预计该博文篇幅较长，这里不再废话，对ganglia不太了解的去问谷老师，直接看环境：

hadoop1.updb.com    192.168.0.101

hadoop2.updb.com    192.168.0.102

hadoop3.updb.com    192.168.0.103

hadoop4.updb.com    192.168.0.104

hadoop5.updb.com    192.168.0.105

操作系统：centos 6.5 x86_64，使用自带网络yum源，同时配置epel扩展源。

在安装ganglia之前，确保你的hadoop及hbase已经安装成功，看我的安装规划：

hadoop1.updb.com    NameNode|HMaster|gmetad|gmond|ganglia-web|nagios

hadoop2.updb.com    DataNode|Regionserver|gmond|nrpe

hadoop3.updb.com    DataNode|Regionserver|gmond|nrpe

hadoop4.updb.com    DataNode|Regionserver|gmond|nrpe

hadoop5.updb.com    DataNode|Regionserver|gmond|nrpe

hadoop1作为ganglia和nagios的主控端，安装的软件为ganglia的服务端gmetad、由于要监控自身节点，所以还需要安装ganglia的客户端gmond以及ganglia的web应用ganglia-web和nagios服务端；hadoop2、hadoop3、hadoop4、hadoop5作为被控端，安装的软件有ganglia的客户端gmond以及nagios的客户端nrpe。注意这里的nrpe不是一定要安装的，这里是因为我要监控hadoop2、hadoop3、hadoop4、hadoop5节点上的mysql及其他一些服务，所以选择安装nrpe。

1、hadoop1安装ganglia的gmetad、gmond及ganglia-web

首先安装ganglia所需要的依赖包

[root@hadoop1 ~]# cat ganglia.rpm
apr-devel
apr-util
check-devel
cairo-devel
pango-devel
libxml2-devel
glib2-devel
dbus-devel
freetype-devel
fontconfig-devel
gcc-c++
expat-devel
python-devel
libXrender-devel
zlib
libart_lgpl
libpng
dejavu-lgc-sans-mono-fonts
dejavu-sans-mono-fonts
perl-ExtUtils-CBuilder
perl-ExtUtils-MakeMaker[root@hadoop1 ~]# yum install -y `cat ganglia.rpm`

除了上面的依赖，还需要安装confuse-2.7.tar.gz、rrdtool-1.4.8.tar.gz两个软件

## 解压软件
[root@hadoop1 pub]# tar xf rrdtool-1.4.8.tar.gz  -C /opt/soft/
[root@hadoop1 pub]# tar xf confuse-2.7.tar.gz  -C /opt/soft/
## 安装rrdtool
[root@hadoop1 rrdtool-1.4.8]# ./configure  -prefix=/usr/local/rrdtool
[root@hadoop1 rrdtool-1.4.8]# make && make install
[root@hadoop1 rrdtool-1.4.8]# mkdir  /usr/local/rrdtool/lib64
[root@hadoop1 rrdtool-1.4.8]# cp /usr/local/rrdtool/lib/* /usr/local/rrdtool/lib64/ -rf
[root@hadoop1 rrdtool-1.4.8]# cp /usr/local/rrdtool/lib/librrd.so /usr/lib/
[root@hadoop1 rrdtool-1.4.8]# cp /usr/local/rrdtool/lib/librrd.so /usr/lib64/
## 安装confuse
[root@hadoop1 rrdtool-1.4.8]# cd ../confuse-2.7/
[root@hadoop1 confuse-2.7]# ./configure CFLAGS=-fPIC --disable-nls --prefix=/usr/local/confuse
[root@hadoop1 confuse-2.7]# make && make install
[root@hadoop1 confuse-2.7]# mkdir  /usr/local/confuse/lib64
[root@hadoop1 confuse-2.7]# cp /usr/local/confuse/lib/* /usr/local/confuse/lib64/ -rf

ok，准备工作做好之后，开始安装ganglia软件中的gmetad和gmond

## 解压软件
[root@hadoop1 pub]# tar xf ganglia-3.6.0.tar.gz  -C /opt/soft/
[root@hadoop1 pub]# cd  /opt/soft/ganglia-3.6.0/
## 安装gmetad
[root@hadoop1 ganglia-3.6.0]# ./configure --prefix=/usr/local/ganglia  --with-librrd=/usr/local/rrdtool
--with-libconfuse=/usr/local/confuse --with-gmetad --with-libpcre=no --enable-gexec --enable-status
--sysconfdir=/etc/ganglia
[root@hadoop1 ganglia-3.6.0]# make && make install
[root@hadoop1 ganglia-3.6.0]# cp gmetad/gmetad.init /etc/init.d/gmetad
[root@hadoop1 ganglia-3.6.0]# cp /usr/local/ganglia/sbin/gmetad /usr/sbin/
[root@hadoop1 ganglia-3.6.0]# chkconfig --add gmetad
## 安装gmond
[root@hadoop1 ganglia-3.6.0]# cp gmond/gmond.init /etc/init.d/gmond
[root@hadoop1 ganglia-3.6.0]# cp /usr//local/ganglia/sbin/gmond /usr/sbin/
[root@hadoop1 ganglia-3.6.0]# gmond --default_config>/etc/ganglia/gmond.conf
[root@hadoop1 ganglia-3.6.0]# chkconfig --add gmond

至此，hadoop1上的gmetad、gmond安装成功，接着安装ganglia-web，首先要安装php和httpd

yum install php httpd -y

修改httpd的配置文件/etc/httpd/conf/httpd.conf，只把监听端口改为8080

Listen 8080

安装ganglia-web

[root@hadoop1 pub]# tar xf ganglia-web-3.6.2.tar.gz  -C /opt/soft/
[root@hadoop1 pub]# cd  /opt/soft/
[root@hadoop1 soft]# mv ganglia-web-3.6.2/ /var/www/html/ganglia
[root@hadoop1 soft]# chmod 777 /var/www/html/ganglia -R
[root@hadoop1 soft]# cd /var/www/html/ganglia
[root@hadoop1 ganglia]# useradd www-data
[root@hadoop1 ganglia]# make install
[root@hadoop1 ganglia]# chmod 777 /var/lib/ganglia-web/dwoo/cache/
[root@hadoop1 ganglia]# chmod 777 /var/lib/ganglia-web/dwoo/compiled/

至此ganglia-web安装完成，修改conf_default.php修改文件，指定ganglia-web的目录及rrds的数据目录，修改如下两行：

     36 # Where gmetad stores the rrd archives.37 $conf['gmetad_root'] = "/var/www/html/ganglia"; ## 改为web程序的安装目录38 $conf['rrds'] = "/var/lib/ganglia/rrds";        ## 指定rrd数据存放的路径

创建rrd数据存放目录并授权

[root@hadoop1 ganglia]# mkdir /var/lib/ganglia/rrds -p
[root@hadoop1 ganglia]# chown nobody:nobody /var/lib/ganglia/rrds/ -R

到这里，hadoop1上的ganglia的所有安装工作就完成了，接下来就是要在hadoop2、hadoop3、hadoop4、hadoop5上安装ganglia的gmond客户端。

2、在hadoop2、hadoop3、hadoop4、hadoop5上安装gmond

首先还是需要安装依赖，参照hadoop1中的前两步来安装所需依赖

ok，准备工作做好之后，开始安装gmond,4个节点的操作是一样的，这里以hadoop2为例

## 解压软件
[root@hadoop2 pub]# tar xf ganglia-3.6.0.tar.gz  -C /opt/soft/
[root@hadoop2 pub]# cd  /opt/soft/ganglia-3.6.0/
## 安装gmond，注意这里的编译和gmetad相比少了--with-gmetad
[root@hadoop2 ganglia-3.6.0]# ./configure --prefix=/usr/local/ganglia  --with-librrd=/usr/local/rrdtool
--with-libconfuse=/usr/local/confuse --with-libpcre=no --enable-gexec --enable-status
--sysconfdir=/etc/ganglia
[root@hadoop2 ganglia-3.6.0]# make && make install
[root@hadoop2 ganglia-3.6.0]# cp gmond/gmond.init /etc/init.d/gmond
[root@hadoop2 ganglia-3.6.0]# cp /usr//local/ganglia/sbin/gmond /usr/sbin/
[root@hadoop2 ganglia-3.6.0]# gmond --default_config>/etc/ganglia/gmond.conf
[root@hadoop2 ganglia-3.6.0]# chkconfig --add gmond

到这里hadoop2上的gmond已经安装成功，hadoop3、hadoop4、hadoop5依次安装成功。

3、配置ganglia，分为服务端和客户端的配置，服务端的配置文件为gmetad.conf,客户端的配置文件为gmond.conf

首先配置hadoop1上的gmetad.conf

[root@hadoop1 ~]# vi  /etc/ganglia/gmetad.conf
## 定义数据源的名字及监听地址，gmond会将收集的数据发送到数据源监听机器上的rrd数据目录中
data_source "hadoop cluster" 192.168.0.101:8649

gmetad.conf的配置相当的简单，注意gmetad.conf只有hadoop1上有，因此只在hadoop1上配置。接着配置hadoop1上的gmond.conf

[root@hadoop1 ~]# head -n 80  /etc/ganglia/gmond.conf
/* This configuration is as close to 2.5.x default behavior as possibleThe values closely match ./gmond/metric.h definitions in 2.5.x */
globals {daemonize = yes        ## 以守护进程运行setuid = yes           user = nobody          ## 运行gmond的用户debug_level = 0        ## 改为1会在启动时打印debug信息max_udp_msg_len = 1472mute = no              ## 哑巴，本节点将不会再广播任何自己收集到的数据到网络上deaf = no              ## 聋子，本节点将不再接收任何其他节点广播的数据包allow_extra_data = yeshost_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 reasonablesend_metadata_interval = 0 /*secs */}/** The cluster attributes specified will be used as part of the <CLUSTER>* tag that will wrap all hosts collected by this instance.*/
cluster {name = "hadoop cluster"    ## 指定集群的名字owner = "nobody"           ## 集群的所有者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 as you like.  Gmondused to only support 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    ## 单播模式要注释调这行host = 192.168.0.101    ## 单播模式，指定接受数据的主机port = 8649             ## 监听端口ttl = 1
}/* You can specify as many udp_recv_channels as you like as well. */
udp_recv_channel {#mcast_join = 239.2.11.71    ## 单播模式要注释调这行port = 8649#bind = 239.2.11.71          ## 单播模式要注释调这行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 sharean xml description of the state of the cluster */
tcp_accept_channel {port = 8649# If you want to gzip XML outputgzip_output = no
}/* Channel to receive sFlow datagrams */
#udp_recv_channel {
#  port = 6343
#}/* Optional sFlow settings */

好了，hadoop1上的gmetad.conf和gmond.conf配置文件已经修改完成，这时，直接将hadoop1上的gmond.conf文件scp到hadoop2、hadoop3、hadoop4、hadoop5上相同的路径下覆盖原来的gmond.conf即可。

4、启动服务

启动hadoop1、hadoop2、hadoop3、hadoop4、hadoop5上的gmond服务

/etc/init.d/gmond start

启动hadoop1上的httpd服务和gmetad服务

/etc/init.d/gmetad start
/etc/init.d/httpd start

5、在浏览器中访问192.168.0.101:8080/ganglia,就会出现下面的页面

但此时，ganglia只是监控了各主机基本的性能，并没有监控到hadoop和hbase，接下来需要配置hadoop和hbase的配置文件，这里以hadoop1上的配置文件为例，其他节点对应的配置文件应从hadoop1上拷贝，首先需要修改的是hadoop配置目录下的hadoop-metrics2.properties

[root@hadoop1 ~]# cd /opt/hadoop-2.4.1/etc/hadoop/
[root@hadoop1 hadoop]# cat hadoop-metrics2.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.
## syntax: [prefix].[source|sink].[instance].[options]
# See javadoc of package-info.java for org.apache.hadoop.metrics2 for details#*.sink.file.class=org.apache.hadoop.metrics2.sink.FileSink
# default sampling period, in seconds
#*.period=10# The namenode-metrics.out will contain metrics from all context
#namenode.sink.file.filename=namenode-metrics.out
# Specifying a special sampling period for namenode:
#namenode.sink.*.period=8#datanode.sink.file.filename=datanode-metrics.out# the following example split metrics of different
# context to different sinks (in this case files)
#jobtracker.sink.file_jvm.context=jvm
#jobtracker.sink.file_jvm.filename=jobtracker-jvm-metrics.out
#jobtracker.sink.file_mapred.context=mapred
#jobtracker.sink.file_mapred.filename=jobtracker-mapred-metrics.out#tasktracker.sink.file.filename=tasktracker-metrics.out#maptask.sink.file.filename=maptask-metrics.out#reducetask.sink.file.filename=reducetask-metrics.out*.sink.ganglia.class=org.apache.hadoop.metrics2.sink.ganglia.GangliaSink31
*.sink.ganglia.period=10
*.sink.ganglia.supportsparse=true
*.sink.ganglia.slope=jvm.metrics.gcCount=zero,jvm.metrics.memHeapUsedM=both
*.sink.ganglia.dmax=jvm.metrics.threadsBlocked=70,jvm.metrics.memHeapUsedM=40
namenode.sink.ganglia.servers=192.168.0.101:8649
datanode.sink.ganglia.servers=192.168.0.101:8649
resourcemanager.sink.ganglia.servers=192.168.0.101:8649
secondarynamenode.sink.ganglia.servers=192.168.0.101:8649
nodemanager.sink.ganglia.servers=192.168.0.101:8649

接着需要修改hbase配置目录下的hadoop-metrics2-hbase.properties

[root@hadoop1 hadoop]# cd /opt/hbase-0.98.4-hadoop2/conf/
[root@hadoop1 conf]# cat hadoop-metrics2-hbase.properties
# syntax: [prefix].[source|sink].[instance].[options]
# See javadoc of package-info.java for org.apache.hadoop.metrics2 for details#*.sink.file*.class=org.apache.hadoop.metrics2.sink.FileSink
# default sampling period
#*.period=10# Below are some examples of sinks that could be used
# to monitor different hbase daemons.# hbase.sink.file-all.class=org.apache.hadoop.metrics2.sink.FileSink
# hbase.sink.file-all.filename=all.metrics# hbase.sink.file0.class=org.apache.hadoop.metrics2.sink.FileSink
# hbase.sink.file0.context=hmaster
# hbase.sink.file0.filename=master.metrics# hbase.sink.file1.class=org.apache.hadoop.metrics2.sink.FileSink
# hbase.sink.file1.context=thrift-one
# hbase.sink.file1.filename=thrift-one.metrics# hbase.sink.file2.class=org.apache.hadoop.metrics2.sink.FileSink
# hbase.sink.file2.context=thrift-two
# hbase.sink.file2.filename=thrift-one.metrics# hbase.sink.file3.class=org.apache.hadoop.metrics2.sink.FileSink
# hbase.sink.file3.context=rest
# hbase.sink.file3.filename=rest.metrics
*.sink.ganglia.class=org.apache.hadoop.metrics2.sink.ganglia.GangliaSink31
*.sink.ganglia.period=10
hbase.sink.ganglia.period=10
hbase.sink.ganglia.servers=192.168.0.101:8649

将hadoop1上的这两个文件，scp到hadoop2~5这4个节点上相同的路径下，覆盖原来的文件，然后重启hadoop、hbase，这时ganglia就能够监控到hadoop和hbase了，如下如

到这里位置ganglia已经完全安装配置完成了，且已经成功的监控到了hadoop和hbase。

接下来，安装nagios，在hadoop1上安装nagios服务端、在hadoop2、hadoop3、hadoop4、hadoop5上安装客户端nrpe。

首先在hadoop1上安装nagios、及相关插件

yum install nagios nagios-plugins nagios-plugins-all nagios-plugins-nrpe -y

设置nagios web界面的登录口令

[root@hadoop1 ~]# cat /etc/httpd/conf.d/nagios.conf
# SAMPLE CONFIG SNIPPETS FOR APACHE WEB SERVER
# Last Modified: 11-26-2005
#
# This file contains examples of entries that need
# to be incorporated into your Apache web server
# configuration file.  Customize the paths, etc. as
# needed to fit your system.ScriptAlias /nagios/cgi-bin/ "/usr/lib64/nagios/cgi-bin/"<Directory "/usr/lib64/nagios/cgi-bin/">
#  SSLRequireSSLOptions ExecCGIAllowOverride NoneOrder allow,denyAllow from all
#  Order deny,allow
#  Deny from all
#  Allow from 127.0.0.1
## 这里的用户名必须是nagiosadminAuthName "nagiosadmin"AuthType Basic
## 这里指定密码文件的路径AuthUserFile /etc/nagios/htpasswd.usersRequire valid-user
</Directory>Alias /nagios "/usr/share/nagios/html"<Directory "/usr/share/nagios/html">
#  SSLRequireSSLOptions NoneAllowOverride NoneOrder allow,denyAllow from all
#  Order deny,allow
#  Deny from all
#  Allow from 127.0.0.1
## 这里的用户名必须是nagiosadminAuthName "nagiosadmin"AuthType Basic
## 这里指定密码文件的路径AuthUserFile /etc/nagios/htpasswd.usersRequire valid-user
</Directory>

保存，退出，生成密码文件

[root@hadoop1 ~]#  htpasswd -c /etc/nagios/htpasswd.users nagiosadmin
New password:
Re-type new password:
Adding password for user nagiosadmin [root@hadoop1 ~]# cat   /etc/nagios/htpasswd.users
nagiosadmin:qWrXYKDlycqHM

生成密码成功，接着在hadoop2、hadoop3、hadoop4、hadoop5上安装客户端nrpe及相关插件

yum install nagios-plugins nagios-plugins-nrpe nrpe nagios-plugins-all -y

所有节点的插件均位于/usr/lib64/nagios/plugins/下

[root@hadoop2 ~]# ls /usr/lib64/nagios/plugins/
check_breeze    check_game       check_mrtgtraf     check_overcr   check_swap
check_by_ssh    check_hpjd       check_mysql        check_pgsql    check_tcp
check_clamd     check_http       check_mysql_query  check_ping     check_time
check_cluster   check_icmp       check_nagios       check_pop      check_udp
check_dhcp      check_ide_smart  check_nntp         check_procs    check_ups
check_dig       check_imap       check_nntps        check_real     check_users
check_disk      check_ircd       check_nrpe         check_rpc      check_wave
check_disk_smb  check_jabber     check_nt           check_sensors  negate
check_dns       check_ldap       check_ntp          check_simap    urlize
check_dummy     check_ldaps      check_ntp_peer     check_smtp     utils.pm
check_file_age  check_load       check_ntp.pl       check_snmp     utils.sh
check_flexlm    check_log        check_ntp_time     check_spop
check_fping     check_mailq      check_nwstat       check_ssh
check_ftp       check_mrtg       check_oracle       check_ssmtp

要想让nagios与ganglia整合起来，就需要在hadoop1上把ganglia安装包中的ganglia的插件放到nagios的插件目录下

[root@hadoop1 ~]# cd /opt/soft/ganglia-3.6.0/
[root@hadoop1 ganglia-3.6.0]# ls contrib/check_ganglia.py
contrib/check_ganglia.py
[root@hadoop1 ganglia-3.6.0]# cp contrib/check_ganglia.py /usr/lib64/nagios/plugins/

默认的check_ganglia.py 插件中只有监控项的实际值大于critical阀值的情况，这里需要增加监控项的实际值小于critical阀值的情况，即最后添加的一段代码

[root@hadoop1 plugins]# cat check_ganglia.py
#!/usr/bin/env pythonimport sys
import getopt
import socket
import xml.parsers.expatclass GParser:def __init__(self, host, metric):self.inhost =0self.inmetric = 0self.value = Noneself.host = hostself.metric = metricdef parse(self, file):p = xml.parsers.expat.ParserCreate()p.StartElementHandler = parser.start_elementp.EndElementHandler = parser.end_elementp.ParseFile(file)if self.value == None:raise Exception('Host/value not found')return float(self.value)def start_element(self, name, attrs):if name == "HOST":if attrs["NAME"]==self.host:self.inhost=1elif self.inhost==1 and name == "METRIC" and attrs["NAME"]==self.metric:self.value=attrs["VAL"]def end_element(self, name):if name == "HOST" and self.inhost==1:self.inhost=0def usage():print """Usage: check_ganglia \
-h|--host= -m|--metric= -w|--warning= \
-c|--critical= [-s|--server=] [-p|--port=] """sys.exit(3)if __name__ == "__main__":
##############################################################ganglia_host = '192.168.0.101'ganglia_port = 8649host = Nonemetric = Nonewarning = Nonecritical = Nonetry:options, args = getopt.getopt(sys.argv[1:],"h:m:w:c:s:p:",["host=", "metric=", "warning=", "critical=", "server=", "port="],)except getopt.GetoptError, err:print "check_gmond:", str(err)usage()sys.exit(3)for o, a in options:if o in ("-h", "--host"):host = aelif o in ("-m", "--metric"):metric = aelif o in ("-w", "--warning"):warning = float(a)elif o in ("-c", "--critical"):critical = float(a)elif o in ("-p", "--port"):ganglia_port = int(a)elif o in ("-s", "--server"):ganglia_host = aif critical == None or warning == None or metric == None or host == None:usage()sys.exit(3)try:s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)s.connect((ganglia_host,ganglia_port))parser = GParser(host, metric)value = parser.parse(s.makefile("r"))s.close()except Exception, err:print "CHECKGANGLIA UNKNOWN: Error while getting value \"%s\"" % (err)sys.exit(3)if critical > warning:if value >= critical:print "CHECKGANGLIA CRITICAL: %s is %.2f" % (metric, value)sys.exit(2)elif value >= warning:print "CHECKGANGLIA WARNING: %s is %.2f" % (metric, value)sys.exit(1)else:print "CHECKGANGLIA OK: %s is %.2f" % (metric, value)sys.exit(0)else:if critical >=value:print "CHECKGANGLIA CRITICAL: %s is %.2f" % (metric, value)sys.exit(2)elif warning >=value:print "CHECKGANGLIA WARNING: %s is %.2f" % (metric, value)sys.exit(1)else:print "CHECKGANGLIA OK: %s is %.2f" % (metric, value)sys.exit(0)

配置hadoop2、hadoop3、hadoop4、hadoop5上的nrpe客户端，这里以hadoop2为例演示，其他节点直接从hadoop2上scp，然后覆盖相同路径的下的文件即可

[root@hadoop2 ~]# cat /etc/nagios/nrpe.cfg
log_facility=daemon
pid_file=/var/run/nrpe/nrpe.pid
## nagios的监听端口
server_port=5666
nrpe_user=nrpe
nrpe_group=nrpe
## nagios服务器主机地址
allowed_hosts=192.168.0.101
dont_blame_nrpe=0
allow_bash_command_substitution=0
debug=0
command_timeout=60
connection_timeout=300## 监控负载
command[check_load]=/usr/lib64/nagios/plugins/check_load -w 15,10,5 -c 30,25,20
## 当前系统用户数
command[check_users]=/usr/lib64/nagios/plugins/check_users -w 5 -c 10
## 根分区空闲容量
command[check_sda2]=/usr/lib64/nagios/plugins/check_disk -w 20% -c 10% -p /dev/sda2
## mysql状态
command[check_mysql]=/usr/lib64/nagios/plugins/check_mysql -H hadoop2.updb.com -P 3306 -d kora -u kora -p upbjsxt
## 主机是否存活
command[check_ping]=/usr/lib64/nagios/plugins/check_ping -H hadoop2.updb.com -w  100.0,20% -c  500.0,60%
## 当前系统的进程总数
command[check_total_procs]=/usr/lib64/nagios/plugins/check_procs -w 150 -c 200include_dir=/etc/nrpe.d/

scp该文件到hadoop3、hadoop4、hadoop5上相同路径下覆盖源文件，注意要将文件中的主机名改为对应的主机名。

hadoop1上配置各个主机及对应的监控项，有如下配置文件

[root@hadoop1 plugins]# cd /etc/nagios/objects/
## 每个节点对应一个host文件和一个监控项文件，如hadoop2对应的是hadoop2.cfg和service2.cfg
[root@hadoop1 objects]# ls
commands.cfg  hadoop3.cfg  localhost.cfg  service3.cfg  templates.cfg
contacts.cfg  hadoop4.cfg  printer.cfg    service4.cfg  timeperiods.cfg
hadoop1.cfg   hadoop5.cfg  service1.cfg   service5.cfg  windows.cfg
hadoop2.cfg   hosts.cfg    service2.cfg   switch.cfg

首先在commond.cfg中声明check_ganglia、check_nrpe命令，在文件最后追加如下内容

# 'check_ganglia' command definition
define command{command_name    check_gangliacommand_line    $USER1$/check_ganglia.py -h $HOSTADDRESS$ -m $ARG1$ -w $ARG2$ -c $ARG3$}# 'check_nrpe' command definition
define command{command_name    check_nrpecommand_line    $USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$}

然后修改templates.cfg模版配置文件，在最后追加如下内容

define service { use generic-service name ganglia-service1    ## 这里的配置在service1.cfg中用到hostgroup_name hadoop1    ## 这里的配置在hadoop1.cfg中用到service_groups ganglia-metrics1    ## 这里的配置在service1.cfg中用到register        0
}define service { use generic-service    name ganglia-service2    ## 这里的配置在service2.cfg中用到 hostgroup_name hadoop2    ## 这里的配置在hadoop2.cfg中用到service_groups ganglia-metrics2    ## 这里的配置在service2.cfg中用到register        0
}
define service { use generic-service name ganglia-service3    ## 这里的配置在service3.cfg中用到 hostgroup_name hadoop3    ## 这里的配置在hadoop3.cfg中用到service_groups ganglia-metrics3    ## 这里的配置在service3.cfg中用到register        0
}
define service { use generic-service name ganglia-service4    ## 这里的配置在service4.cfg中用到 hostgroup_name hadoop4    ## 这里的配置在hadoop4.cfg中用到service_groups ganglia-metrics4    ## 这里的配置在service4.cfg中用到register        0
}
define service { use generic-service     name ganglia-service5    ## 这里的配置在service5.cfg中用到     hostgroup_name hadoop5    ## 这里的配置在hadoop5.cfg中用到    service_groups ganglia-metrics5    ## 这里的配置在service5.cfg中用到register        0
}

hadoop1的配置如下,由于hadoop1是服务端，无需使用nrpe来监控自己，配置如下

## hadoop1.cfg中的监控项为常规的本机监控项，而service1.cfg中的监控项为ganglia的监控项
[root@hadoop1 objects]# cat hadoop1.cfg
define host{   use                     linux-server host_name               hadoop1.updb.comalias                   hadoop1.updb.comaddress                 hadoop1.updb.com}define hostgroup { hostgroup_name  hadoop1alias  hadoop1members hadoop1.updb.com}
define service{use                             local-servicehost_name                       hadoop1.updb.comservice_description             PINGcheck_command                   check_ping!100,20%!500,60%}define service{use                             local-servicehost_name                       hadoop1.updb.comservice_description             根分区check_command                   check_local_disk!20%!10%!/
#       contact_groups                  admins}define service{use                             local-servicehost_name                       hadoop1.updb.comservice_description             用户数量check_command                   check_local_users!20!50}define service{use                             local-servicehost_name                       hadoop1.updb.comservice_description             进程数check_command                   check_local_procs!250!400!RSZDT}define service{ use                             local-service         host_name                       hadoop1.updb.comservice_description             系统负载check_command                   check_local_load!5.0,4.0,3.0!10.0,6.0,4.0
} ## services.cfg
[root@hadoop1 objects]# cat  service1.cfg
define servicegroup { servicegroup_name ganglia-metrics1alias Ganglia Metrics1
}
## 这里的check_ganglia为commonds.cfg中声明的check_ganglia命令
define service{ use                             ganglia-service1 service_description             HMaster负载check_command                   check_ganglia!master.Server.averageLoad!5!10
} define service{ use                             ganglia-service1service_description             内存空闲check_command                   check_ganglia!mem_free!200!50
} define service{use                             ganglia-service1service_description             NameNode同步check_command                   check_ganglia!dfs.namenode.SyncsAvgTime!10!50
}

hadoop2的配置如下，需要注意使用check_nrpe插件的监控项必须要在hadoop2上的nrpe.cfg中声明

## 这里的监控项就使用了远程客户端节点上的nrpe来收集数据并周期性的发送给hadoop1的nagios server
[root@hadoop1 objects]# cat hadoop2.cfg
define host{   use                     linux-server host_name               hadoop2.updb.comalias                   hadoop2.updb.comaddress                 hadoop2.updb.com}define hostgroup { hostgroup_name  hadoop2alias  hadoop2members hadoop2.updb.com}
## 这里的check_nrpe为commonds.cfg中声明的check_nrpe
define service{use                             local-service         host_name                       hadoop2.updb.comservice_description             Mysql状态check_command                   check_nrpe!check_mysql}define service{use                             local-servicehost_name                       hadoop2.updb.comservice_description             PINGcheck_command                   check_nrpe!check_ping}define service{use                             local-servicehost_name                       hadoop2.updb.comservice_description             根分区check_command                   check_nrpe!check_sda2}define service{use                             local-servicehost_name                       hadoop2.updb.comservice_description             用户数量check_command                   check_nrpe!check_users}define service{use                             local-servicehost_name                       hadoop2.updb.comservice_description             进程数check_command                   check_nrpe!check_total_procs}define service{ use                             local-service         host_name                       hadoop2.updb.comservice_description             系统负载check_command                   check_nrpe!check_load
} ## 这里的监控项为ganglia的监控项使用check_ganglia插件
[root@hadoop1 objects]# cat service2.cfg
define servicegroup {servicegroup_name ganglia-metrics2alias Ganglia Metrics2
}define service{ use                             ganglia-service2 service_description             内存空闲check_command                   check_ganglia!mem_free!200!50
} define service{ use                             ganglia-service2 service_description             RegionServer_Getcheck_command                   check_ganglia!regionserver.Server.Get_min!5!15
} define service{ use                             ganglia-service2service_description             DateNode_Heartbeatcheck_command                   check_ganglia!dfs.datanode.HeartbeatsAvgTime!15!40
}

hadoop3、hadoop4、hadoop5的配置与hadoop2一样，除了主机名要改之外。

最后，还要将这些配置include到nagios的主配置文件中去

## 涉及修改的内容如下，其他的保持不变即可
[root@hadoop1 objects]# vi  ../nagios.cfg
# You can specify individual object config files as shown below:
#cfg_file=/etc/nagios/objects/localhost.cfg
cfg_file=/etc/nagios/objects/commands.cfg
cfg_file=/etc/nagios/objects/contacts.cfg
cfg_file=/etc/nagios/objects/timeperiods.cfg
cfg_file=/etc/nagios/objects/templates.cfg
## 将host文件引入进来
cfg_file=/etc/nagios/objects/hadoop1.cfg
cfg_file=/etc/nagios/objects/hadoop2.cfg
cfg_file=/etc/nagios/objects/hadoop3.cfg
cfg_file=/etc/nagios/objects/hadoop4.cfg
cfg_file=/etc/nagios/objects/hadoop5.cfg
## 将监控项的文件引入进来
cfg_file=/etc/nagios/objects/service1.cfg
cfg_file=/etc/nagios/objects/service2.cfg
cfg_file=/etc/nagios/objects/service3.cfg
cfg_file=/etc/nagios/objects/service4.cfg
cfg_file=/etc/nagios/objects/service5.cfg

接下来验证，配置项是否正确

[root@hadoop1 objects]# nagios -v ../nagios.cfg Nagios Core 3.5.1
Copyright (c) 2009-2011 Nagios Core Development Team and Community Contributors
Copyright (c) 1999-2009 Ethan Galstad
Last Modified: 08-30-2013
License: GPLWebsite: http://www.nagios.org
Reading configuration data...Read main config file okay...
Processing object config file '/etc/nagios/objects/commands.cfg'...
Processing object config file '/etc/nagios/objects/contacts.cfg'...
Processing object config file '/etc/nagios/objects/timeperiods.cfg'...
Processing object config file '/etc/nagios/objects/templates.cfg'...
Processing object config file '/etc/nagios/objects/hadoop1.cfg'...
Processing object config file '/etc/nagios/objects/hadoop2.cfg'...
Processing object config file '/etc/nagios/objects/hadoop3.cfg'...
Processing object config file '/etc/nagios/objects/hadoop4.cfg'...
Processing object config file '/etc/nagios/objects/hadoop5.cfg'...
Processing object config file '/etc/nagios/objects/service1.cfg'...
Processing object config file '/etc/nagios/objects/service2.cfg'...
Processing object config file '/etc/nagios/objects/service3.cfg'...
Processing object config file '/etc/nagios/objects/service4.cfg'...
Processing object config file '/etc/nagios/objects/service5.cfg'...
Processing object config directory '/etc/nagios/conf.d'...Read object config files okay...Running pre-flight check on configuration data...Checking services...Checked 44 services.
Checking hosts...Checked 5 hosts.
Checking host groups...Checked 5 host groups.
Checking service groups...Checked 5 service groups.
Checking contacts...Checked 1 contacts.
Checking contact groups...Checked 1 contact groups.
Checking service escalations...Checked 0 service escalations.
Checking service dependencies...Checked 0 service dependencies.
Checking host escalations...Checked 0 host escalations.
Checking host dependencies...Checked 0 host dependencies.
Checking commands...Checked 26 commands.
Checking time periods...Checked 5 time periods.
Checking for circular paths between hosts...
Checking for circular host and service dependencies...
Checking global event handlers...
Checking obsessive compulsive processor commands...
Checking misc settings...Total Warnings: 0
Total Errors:   0Things look okay - No serious problems were detected during the pre-flight check

ok，没有错误，这时就可以启动hadoop1上的nagios服务

[root@hadoop1 objects]# /etc/init.d/nagios start
Starting nagios: done.

启动hadoop2-5上的nrpe服务

[root@hadoop2 ~]# /etc/init.d/nrpe   start
Starting nrpe:                                             [  OK  ]

在hadoop1上测试nagios与nrpe通信是否正常

[root@hadoop1 objects]# cd /usr/lib64/nagios/plugins/
[root@hadoop1 plugins]# ./check_nrpe -H hadoop1.updb.com
[root@hadoop1 plugins]# ./check_nrpe -H hadoop2.updb.com
NRPE v2.15
[root@hadoop1 plugins]# ./check_nrpe -H hadoop3.updb.com
NRPE v2.15
[root@hadoop1 plugins]# ./check_nrpe -H hadoop4.updb.com
NRPE v2.15
[root@hadoop1 plugins]# ./check_nrpe -H hadoop5.updb.com
NRPE v2.15

ok，通信正常，验证check_ganglia.py插件是否工作正常

[root@hadoop1 ~]# cd /usr/lib64/nagios/plugins/
[root@hadoop1 plugins]# ./check_ganglia.py -h hadoop2.updb.com -m mem_free -w 200 -c 50
CHECKGANGLIA OK: mem_free is 72336.00

工作正常，现在我们可以nagios的web页面，看是否监控成功。

可以看到已经成功监控，如果需要监控hadoop、hbase更多更加详细的监控项，可以在这里查找，nagios中配置监控项的名字和这里出现的一致即可

你会发现，ganglia中出现的可监控项非常的多，只要你愿意，你可以监控任何你想要监控的集群选项，前提是要弄明白这些监控项的含义，以及阀值设置为多少最为合适，显然这才是真正的挑战，实验中的集群监控项我是随便选取的，并没有特别的含义，因为鄙人也还是屌丝一枚，正在慢慢学习这些监控项，好了，到这里nagios已经成功监控到了我们想要监控的内容了，那么怎么实现手机短信报警呢？

看下面，首先我们要配置contacts.cfg，添加邮件的接受人

[root@hadoop1 plugins]# cd /etc/nagios/objects/
[root@hadoop1 objects]# cat contacts.cfg
define contact{contact_name                    nagiosadminuse                             generic-contactalias                           Nagios Admin## 告警时间段service_notification_period     24x7host_notification_period        24x7## 告警信息格式service_notification_options    w,u,c,r,f,shost_notification_options       d,u,r,f,s## 告警方式为邮件service_notification_commands   notify-service-by-email                            host_notification_commands      notify-host-by-email ## 联系人的139邮箱email                           1820280----@139.com}## 定义联系人所属组
define contactgroup{contactgroup_name       adminsalias                   Nagios Administratorsmembers                 nagiosadmin}

重启nagios服务，要保证你的139邮箱开启了长短信转发格式，确保邮件到达邮箱后会转发到你的手机，这里上一张手机接受到邮件后的截图：

好了，大功告成。看到此刻的你，是否犹如孙悟空的精魂刚刚回归肉体，与周围的世界有片刻的神离。哈哈，好好学习ganglia的监控项，以后嘛嘛就再也不用担心你的运维工作了，上班时间就是喝茶看新闻了，坐等短信告诉你哪个地方有问题。