saltstack自动部署hadoop集群

1.1 操作系统

操作系统为：linux-3.10.0-327.el7.x86_64-x86_64-with-centos-7.2.1511-core

        或：linux-2.6.32-504.el6.x86_64-x86_64-with-centos-6.6-final

2 软件环境

2.1 防火墙

关闭各节点防火墙

#service iptables stop

#chkconfig iptables off

3 安装部署

3.1 Saltstack安装

3.1.1 服务分配

	系统版本	角色
10.10.2.34	CentosOS 6.5 x86_64	Master,Minion
10.10.2.35	CentosOS 6.5 x86_64	Minion
10.10.2.36	CentosOS 6.5 x86_64	Minion

 

3.1.2 软件包安装

Master 端安装：

Master#:curl -L https://bootstrap.saltstack.com -o install_salt.sh

Master#:sh install_salt.sh -M -N

 

Minion端安装：

Minion#:curl -L https://bootstrap.saltstack.com -o install_salt.sh

Minion#:sh install_salt.sh

3.1.3 配置文件修改

/etc/salt/master:

#####      File Server settings      #####

##########################################

# Salt runs a lightweight file server written in zeromq to deliver files to

# minions. This file server is built into the master daemon and does not

# require a dedicated port.

 

# The file server works on environments passed to the master, each environment

# can have multiple root directories, the subdirectories in the multiple file

# roots cannot match, otherwise the downloaded files will not be able to be

# reliably ensured. A base environment is required to house the top file.

# Example:

file_roots:

   base:

     - /srv/salt/

#   dev:

#     - /srv/salt/dev/services

#     - /srv/salt/dev/states

#   prod:

#     - /srv/salt/prod/services

#     - /srv/salt/prod/states

#

#file_roots:

#  base:

#    - /srv/salt

#

#####         Pillar settings        #####

##########################################

# Salt Pillars allow for the building of global data that can be made selectively

# available to different minions based on minion grain filtering. The Salt

# Pillar is laid out in the same fashion as the file server, with environments,

# a top file and sls files. However, pillar data does not need to be in the

# highstate format, and is generally just key/value pairs.

pillar_roots:

  base:

    - /srv/pillar

#

#ext_pillar:

#  - hiera: /etc/hiera.yaml

#  - cmd_yaml: cat /etc/salt/yaml

 

 

/etc/salt/minion:

# Set the location of the salt master server. If the master server cannot be

# resolved, then the minion will fail to start.

master: 10.10.2.34

 

# Set http proxy information for the minion when doing requests

#proxy_host:

#proxy_port:

#proxy_username:

#proxy_password:

# will be the hostname as returned by the python call: socket.getfqdn()

# Since salt uses detached ids it is possible to run multiple minions on the

# same machine but with different ids, this can be useful for salt compute

# clusters.

id: 三台minion分别设置为hadoop001，hadoop002，hadoop003

 

# Cache the minion id to a file when the minion's id is not statically defined

# in the minion config. Defaults to "True". This setting prevents potential

# problems when automatic minion id resolution changes, which can cause the

# minion to lose connection with the master. To turn off minion id caching,

# set this config to ``False``.

#minion_id_caching: True

 

3.1.4 服务认证

Master#：service salt-master start

（所有）Minion#：service salt-minion start

 

Master#：salt-key -L

Accepted Keys:

Denied Keys:

Unaccepted Keys:

hadoop001

hadoop002

hadoop003

Rejected Keys:

 

Master#：salt-key -A

The following keys are going to be accepted:

Unaccepted Keys:

hadoop001

hadoop002

hadoop003

Proceed? [n/Y] Y

Key for minion hadoop001 accepted.

Key for minion hadoop002 accepted.

Key for minion hadoop003 accepted.

 

Master#：salt ‘*’ test.ping

hadoop003:

    True

hadoop001:

    True

hadoop002:

True

3.2 Saltstack部署三节点hadoop集群

 

3.2.1  软件包

拷贝脚本至/srv/salt目录下，同时拷贝beh.tar.gz至该目录。

3.2.2 添加用户

# salt ‘*’ state.sls useradd

3.2.3 赋予权限

客户端均执行

#echo "hadoop ALL = (root) NOPASSWD:ALL" | sudo tee /etc/sudoers.d/hadoop

3.2.4 Grains值初始化

因为判断myid用到了自定义常量，如果没有该值，saltstack会报错。

$cd /srv/salt/python_work/

$python write_zookeeper.py

3.2.5 文件分发

$cd /srv/salt/python_work/

$python main.py copy

执行该命令后，会将beh.tar.gz同时分配至三台机器的/opt/beh目录下，并解压更改权限。

3.2.6 系统初始化

/srv/salt/hosts_add.sls：

修改为自己集群的ip

/etc/hosts:

    file.append:

        - text:

           - "10.10.2.43 hadoop001"

           - "10.10.2.44 hadoop002"

           - "10.10.2.45 hadoop003"

 

/srv/pillar/ip4.sls：

此文件为ntp安装时所需要的参数

Ip4修改为ntp master ip

Ip代表这一段ip地址

ip4: 10.10.2.43

ip: 10.10.2.0

 

$cd /srv/salt/python_work/

$Python main.py init

执行该命令后，三台机器会自动添加ip列表至/etc/hosts,会对应自己的ip写入正确的myid文件，关闭防火墙。根据/srv/pillar/disk.sls文件中的disk值作mkfs.ext4，挂载磁盘并写入fstab文件。安装ntp服务，修改文件并启动。修改最大进程数，最大文件数，修改mysql配置文件，为hadoop赋权，启动mysql服务。

Ø 

3.2.7 配置文件初始化

$cd /srv/salt/python_work/

$Python main.py create

执行该命令后，会将/opt/beh下所有组件的配置文件拷贝至/srv/salt/files/dir

 

3.2.8 配置文件更新

$cd /srv/salt/python_work/

$Python main.py refresh

在/srv/salt/files/dir目录下修改所需组件的配置文件后即可更新所有节点配置文件。

3.2.9 集群孵化

$cd /srv/salt/python_work/

$Python main.py hatch

集群会自动启动zookeeper服务，格式化HDFS的Zookeeper存储目录，启动JournalNode集群，格式化并启动第一个NameNode，格式化并启动第二个NameNode，启动所有DataNode，启动ZooKeeperFailoverController，启动yarn，jobhistory server，在hadoop001上启动timeline server，启动metastore，hmaster，hregionserver。

 软件包下载 可加qq 493854112

转载于:https://my.oschina.net/u/2547078/blog/810306