.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../clients/build/libs//kafka-clients*.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../examples/build/libs//kafka-examples*.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../contrib/hadoop-consumer/build/libs//kafka-hadoop-consumer*.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../contrib/hadoop-producer/build/libs//kafka-hadoop-producer*.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/jopt-simple-3.2.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/kafka_2.9.2-0.8.1.1.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/kafka_2.9.2-0.8.1.1-javadoc.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/kafka_2.9.2-0.8.1.1-scaladoc.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/kafka_2.9.2-0.8.1.1-sources.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/log4j-1.2.15.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/metrics-core-2.2.0.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/scala-library-2.9.2.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/slf4j-api-1.7.2.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/snappy-java-1.0.5.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/zkclient-0.3.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../libs/zookeeper-3.3.4.jar:/home/hadoop/kafka_2.9.2-0.8.1.1/bin/../core/build/libs/kafka_2.8.0*.jar (org.apache.zookeeper.ZooKeeper)

  30. [2014-08-05 10:03:11,350] INFO Client environment:java.library.path=:/usr/local/lib:/usr/java/packages/lib/amd64:/usr/lib64:/lib64:/lib:/usr/lib (org.apache.zookeeper.ZooKeeper)

  31. [2014-08-05 10:03:11,350] INFO Client environment:java.io.tmpdir=/tmp (org.apache.zookeeper.ZooKeeper)

  32. [2014-08-05 10:03:11,350] INFO Client environment:java.compiler=<NA> (org.apache.zookeeper.ZooKeeper)

  33. [2014-08-05 10:03:11,350] INFO Client environment:os.name=Linux (org.apache.zookeeper.ZooKeeper)

  34. [2014-08-05 10:03:11,350] INFO Client environment:os.arch=amd64 (org.apache.zookeeper.ZooKeeper)

  35. [2014-08-05 10:03:11,351] INFO Client environment:os.version=3.11.0-15-generic (org.apache.zookeeper.ZooKeeper)

  36. [2014-08-05 10:03:11,351] INFO Client environment:user.name=root (org.apache.zookeeper.ZooKeeper)

  37. [2014-08-05 10:03:11,351] INFO Client environment:user.home=/root (org.apache.zookeeper.ZooKeeper)

  38. [2014-08-05 10:03:11,351] INFO Client environment:user.dir=/home/hadoop (org.apache.zookeeper.ZooKeeper)

  39. [2014-08-05 10:03:11,352] INFO Initiating client connection, connectString=m1:2181,m2:2181,s1:2181,s2:2181 sessionTimeout=6000 watcher=org.I0Itec.zkclient.ZkClient@51f782b8 (org.apache.zookeeper.ZooKeeper)

  40. [2014-08-05 10:03:11,380] INFO Opening socket connection to server m2/192.168.1.51:2181 (org.apache.zookeeper.ClientCnxn)

  41. [2014-08-05 10:03:11,386] INFO Socket connection established to m2/192.168.1.51:2181, initiating session (org.apache.zookeeper.ClientCnxn)

  42. [2014-08-05 10:03:11,398] INFO Session establishment complete on server m2/192.168.1.51:2181, sessionid = 0x247a3e09b460000, negotiated timeout = 6000 (org.apache.zookeeper.ClientCnxn)

  43. [2014-08-05 10:03:11,400] INFO zookeeper state changed (SyncConnected) (org.I0Itec.zkclient.ZkClient)

  44. [2014-08-05 10:03:11,652] INFO Loading log 'test-1' (kafka.log.LogManager)

  45. [2014-08-05 10:03:11,681] INFO Recovering unflushed segment 0 in log test-1. (kafka.log.Log)

  46. [2014-08-05 10:03:11,711] INFO Completed load of log test-1 with log end offset 137 (kafka.log.Log)

  47. SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder".

  48. SLF4J: Defaulting to no-operation (NOP) logger implementation

  49. SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details.

  50. [2014-08-05 10:03:11,747] INFO Loading log 'idoall.org-0' (kafka.log.LogManager)

  51. [2014-08-05 10:03:11,748] INFO Recovering unflushed segment 0 in log idoall.org-0. (kafka.log.Log)

  52. [2014-08-05 10:03:11,754] INFO Completed load of log idoall.org-0 with log end offset 5 (kafka.log.Log)

  53. [2014-08-05 10:03:11,760] INFO Loading log 'test-0' (kafka.log.LogManager)

  54. [2014-08-05 10:03:11,765] INFO Recovering unflushed segment 0 in log test-0. (kafka.log.Log)

  55. [2014-08-05 10:03:11,777] INFO Completed load of log test-0 with log end offset 151 (kafka.log.Log)

  56. [2014-08-05 10:03:11,779] INFO Starting log cleanup with a period of 60000 ms. (kafka.log.LogManager)

  57. [2014-08-05 10:03:11,782] INFO Starting log flusher with a default period of 9223372036854775807 ms. (kafka.log.LogManager)

  58. [2014-08-05 10:03:11,800] INFO Awaiting socket connections on m1:9092. (kafka.network.Acceptor)

  59. [2014-08-05 10:03:11,802] INFO [Socket Server on Broker 1], Started (kafka.network.SocketServer)

  60. [2014-08-05 10:03:11,890] INFO Will not load MX4J, mx4j-tools.jar is not in the classpath (kafka.utils.Mx4jLoader$)

  61. [2014-08-05 10:03:11,919] INFO 1 successfully elected as leader (kafka.server.ZookeeperLeaderElector)

  62. [2014-08-05 10:03:12,359] INFO New leader is 1 (kafka.server.ZookeeperLeaderElector$LeaderChangeListener)

  63. [2014-08-05 10:03:12,387] INFO Registered broker 1 at path /brokers/ids/1 with address m1:9092. (kafka.utils.ZkUtils$)

  64. [2014-08-05 10:03:12,392] INFO [Kafka Server 1], started (kafka.server.KafkaServer)

  65. [2014-08-05 10:03:12,671] INFO [ReplicaFetcherManager on broker 1] Removed fetcher for partitions [idoall.org,0],[test,0],[test,1] (kafka.server.ReplicaFetcherManager)

  66. [2014-08-05 10:03:12,741] INFO [ReplicaFetcherManager on broker 1] Removed fetcher for partitions [idoall.org,0],[test,0],[test,1] (kafka.server.ReplicaFetcherManager)

  67. [2014-08-05 10:03:25,327] INFO Partition [test,0] on broker 1: Expanding ISR for partition [test,0] from 1 to 1,2 (kafka.cluster.Partition)

  68. [2014-08-05 10:03:25,334] INFO Partition [test,1] on broker 1: Expanding ISR for partition [test,1] from 1 to 1,2 (kafka.cluster.Partition)

  69. [2014-08-05 10:03:26,905] INFO Partition [test,1] on broker 1: Expanding ISR for partition [test,1] from 1,2 to 1,2,3 (kafka.cluster.Partition)
复制代码

　4、测试kafka的状态
　　　　1)在m1上创建一个idoall_testTopic主题

  1. #KAFKA有几个，replication-factor就填几个

  2. root@m1:/home/hadoop# /home/hadoop/kafka_2.9.2-0.8.1.1/bin/kafka-topics.sh --create --topic idoall_testTopic --replication-factor 4 --partitions 2 --zookeeper m1:2181

  3. Created topic "idoall_testTopic".

  4. [2014-08-05 10:08:29,315] INFO [ReplicaFetcherManager on broker 1] Removed fetcher for partitions [idoall_testTopic,0] (kafka.server.ReplicaFetcherManager)

  5. [2014-08-05 10:08:29,334] INFO Completed load of log idoall_testTopic-0 with log end offset 0 (kafka.log.Log)

  6. [2014-08-05 10:08:29,373] INFO Created log for partition [idoall_testTopic,0] in /home/hadoop/kafka_2.9.2-0.8.1.1/kafka-logs with properties {segment.index.bytes -> 10485760, file.delete.delay.ms -> 60000, segment.bytes -> 536870912, flush.ms -> 9223372036854775807, delete.retention.ms -> 86400000, index.interval.bytes -> 4096, retention.bytes -> -1, cleanup.policy -> delete, segment.ms -> 604800000, max.message.bytes -> 1000012, flush.messages -> 9223372036854775807, min.cleanable.dirty.ratio -> 0.5, retention.ms -> 604800000}. (kafka.log.LogManager)

  7. [2014-08-05 10:08:29,384] WARN Partition [idoall_testTopic,0] on broker 1: No checkpointed highwatermark is found for partition [idoall_testTopic,0] (kafka.cluster.Partition)

  8. [2014-08-05 10:08:29,415] INFO Completed load of log idoall_testTopic-1 with log end offset 0 (kafka.log.Log)

  9. [2014-08-05 10:08:29,416] INFO Created log for partition [idoall_testTopic,1] in /home/hadoop/kafka_2.9.2-0.8.1.1/kafka-logs with properties {segment.index.bytes -> 10485760, file.delete.delay.ms -> 60000, segment.bytes -> 536870912, flush.ms -> 9223372036854775807, delete.retention.ms -> 86400000, index.interval.bytes -> 4096, retention.bytes -> -1, cleanup.policy -> delete, segment.ms -> 604800000, max.message.bytes -> 1000012, flush.messages -> 9223372036854775807, min.cleanable.dirty.ratio -> 0.5, retention.ms -> 604800000}. (kafka.log.LogManager)

  10. [2014-08-05 10:08:29,422] WARN Partition [idoall_testTopic,1] on broker 1: No checkpointed highwatermark is found for partition [idoall_testTopic,1] (kafka.cluster.Partition)

  11. [2014-08-05 10:08:29,430] INFO [ReplicaFetcherManager on broker 1] Removed fetcher for partitions [idoall_testTopic,1] (kafka.server.ReplicaFetcherManager)

  12. [2014-08-05 10:08:29,438] INFO Truncating log idoall_testTopic-1 to offset 0. (kafka.log.Log)

  13. [2014-08-05 10:08:29,473] INFO [ReplicaFetcherManager on broker 1] Added fetcher for partitions ArrayBuffer([[idoall_testTopic,1], initOffset 0 to broker id:2,host:m2,port:9092] ) (kafka.server.ReplicaFetcherManager)

  14. [2014-08-05 10:08:29,475] INFO [ReplicaFetcherThread-0-2], Starting  (kafka.server.ReplicaFetcherThread)
复制代码

　　　　2)在m1上查看刚才创建的idoall_testTopic主题
  1. root@m1:/home/hadoop# /home/hadoop/kafka_2.9.2-0.8.1.1/bin/kafka-topics.sh --list --zookeeper m1:2181   

  2. idoall_testTopic
复制代码

　　　　3)在m2上发送消息至kafka（m2模拟producer），发送消息“hello idoall.org”
  1. root@m2:/home/hadoop# /home/hadoop/kafka_2.9.2-0.8.1.1/bin/kafka-console-producer.sh --broker-list m1:9092 --sync --topic idoall_testTopic

  2. SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder".

  3. SLF4J: Defaulting to no-operation (NOP) logger implementation

  4. SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details.

  5. hello idoall.org
复制代码

　　　　4)在s1上开启一个消费者（s1模拟consumer），可以看到刚才发送的消息
  1. root@s1:/home/hadoop# /home/hadoop/kafka_2.9.2-0.8.1.1/bin/kafka-console-consumer.sh --zookeeper m1:2181 --topic idoall_testTopic --from-beginning

  2. SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder".

  3. SLF4J: Defaulting to no-operation (NOP) logger implementation

  4. SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details.

  5. hello idoall.org 
复制代码

　　　　5)删除掉一个Topic，这里我们测试创建一个idoall的主题，再删除掉
  1. root@m1:/home/hadoop# /home/hadoop/kafka_2.9.2-0.8.1.1/bin/kafka-topics.sh --create --topic idoall --replication-factor 4 --partitions 2 --zookeeper m1:2181          

  2. Created topic "idoall".

  3. [2014-08-05 10:38:30,862] INFO Completed load of log idoall-1 with log end offset 0 (kafka.log.Log)

  4. [2014-08-05 10:38:30,864] INFO Created log for partition [idoall,1] in /home/hadoop/kafka_2.9.2-0.8.1.1/kafka-logs with properties {segment.index.bytes -> 10485760, file.delete.delay.ms -> 60000, segment.bytes -> 536870912, flush.ms -> 9223372036854775807, delete.retention.ms -> 86400000, index.interval.bytes -> 4096, retention.bytes -> -1, cleanup.policy -> delete, segment.ms -> 604800000, max.message.bytes -> 1000012, flush.messages -> 9223372036854775807, min.cleanable.dirty.ratio -> 0.5, retention.ms -> 604800000}. (kafka.log.LogManager)

  5. [2014-08-05 10:38:30,870] WARN Partition [idoall,1] on broker 1: No checkpointed highwatermark is found for partition [idoall,1] (kafka.cluster.Partition)

  6. [2014-08-05 10:38:30,878] INFO [ReplicaFetcherManager on broker 1] Removed fetcher for partitions [idoall,1] (kafka.server.ReplicaFetcherManager)

  7. [2014-08-05 10:38:30,880] INFO Truncating log idoall-1 to offset 0. (kafka.log.Log)

  8. [2014-08-05 10:38:30,885] INFO [ReplicaFetcherManager on broker 1] Added fetcher for partitions ArrayBuffer([[idoall,1], initOffset 0 to broker id:3,host:s1,port:9092] ) (kafka.server.ReplicaFetcherManager)

  9. [2014-08-05 10:38:30,887] INFO [ReplicaFetcherThread-0-3], Starting  (kafka.server.ReplicaFetcherThread)

  10. root@m1:/home/hadoop# /home/hadoop/kafka_2.9.2-0.8.1.1/bin/kafka-topics.sh --list --zookeeper m1:2181

  11. idoall

  12. idoall_testTopic

  13. root@m1:/home/hadoop# /home/hadoop/kafka_2.9.2-0.8.1.1/bin/kafka-run-class.sh kafka.admin.DeleteTopicCommand --topic idoall --zookeeper m2:2181    

  14. deletion succeeded!

  15. root@m1:/home/hadoop# /home/hadoop/kafka_2.9.2-0.8.1.1/bin/kafka-topics.sh --list --zookeeper m1:2181                  idoall_testTopic

  16. root@m1:/home/hadoop#
复制代码

　　　　同样也可以进入到zookeeper中查看主题是否已经删除掉。
  1. root@m1:/home/hadoop# /home/hadoop/zookeeper-3.4.5/bin/zkCli.sh

  2. Connecting to localhost:2181

  3. 2014-08-05 10:15:21,863 [myid:] - INFO  [main:Environment@100] - Client environment:zookeeper.version=3.4.5-1392090, built on 09/30/2012 17:52 GMT

  4. 2014-08-05 10:15:21,871 [myid:] - INFO  [main:Environment@100] - Client environment:host.name=m1

  5. 2014-08-05 10:15:21,871 [myid:] - INFO  [main:Environment@100] - Client environment:java.version=1.7.0_65

  6. 2014-08-05 10:15:21,872 [myid:] - INFO  [main:Environment@100] - Client environment:java.vendor=Oracle Corporation

  7. 2014-08-05 10:15:21,872 [myid:] - INFO  [main:Environment@100] - Client environment:java.home=/usr/lib/jvm/java-7-oracle/jre

  8. 2014-08-05 10:15:21,873 [myid:] - INFO  [main:Environment@100] - Client environment:java.class.path=/home/hadoop/zookeeper-3.4.5/bin/../build/classes:/home/hadoop/zookeeper-3.4.5/bin/../build/lib/*.jar:/home/hadoop/zookeeper-3.4.5/bin/../lib/slf4j-log4j12-1.6.1.jar:/home/hadoop/zookeeper-3.4.5/bin/../lib/slf4j-api-1.6.1.jar:/home/hadoop/zookeeper-3.4.5/bin/../lib/netty-3.2.2.Final.jar:/home/hadoop/zookeeper-3.4.5/bin/../lib/log4j-1.2.15.jar:/home/hadoop/zookeeper-3.4.5/bin/../lib/jline-0.9.94.jar:/home/hadoop/zookeeper-3.4.5/bin/../zookeeper-3.4.5.jar:/home/hadoop/zookeeper-3.4.5/bin/../src/java/lib/*.jar:/home/hadoop/zookeeper-3.4.5/bin/../conf:.:/usr/lib/jvm/java-7-oracle/lib/tools.jar:/usr/lib/jvm/java-7-oracle/lib/dt.jar

  9. 2014-08-05 10:15:21,874 [myid:] - INFO  [main:Environment@100] - Client environment:java.library.path=:/usr/local/lib:/usr/java/packages/lib/amd64:/usr/lib64:/lib64:/lib:/usr/lib

  10. 2014-08-05 10:15:21,874 [myid:] - INFO  [main:Environment@100] - Client environment:java.io.tmpdir=/tmp

  11. 2014-08-05 10:15:21,874 [myid:] - INFO  [main:Environment@100] - Client environment:java.compiler=<NA>

  12. 2014-08-05 10:15:21,875 [myid:] - INFO  [main:Environment@100] - Client environment:os.name=Linux

  13. 2014-08-05 10:15:21,875 [myid:] - INFO  [main:Environment@100] - Client environment:os.arch=amd64

  14. 2014-08-05 10:15:21,876 [myid:] - INFO  [main:Environment@100] - Client environment:os.version=3.11.0-15-generic

  15. 2014-08-05 10:15:21,876 [myid:] - INFO  [main:Environment@100] - Client environment:user.name=root

  16. 2014-08-05 10:15:21,877 [myid:] - INFO  [main:Environment@100] - Client environment:user.home=/root

  17. 2014-08-05 10:15:21,878 [myid:] - INFO  [main:Environment@100] - Client environment:user.dir=/home/hadoop

  18. 2014-08-05 10:15:21,879 [myid:] - INFO  [main:ZooKeeper@438] - Initiating client connection, connectString=localhost:2181 sessionTimeout=30000 watcher=org.apache.zookeeper.ZooKeeperMain$MyWatcher@666c211a

  19. Welcome to ZooKeeper!

  20. 2014-08-05 10:15:21,920 [myid:] - INFO  [main-SendThread(localhost:2181):ClientCnxn$SendThread@966] - Opening socket connection to server localhost/127.0.0.1:2181. Will not attempt to authenticate using SASL (unknown error)

  21. 2014-08-05 10:15:21,934 [myid:] - INFO  [main-SendThread(localhost:2181):ClientCnxn$SendThread@849] - Socket connection established to localhost/127.0.0.1:2181, initiating session

  22. JLine support is enabled

  23. 2014-08-05 10:15:21,966 [myid:] - INFO  [main-SendThread(localhost:2181):ClientCnxn$SendThread@1207] - Session establishment complete on server localhost/127.0.0.1:2181, sessionid = 0x147a3e1246b0007, negotiated timeout = 30000

  24.  

  25. WATCHER::

  26.  

  27. WatchedEvent state:SyncConnected type:None path:null

  28. [zk: localhost:2181(CONNECTED) 0] ls /

  29. [hbase, hadoop-ha, admin, zookeeper, consumers, config, controller, storm, brokers, controller_epoch]

  30. [zk: localhost:2181(CONNECTED) 1] ls /brokers

  31. [topics, ids]

  32. [zk: localhost:2181(CONNECTED) 2] ls /brokers/topics

  33. [idoall_testTopic]
复制代码

　5、使用Eclipse来调用kafka的JAVA API来测试kafka的集群状态
　　　　1)消息生产端：Producertest.java

  1. package idoall.testkafka;

  2.  

  3. import java.util.Date;

  4. import java.util.Properties;

  5. import java.text.SimpleDateFormat;   

  6.  

  7. import kafka.javaapi.producer.Producer;

  8. import kafka.producer.KeyedMessage;

  9. import kafka.producer.ProducerConfig;

  10.  

  11.  

  12. /**

  13. * 消息生产端

  14. * @author 迦壹

  15. * @Time 2014-08-05

  16. */

  17. public class Producertest {

  18.      

  19.      public static void main(String[] args) {

  20.          Properties props = new Properties();

  21.          props.put("zk.connect", "m1:2181,m2:2181,s1:2181,s2:2181");

  22.          // serializer.class为消息的序列化类

  23.          props.put("serializer.class", "kafka.serializer.StringEncoder");

  24.          // 配置metadata.broker.list, 为了高可用, 最好配两个broker实例

  25.          props.put("metadata.broker.list", "m1:9092,m2:9092,s1:9092,s2:9092");

  26.          // 设置Partition类, 对队列进行合理的划分

  27.          //props.put("partitioner.class", "idoall.testkafka.Partitionertest");

  28.          // ACK机制, 消息发送需要kafka服务端确认

  29.          props.put("request.required.acks", "1");

  30.  

  31.           props.put("num.partitions", "4");

  32.          ProducerConfig config = new ProducerConfig(props);

  33.          Producer<String, String> producer = new Producer<String, String>(config);

  34.          for (int i = 0; i < 10; i++)

  35.          {

  36.            // KeyedMessage<K, V>

  37.            // 　　K对应Partition Key的类型

  38.            // 　　V对应消息本身的类型

  39. //　　 topic: "test", key: "key", message: "message"

  40.            SimpleDateFormat formatter = new SimpleDateFormat   ("yyyy年MM月dd日 HH:mm:ss SSS");       

  41.            Date curDate = new Date(System.currentTimeMillis());//获取当前时间       

  42.            String str = formatter.format(curDate);   

  43.             

  44.            String msg = "idoall.org" + i+"="+str;

  45.            String key = i+"";

  46.            producer.send(new KeyedMessage<String, String>("idoall_testTopic",key, msg));

  47.          }

  48.        }

  49. }
复制代码

　　　　2)消息消费端：Consumertest.java
  1. package idoall.testkafka;

  2.  

  3. import java.util.HashMap; 

  4. import java.util.List;  

  5. import java.util.Map;  

  6. import java.util.Properties;  

  7.    

  8. import kafka.consumer.ConsumerConfig;  

  9. import kafka.consumer.ConsumerIterator;  

  10. import kafka.consumer.KafkaStream;  

  11. import kafka.javaapi.consumer.ConsumerConnector;

  12.  

  13.  

  14. /**

  15. * 消息消费端

  16. * @author 迦壹

  17. * @Time 2014-08-05

  18. */

  19. public class Consumertest extends Thread{

  20.       

  21.      private final ConsumerConnector consumer;  

  22.     private final String topic;  

  23.  

  24.     public static void main(String[] args) {  

  25.       Consumertest consumerThread = new Consumertest("idoall_testTopic");  

  26.         consumerThread.start();  

  27.     }  

  28.     public Consumertest(String topic) {  

  29.         consumer =kafka.consumer.Consumer.createJavaConsumerConnector(createConsumerConfig());  

  30.         this.topic =topic;  

  31.     }  

  32.  

  33. private static ConsumerConfig createConsumerConfig() {  

  34.     Properties props = new Properties();  

  35.     // 设置zookeeper的链接地址

  36.     props.put("zookeeper.connect","m1:2181,m2:2181,s1:2181,s2:2181");  

  37.     // 设置group id

  38.     props.put("group.id", "1");  

  39.     // kafka的group 消费记录是保存在zookeeper上的, 但这个信息在zookeeper上不是实时更新的, 需要有个间隔时间更新

  40.     props.put("auto.commit.interval.ms", "1000");

  41.     props.put("zookeeper.session.timeout.ms","10000");  

  42.     return new ConsumerConfig(props);  

  43. }  

  44.  

  45. public void run(){  

  46.      //设置Topic=>Thread Num映射关系, 构建具体的流

  47.     Map<String,Integer> topickMap = new HashMap<String, Integer>();  

  48.     topickMap.put(topic, 1);  

  49.     Map<String, List<KafkaStream<byte[],byte[]>>>  streamMap=consumer.createMessageStreams(topickMap);  

  50.     KafkaStream<byte[],byte[]>stream = streamMap.get(topic).get(0);  

  51.     ConsumerIterator<byte[],byte[]> it =stream.iterator();  

  52.     System.out.println("*********Results********");  

  53.     while(it.hasNext()){  

  54.         System.err.println("get data:" +new String(it.next().message()));  

  55.         try {  

  56.             Thread.sleep(1000);  

  57.         } catch (InterruptedException e) {  

  58.             e.printStackTrace();  

  59.         }  

  60.     }  

  61. }  

  62. }
复制代码



　　　　3)在Eclipse查看java代码效果，在这之前先在其中一台机器(我使用的s1)，开启消费者，同时观察eclipse和s1上的消费者是否都收到了消息。最后结果如下图：
2.png

----------两张图片之间的分隔线----------
3.png


　　　　可以看到，刚好10条信息，没有丢失。不过消息因为均衡的原因，并非是有序的，在Kafka只提供了分区内部的有序性，不能跨partition. 每个分区的有序性，结合按Key分partition的能力对大多应用都够用了。（如何按key进行分partition，在文章末尾提供的Eclpise代码中有个Partitionertest.java提供了一个Demo）

　　6、在命令行下打包java文件，测试kafka
　　　　1)修改工程目录中的pom.xml文件

  1. <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

  2.   xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">

  3.   <modelVersion>4.0.0</modelVersion>

  4.   <groupId>idoall.testkafka</groupId>

  5.   <artifactId>idoall.testkafka</artifactId>

  6.   <version>0.0.1-SNAPSHOT</version>

  7.   <packaging>jar</packaging>

  8.   <name>idoall.testkafka</name>

  9.   <url>http://maven.apache.org</url>

  10.   <properties>

  11.     <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>

  12.   </properties>

  13.   <dependencies>

  14.     <dependency>

  15.       <groupId>junit</groupId>

  16.       <artifactId>junit</artifactId>

  17.       <version>3.8.1</version>

  18.       <scope>test</scope>

  19.     </dependency>

  20.     <dependency>

  21.       <groupId>log4j</groupId>

  22.       <artifactId>log4j</artifactId>

  23.       <version>1.2.14</version>

  24.     </dependency>

  25.     <dependency>

  26.       <groupId>com.sksamuel.kafka</groupId>

  27.       <artifactId>kafka_2.10</artifactId>

  28.       <version>0.8.0-beta1</version>

  29.     </dependency>

  30.   </dependencies>

  31.   <build>  

  32.         <finalName>idoall.testkafka</finalName>  

  33.         <plugins>  

  34.             <plugin>  

  35.                 <groupId>org.apache.maven.plugins</groupId>  

  36.                 <artifactId>maven-compiler-plugin</artifactId>  

  37.                 <version>2.0.2</version>  

  38.                 <configuration>  

  39.                     <source>1.5</source>  

  40.                     <target>1.5</target>  

  41.                     <encoding>UTF-8</encoding>  

  42.                 </configuration>  

  43.             </plugin>  

  44.             <plugin>  

  45.                 <artifactId>maven-assembly-plugin</artifactId>  

  46.                 <version>2.4</version>  

  47.                 <configuration>  

  48.                     <descriptors>  

  49.                         <descriptor>src/main/src.xml</descriptor>  

  50.                     </descriptors>  

  51.                     <descriptorRefs>  

  52.                         <descriptorRef>jar-with-dependencies</descriptorRef>  

  53.                     </descriptorRefs>  

  54.                 </configuration>  

  55.                 <executions>  

  56.                     <execution>  

  57.                         <id>make-assembly</id> <!-- this is used for inheritance merges --> 

  58.                         <phase>package</phase> <!-- bind to the packaging phase --> 

  59.                         <goals>  

  60.                             <goal>single</goal>  

  61.                         </goals>  

  62.                     </execution>  

  63.                 </executions>  

  64.             </plugin>  

  65.         </plugins>  

  66.     </build>

  67. </project>
复制代码

　　　　2)修改工程目录中的src/main/src.xml文件

  1. <?xml version="1.0" encoding="UTF-8"?>

  2. <assembly xmlns="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0"

  3.           xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

  4.           xsi:schemaLocation="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0 http://maven.apache.org/xsd/assembly-1.1.0.xsd ">

  5.     <id>jar-with-dependencies</id>

  6.     <formats>

  7.         <format>jar</format>

  8.     </formats>

  9.     <includeBaseDirectory>false</includeBaseDirectory>

  10.     <dependencySets>

  11.         <dependencySet>

  12.             <unpack>false</unpack>

  13.             <scope>runtime</scope>

  14.         </dependencySet>

  15.     </dependencySets>

  16.     <fileSets>

  17.         <fileSet>

  18.             <directory>/lib</directory>

  19.         </fileSet>

  20.     </fileSets>

  21. </assembly>
复制代码

　　　　3)制作依赖包，在工程目录执行mvn package，得到idoall.testkafka-jar-with-dependencies.jar，下面是部分执行后的结果：
  1. Running idoall.testkafka.idoall.testkafka.AppTest

  2. Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.004 sec

  3.  

  4. Results :

  5.  

  6. Tests run: 1, Failures: 0, Errors: 0, Skipped: 0

  7.  

  8. [INFO] 

  9. [INFO] --- maven-jar-plugin:2.4:jar (default-jar) @ idoall.testkafka ---

  10. [INFO] Building jar: /Users/lion/Documents/_my_project/java/idoall.testkafka/target/idoall.testkafka.jar

  11. [INFO] 

  12. [INFO] --- maven-assembly-plugin:2.4:single (make-assembly) @ idoall.testkafka ---

  13. [INFO] Reading assembly descriptor: src/main/src.xml

  14. [WARNING] The assembly id jar-with-dependencies is used more than once.

  15. [INFO] Building jar: /Users/lion/Documents/_my_project/java/idoall.testkafka/target/idoall.testkafka-jar-with-dependencies.jar

  16. [INFO] Building jar: /Users/lion/Documents/_my_project/java/idoall.testkafka/target/idoall.testkafka-jar-with-dependencies.jar

  17. [INFO] ------------------------------------------------------------------------

  18. [INFO] BUILD SUCCESS

  19. [INFO] ------------------------------------------------------------------------

  20. [INFO] Total time: 9.074 s

  21. [INFO] Finished at: 2014-08-05T12:22:47+08:00

  22. [INFO] Final Memory: 63M/836M

  23. [INFO] ------------------------------------------------------------------------
复制代码

　　　　4)编译文件，进入到工程目录，执行命令
  1. liondeMacBook-Pro:idoall.testkafka lion$ pwd

  2. /Users/lion/Documents/_my_project/java/idoall.testkafka

  3. liondeMacBook-Pro:idoall.testkafka lion$ javac -classpath target/idoall.testkafka-jar-with-dependencies.jar -d . src/main/java/idoall/testkafka/*.java

复制代码



　　　　5)执行编译后的文件。分别打开两个窗口，一个用来消费，一个用来生产。可以看到消费窗口可以正常显示消息。
  1. java -classpath .:target/idoall.testkafka-jar-with-dependencies.jar idoall.testkafka.Producertest

  2.  

  3. java -classpath .:target/idoall.testkafka-jar-with-dependencies.jar idoall.testkafka.Consumertest

复制代码

4.png

----------两张图片之间的分隔线----------
5.png


　　五、FAQ
　　1、如果在创建主题时出现下面的错误 ，那就是启动的brokers的个数达不到你所指定的--replication-factor值：


  1. Error while executing topic command replication factor: 3 larger than available brokers: 1

  2. kafka.admin.AdminOperationException: replication factor: 3 larger than available brokers: 1

  3.         at kafka.admin.AdminUtils$.assignReplicasToBrokers(AdminUtils.scala:70)

  4.         at kafka.admin.AdminUtils$.createTopic(AdminUtils.scala:155)

  5.         at kafka.admin.TopicCommand$.createTopic(TopicCommand.scala:86)

  6.         at kafka.admin.TopicCommand$.main(TopicCommand.scala:50)

  7.         at kafka.admin.TopicCommand.main(TopicCommand.scala)
复制代码

　2、如果出现下面的错误，可以先启动kafka，再启动hadoop中的zkfc(DFSZKFailoverController)：
  1. Java HotSpot(TM) 64-Bit Server VM warning: INFO: os::commit_memory(0x00000000c0130000, 986513408, 0) failed; error='Cannot allocate memory' (errno=12)

  2. #

  3. # There is insufficient memory for the Java Runtime Environment to continue.

  4. # Native memory allocation (malloc) failed to allocate 986513408 bytes for committing reserved memory.

  5. # An error report file with more information is saved as:

  6. # /home/hadoop/hs_err_pid13558.log
复制代码


代码下载：
链接: http://pan.baidu.com/s/1bnriYK3 密码: x20l





博文作者：迦壹
博客地址：http://idoall.org/home.php?mod=space&uid=1&do=blog&id=547
关注公众号，获取大数据、人工智能20套、区块链资源5阶段等资源，随时更新，获取最新技术资源
回复

使用道具 举报

自定义广告语

已有(11)人评论

跳转到指定楼层
正序浏览
ohano_javaee 发表于 2014-10-10 17:56:40
好东西先收着了
回复

使用道具 举报

hadoop小象 发表于 2014-12-1 11:12:35
Consumer Group （CG）：这是kafka用来实现一个topic消息的广播（发给所有的consumer）和单播（发给任意一个consumer）的手段。一个topic可以有多个CG。topic的消息会复制（不是真的复制，是概念上的）到所有的CG，但每个CG只会把消息发给该CG中的一个consumer。如果需要实现广播，只要每个consumer有一个独立的CG就可以了。要实现单播只要所有的consumer在同一个CG。用CG还可以将consumer进行自由的分组而不需要多次发送消息到不同的topic。
这句我怎么看着别扭呢，如果是广播，每个consumer的CG不同，而每个CG只会把消息发给该CG中的一个consumer，那广播还能把消息发给所有的consumer吗？
回复

使用道具 举报

longaili 发表于 2014-12-24 16:06:23
好东西，支持！
回复

使用道具 举报

ohano_javaee 发表于 2015-1-31 13:43:50
写的很好，正需要~
回复

使用道具 举报

ainubis 发表于 2015-3-29 01:34:08
好东西，支持！
回复

使用道具 举报

stephenKang 发表于 2016-8-24 16:16:37
谢谢分享，写的很详细
回复

使用道具 举报

shuijun1106 发表于 2016-9-27 11:22:23
楼主写的很详细，每一步都有记录
回复

使用道具 举报

空空未空 发表于 2016-11-12 14:57:39
这个资料非常棒，刚好解决我遇到的问题
回复

使用道具 举报

12下一页
高级模式
B Color Image Link Quote Code Smilies
您需要登录后才可以回帖 登录 | 立即注册

本版积分规则

发表新帖
pig2

超级版主

关注

  • 2329

    主题

  • 3339

    帖子

  • 1972

    粉丝

欢迎大家如about云官方2群322273151，更新咨询，更新资源，随时关注
TA的主题
关闭

推荐上一条 /2 下一条

关于我们· 联系我们· 加入我们· 小黑屋· 合作伙伴   

Copyright © 2001-2024 About云-梭伦科技 Powered by Discuz! X3.4 Licensed Discuz Team.

简书 / Smrz 京ICP备2020039040号 Wxb 简书网举报电话：021-34700000