Ganglia：分布式监控系统
Ganglia：分布式监控系统
Ganglia的核心包含gmond、gmetad以及一个Web前端。主要是用来监控系统性能，如：cpu 、mem、硬盘利用率， I/O负载、网络流量情况等，通过曲线很容易见到每个节点的工作状态，对合理调整、分配系统资源，提高系统整体性能起到重要作用
Ganglia监控端安装
1、安装依赖的软件包
yum install ntp vim-enhanced gcc gcc-c++ flex bison autoconf automake bzip2-devel ncurses-devel zlib-devel libjpeg-devel libpng-devel libtiff-devel freetype-devel libXpm-devel gettext-devel pam-devel python-devel perl perl-devel expat expat-devel pcre pcre-devel apr apr-devel cairo-devel和pango-devel
需要安装rrdtool工具 最新版本的ganglia已可以直接使用yum安装rrdtool工具即可，
2、安装confuse
wget http://download.savannah.gnu.org/releases/confuse/confuse-2.7.tar.gz tar zxf confuse-2.7.tar.gz cd confuse-2.7 ./configure CFLAGS=-fPIC --disable-nls ;make;make install cd ..
3、安装ganglia
wget http://downloads.sourceforge.net/project/ganglia/ganglia%20monitoring%20core/3.3.1/ganglia-3.3.1.tar.gz tar zxf ganglia-3.3.1.tar.gz cd ganglia-3.3.1 #server 监控端 ./configure --prefix=/usr/local/ganglia --with-static-modules --enable-gexec --enable-status --with-gmetad --with-python=/usr --with-librrd=/usr/local/rrdtool --with-libexpat=/usr --with-libconfuse=/usr/local --with-libpcre make make install cd gmetad cp gmetad.conf /opt/modules/ganglia/etc/ cp gmetad.init /etc/init.d/gmetad vim /etc/init.d/gmetad 修改为GMETAD=/usr/local/ganglia/sbin/gmetad ip route add 239.2.11.71 dev eth0 ##添加广播路由
Ganglia：被监控端安装：
1、安装依赖的软件包
yum install ntp vim-enhanced gcc gcc-c++ flex bison autoconf automake bzip2-devel ncurses-devel zlib-devel libjpeg-devel libpng-devel libtiff-devel freetype-devel libXpm-devel gettext-devel pam-devel python-devel perl perl-devel expat expat-devel pcre pcre-devel apr apr-devel
wget http://download.savannah.gnu.org/releases/confuse/confuse-2.7.tar.gz tar zxf confuse-2.7.tar.gz cd confuse-2.7 ./configure CFLAGS=-fPIC --disable-nls ;make;make install cd ..
2、安装ganglia
wget http://downloads.sourceforge.net/project/ganglia/ganglia%20monitoring%20core/3.3.1/ganglia-3.6.0.tar.gz tar zxf ganglia-3.6.0.tar.gz cd ganglia-3.6.0 ./configure --prefix=/usr/local/ganglia --enable-gexec --enable-status --with-python=/usr --with-libapr --with-libconfuse=/usr/local --with-libexpat=/usr --with-libpcre make make install cd gmond ./gmond -t > /usr/local/ganglia/etc/gmond.conf cp gmetad.init /etc/init.d/gmond vim /etc/init.d/gmond 修改为： GMETAD=/usr/local/ganglia/sbin/gmond mkdir /usr/local/ganglia/lib64/ganglia/python_modules cp python_modules/*/*.py /usr/local/ganglia/lib64/ganglia/python_modules ip route add 239.2.11.71 dev eth0
安装已完成，过程简单。
Web网页页面在https://github.com/ganglia/ganglia-web
自行下载配置使用即可。
下面来监控，使其能正常工作和使用。
ganglia是分布式监控系统，也可以不使用分布来用。下面就分两种方式来解说。
1、不使用分布式来监控的方式
服务器配置文件 修改两项： data_source "test1" 192.168.107.2 data_source "test2" 172.16.1.4 ##这里分了两个监控组 data_source是关键字， “test1,test2” 是监控主机组的名字，全局要唯一。 后面跟着要监控的ip或主机名,如果有多个可以用空格隔开就可以了 gridname "Test" 这个是定义监控集的名字，
如下所示：
现在ganglia还没有启动，在启动之前执行下面命令：
可以使用命令查看是不是有配置错误的地方导致启动不了
/usr/local/ganglia/sbin/gmetad -d 5
出错的地方应该可以修改配置文件gmetad.conf得到修正、
service gmetad start 来启动服务
好了，再看客户端的配置吧
在客户端配置
globals {
daemonize = yes
setuid = yes
user = nobody
debug_level = 0
max_udp_msg_len = 1472
mute = no
deaf = no
allow_extra_data = yes
host_dmax = 86400 /*secs. Expires (removes from web interface) hosts in 1 day */
host_tmax = 20 /*secs */
cleanup_threshold = 300 /*secs */
gexec = no
# By default gmond will use reverse DNS resolution when displaying your hostname
# Uncommeting following value will override that value.
# override_hostname = "mywebserver.domain.com"
# If you are not using multicast this value should be set to something other than 0.
# Otherwise if you restart aggregator gmond you will get empty graphs. 60 seconds is reasonable
send_metadata_interval = 0 /*secs */
}
cluster {
name = "test1" ####需要修改与服务器端设定相同的名字 ###
owner = "nobody" ###修改为nobody
latlong = "unspecified"
url = "unspecified"
}
host {
location = "unspecified"
}
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 ##与设定的要一样哦
port = 8649 ##默认端口
ttl = 1
}
udp_recv_channel {
mcast_join = 239.2.11.71 ##与设定的要一样哦
port = 8649 ##默认端口
bind = 239.2.11.71
retry_bind = true
# Size of the UDP buffer. If you are handling lots of metrics you really
# should bump it up to e.g. 10MB or even higher.
# buffer = 10485760
}
/* You can specify as many tcp_accept_channels as you like to share
an xml description of the state of the cluster */
tcp_accept_channel {
port = 8649
# If you want to gzip XML output
gzip_output = no
}
启动客户端
在客户端也是可以使用debug来调试配置是不是有错
/usr/local/ganglia/sbin/gmond -d 5
service gmond start
下面来说第2种分布式监控系统
2、分布式监控
主gmetad 多个次gmetad 被监控点
|------ gmond
_ gmetad---|------ gmond
| |------ gmond
|
gmetad-------|_ gmetad---|-------gmond
| | |------gmond
| |__ gmetad----|------gmond
| |------gmond
|_gmond
|-gmond
从面可以看到多个gmetad点和多个gmond点
主要是从次节点的配置：
下面对次节点配置如下
gmetad.conf
配置如下：
data_source "test2" localhost ip/hostname
gmond.conf配置文件如下：
cluster {
name = "test2"
owner = "nobody"
latlong = "unspecified"
url = "unspecified"
}
host {
location = "unspecified"
}
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 = 172.16.1.4 ####次节点的ip地址。也就次节点自己的ip地址
port = 8649
}
udp_recv_channel {
port = 8649
family = inet4
}
/* You can specify as many tcp_accept_channels as you like to share
an xml description of the state of the cluster */
tcp_accept_channel {
port = 8649
}
次节点下的gmond节点配置与次节点gmond的配置一样，直接copy一份到gmond上面就好。
关于Ganglia的主要配置到此为止了。Ganglia的views功能还在研究当中，不过这个view添加很麻烦，我配置的好久也，只是在配置文件中添加，在页面上没有添加成功，因为页面上没有添加view的功能，，这个很不爽，网上查了很多资料，只是在官网上找到一点相关E文，，，，，，希路过的大牛能指点一二，多谢！