大数据技术交流群 :494721467

安装storm

-------------------

1.准备

jdk + zookeeper

2.storm安装

下载apache-storm-1.0.1.tar.gz + apache-storm-1.0.1.-src-tar.gz

3.tar

4.移动文件

$>mv /soft/..

5.配置环境变量

[/etc/environment]

path

STORM_HOME=

6.配置storm

[conf/yaml]

storm.zookeeper.servers:

 - "s2"

 - "s3"

 - "s4"

storm.local.dir: "/home/ubuntu/storm"

nimbus.host: "s1"

supervisor.slots.ports:

- 6700

- 6701

- 6703

#

# nimbus.seeds: ["host1", "host2", "host3"]

#

#

# ##### These may optionally be filled in:

#    

## List of custom serializations

# topology.kryo.register:

#     - org.mycompany.MyType

#     - org.mycompany.MyType2: org.mycompany.MyType2Serializer

#

## List of custom kryo decorators

# topology.kryo.decorators:

#     - org.mycompany.MyDecorator

#

## Locations of the drpc servers

# drpc.servers:

#     - "server1"

#     - "server2"

## Metrics Consumers

# topology.metrics.consumer.register:

#   - class: "org.apache.storm.metric.LoggingMetricsConsumer"

#     parallelism.hint: 1

#   - class: "org.mycompany.MyMetricsConsumer"

#     parallelism.hint: 1

#     argument:

#       - endpoint: "metrics-collector.mycompany.org"

7.启动zk集群

$>zkServer.sh start

$>zkServer.sh status

8.启动storm集成

$>storm nimbus

$>storm supervisor

$>storm ui

本文出自 “11009607” 博客，转载请与作者联系！

storm