系统监控软件ganglia3.2.0

fgew2

实验环境如下图所示

1.上图中三台机器均为rhel 5.8 x86_64,三台机器IP地址和主机名已按上图设置完毕
2.各软件版本如下
◆.confuse-2.7.tar.gz
◆.ganglia-3.2.0.tar.gz
◆.rrdtool-1.4.5.tar.gz
3.管理节点(admin),hosts文件已添加完毕，如下所示
192.168.10.11       admin
192.168.10.100  node1
192.168.10.101  node2
4.本次实验ganglia通信采用单播方式

【管理端（admin）安装ganglia】
一：安装所需软件
[iyunv@admin ~]#yum -y install apr-devel expat-devel
> pcre-devellibxml2-devel cairo-devel
> pango-develpython-devel

注：RHEL 6系列不能只安装python-devel,可以直接python*,还要加装perl-ExtUtils*

二：编译安装confuse
[root@adminganglia-3.2.0]# tar xzf confuse-2.7.tar.gz
[root@adminganglia-3.2.0]# cd confuse-2.7
[root@adminconfuse-2.7]# ./configure --enable-shared --libdir=/usr/lib64
[root@adminconfuse-2.7]# make
[root@adminconfuse-2.7]# make install

三:编译安装rrdtool
[root@adminganglia-3.2.0]# tar xzf rrdtool-1.4.5.tar.gz
[root@adminganglia-3.2.0]# cd rrdtool-1.4.5
[root@adminrrdtool-1.4.5]# ./configure --prefix=/usr/local/rrdtool
[root@adminrrdtool-1.4.5]# make
[root@adminrrdtool-1.4.5]# make install

四：编译安装ganglia
[root@adminganglia-3.2.0]# ./configure --prefix=/usr/local/ganglia
>--sysconfdir=/etc/ganglia
>--with-librrd=/usr/local/rrdtool
> --with-gmetad
>--enable-gexec
>--enable-status
[root@adminganglia-3.2.0]# make
[root@adminganglia-3.2.0]# make install

五：配置ganglia
◆.链接命令文件
[root@adminganglia-3.2.0]# ln -s /usr/local/ganglia/bin/* /usr/local/bin/
[root@adminganglia-3.2.0]# ln -s /usr/local/ganglia/sbin/* /usr/local/sbin/
◆.配置gmetad
1.[root@adminganglia-3.2.0]# cd gmetad
2.[iyunv@admingmetad]# cp gmetad.init /etc/init.d/gmetad
3.[iyunv@admingmetad]# chkconfig --add gmetad;chkconfig --list gmetad
gmetad          0:off   1:off  2:on    3:on    4:on   5:on    6:off
4.[iyunv@admingmetad]# vi /etc/init.d/gmetad 修改GMETAD=/usr/sbin/gmetad为GMETAD=/usr/local/ganglia/sbin/gmetad
5.[iyunv@admingmetad]# vi /etc/ganglia/gmetad.conf，找到第45行：data_source "my cluster" localhost，把"my cluster"可一更改为自己喜欢的名称，比如wjcyf,其他默认即可
6.创建所需目录，并设置所有者和权限
[iyunv@admin gmetad]#mkdir -p /var/lib/ganglia/rrds
[iyunv@admingmetad]# chown nobody:nobody /var/lib/ganglia/rrds/
[iyunv@admingmetad]# chmod 1777 /var/lib/ganglia/rrds/
7.启动gmetad服务
[iyunv@admingmetad]# /etc/init.d/gmetad start
Starting GANGLIAgmetad:                                   [  OK  ]
◆.配置gmond
1.[iyunv@admingmetad]# cd ../gmond/
2.[iyunv@admingmond]# cp gmond.init /etc/init.d/gmond
3.[iyunv@admingmond]# chkconfig --add gmond ;chkconfig --list gmond
gmond           0:off   1:off  2:on    3:on    4:on   5:on    6:off
4. [iyunv@admingmond]# vi /etc/init.d/gmond修改GMOND=/usr/sbin/gmond为GMOND=/usr/local/ganglia/sbin/gmond
5.生成默认配置文件gmond.conf
[iyunv@admingmond]# ./gmond -t >/etc/ganglia/gmond.conf
6. [iyunv@admingmond]# vi /etc/ganglia/gmond.conf,本次实验配置ganglia server节点为“单收”，配置文件需要修改的地方为下面红色文字所示，其余均默认
/* This configuration is as close to 2.5.x defaultbehavior as possible
   The valuesclosely match ./gmond/metric.h definitions in 2.5.x */
globals {
  daemonize =yes
  setuid = yes
  user = nobody
  debug_level =0
max_udp_msg_len = 1472
  mute = yes
  deaf = no
allow_extra_data = yes
  host_dmax = 0/*secs */
  host_tmax = 20/*secs */
cleanup_threshold = 300 /*secs */
  gexec = no
send_metadata_interval = 0 /*secs */
}

/*
* The clusterattributes specified will be used as part of the
* tag that willwrap all hosts collected by this instance.
*/
cluster {
  name = "wjcyf"
  owner ="unspecified"
  latlong ="unspecified"
  url ="unspecified"
}

/* The host section describes attributes of the host,like the location */
host {
  location ="unspecified"
}

/* Feel free to specify as many udp_send_channels asyou like.  Gmond
   used to onlysupport having a single channel */
udp_send_channel {
  #bind_hostname= yes # Highly recommended, soon to be default.
                      # This option tells gmond to use a source address
                      # that resolves to the machine's hostname.  Without
                      # this, the metrics may appear to come from any
                      # interface and the DNS names associated with
                      # those IPs will be used to create the RRDs.
  mcast_join =239.2.11.71
  port = 8649
  ttl = 1
}

/* You can specify as many udp_recv_channels as youlike as well. */
udp_recv_channel {
  port= 8649
  bind= 192.168.10.11
}

/* You can specify as many tcp_accept_channels as youlike to share
   an xmldescription of the state of the cluster */
tcp_accept_channel {
  port = 8649
    acl {
       default = "deny"
        access {
           ip= 127.0.0.1
           mask= 32
           action= "allow"
    }
   }
}
以下配置文件略

7.启动gmond服务
[iyunv@admingmond]# /etc/init.d/gmond start
Starting GANGLIAgmond:                                   [  OK  ]

六：安装apache配置ganglia-web
1. [iyunv@admingmond]# yum -y install httpd php php-gd
2. [root@adminganglia-3.2.0]# cp -r web/ /var/www/html/ganglia
3. [root@adminganglia-3.2.0]# cd /var/www/html/ganglia/
4. [iyunv@adminganglia]# cp conf.php conf.php.bak
5.[iyunv@admin ganglia]#vi conf.php，更改第33行define("RRDTOOL","/usr/bin/rrdtool");为
define("RRDTOOL","/usr/local/rrdtool/bin/rrdtool");
6.创建所需目录，并设置所有者和权限
[iyunv@adminganglia]# mkdir /var/lib/ganglia/dwoo
[iyunv@adminganglia]# chown apache:apache /var/lib/ganglia/dwoo/
[iyunv@adminganglia]# chmod 777 /var/lib/ganglia/dwoo/
7. [iyunv@adminganglia]# vi /etc/httpd/conf/httpd.conf，把第256行#ServerName www.example.com:80前面的#号去掉
8.启动apache服务
[iyunv@adminganglia]# /etc/init.d/httpd start ; chkconfig httpd on
Startinghttpd:                                            [  OK  ]

七：访问ganglia server
1.http://192.168.10.11/ganglia,因为此服务器被配置为“单收”，所以会出现下面的网页显示（有被监控端后，显示正常）



【被监控端安装ganglia】
一：安装所需软件
[iyunv@node1 ~]#yum -y install apr-devel expat-devel pcre-devel python-devel
二：编译安装confuse
[root@node1ganglia-3.2.0]# tar xzf confuse-2.7.tar.gz
[root@node1ganglia-3.2.0]# cd confuse-2.7
[root@node1confuse-2.7]# ./configure --enable-shared --libdir=/usr/lib64
[root@node1confuse-2.7]# make
[root@node1confuse-2.7]# make install
三：编译安装ganglia
[root@node1ganglia-3.2.0]# ./configure --prefix=/usr/local/ganglia
>--sysconfdir=/etc/ganglia
>--enable-gexec
>--enable-status
[root@node1ganglia-3.2.0]# make
[root@node1ganglia-3.2.0]# make install
四：配置ganglia
1.创建ganglia配置文件存放目录
[iyunv@node1 ganglia-3.2.0]#mkdir /etc/ganglia
2.[root@node1ganglia-3.2.0]# cd gmond/
3.[iyunv@node1gmond]# cp gmond.init /etc/init.d/gmond
4.[iyunv@node1gmond]# chkconfig --add gmond ;chkconfig --list gmond
gmond           0:off   1:off  2:on    3:on    4:on   5:on    6:off
5. [iyunv@node1gmond]# vi /etc/init.d/gmond修改GMOND=/usr/sbin/gmond为GMOND=/usr/local/ganglia/sbin/gmond
6.生成默认配置文件gmond.conf
[iyunv@node1gmond]# ./gmond -t >/etc/ganglia/gmond.conf
7.[iyunv@node1gmond]# vi /etc/ganglia/gmond.conf,本次实验配置被监控节点为“单发”，配置文件需要修改的地方为下面红色文字所示，其余均默认
/* This configuration is as close to 2.5.x defaultbehavior as possible
   The valuesclosely match ./gmond/metric.h definitions in 2.5.x */
globals {
  daemonize =yes
  setuid = yes
  user = nobody
  debug_level =0
  max_udp_msg_len= 1472
  mute = no
  deaf = yes
allow_extra_data = yes
  host_dmax = 86400/*secs */
  host_tmax = 20/*secs */
cleanup_threshold = 300 /*secs */
  gexec = no
  send_metadata_interval= 300 /*secs */
}

/*
* The clusterattributes specified will be used as part of the
* tag that willwrap all hosts collected by this instance.
*/
cluster {
  name = "wjcyf"
  owner ="unspecified"
  latlong ="unspecified"
  url ="unspecified"
}

/* The host section describes attributes of the host,like the location */
host {
  location ="unspecified"
}

/* Feel free to specify as many udp_send_channels asyou like.  Gmond
   used to onlysupport having a single channel */
udp_send_channel {
  host= 192.168.10.11
  port= 8649
}
以下配置文件略

8.启动gmond服务
[iyunv@node1 ~]#/etc/init.d/gmond start

注：node2操作同node1

五：重新访问ganglia server，则可正常显示node1和node2的信息，如下列图中所示




碰到的问题

• 在启动/etc/init.d/gmond服务时报错：gmond dead but subsys locked
  

A：原来是在目录/etc/ganglia/conf.d/下有个文件modpython.conf,我这样做了一个备份cp modpython.conf ~modpython.conf的缘故，删除~modpython.conf后，gmond服务正常启动