通过nagios监控weblogic服务

1.前言

本文主要介绍如何通过Nagios软件来监控Weblogic服务运行状况，其中主要包括Weblogic Server以及Weblogic JDBC Pool的运行状态。Nagios的插件中本身并不提供对于Weblogic服务监控的功能，所以要根据Nagios Plugin API编写自己的脚本，扩展其插件，完成我们所需要的功能。对于Weblogic运行状态信息的获得需通过JMX。

本文参考了Nagios3的官方文档中有关Nagios Plugin部分，以及Weblogic官方文档有关JMX和命令行部分，具体的Weblogic版本是8.14。

2.Nagios Plugin API概述

作为一个Nagios插件，无论你是用脚本（如shell、perl）还是用c编译后的可执行程序实现，它必须至少完成两件事，

1、退出时有一个返回值。

2、至少向标准输出设备（STDOUT）输出一行文本。

返回值定义：

Plugin Return Code Service State Host State

0 OK UP

1 WARNING UP or DOWN/UNREACHABLE*

2 CRITICAL DOWN/UNREACHABLE

3 UNKNOWN DOWN/UNREACHABLE

输出文本至少要一行，其信息主要反映被监控应用、服务的状态。

例如：DISK OK - free space: / 3326 MB (56%);

3.监控Weblogic的实现方法

对于Weblogic运行状况的获得，我们是通过命令行的方式实现的，通过调用Weblogic 的weblogic.Admin类实现的。这个类的功能很强大，可以通过它管理和配置Weblogic。

以下介绍几个常用的命令写法。

1、获得server运行状态

2、获得JDBC Pool运行状态

将黄色标记部分的变量替换成相应真实环境值即可。

${URL}weblogic的URL，例如t3://192.168.1.2:7002

${USER_NAME}用户名

${PASS_WORD}密码

${DOMAIN_NAME}weblogic域的名称，如mydomain

${SERVER_NAME}Server名

${POOL_NAME}JDBC Pool名称

在运行上述命令前需要设置JAVA_HOME，并且将$JAVA_HOME/bin添加到PATH中，将weblogic的weblogic81/server/lib/weblogic.jar包添加到CLASSPATH中。

4.具体实现的shell脚本

有了监控的方法，根据Nagios Plugin API规则编写自己的shell实现脚本。具体的shell脚本如下：

check_wls.sh

#!/bin/ksh

#check_wls.sh --jdbcpool url username password domainname servername poolname

#check_wls.sh --server url username password domainname servername

PROGNAME=`basename $0`

PROGPATH=`echo $0 | sed -e 's,[\\\\/][^\\\\/][^\\\\/]*$,,'`

REVISION=`echo '$Revision: 1749 $' | sed -e 's/[^0-9.]//g'`

. $PROGPATH/utils.sh

print_usage() {

echo "Usage:"

echo " $PROGNAME --jdbcpool url username password domainname servername poolname

echo " $PROGNAME --server url username password domainname servername

echo " $PROGNAME --help"

echo " $PROGNAME --version"

}

print_revision $PROGNAME $REVISION

echo ""

print_usage

echo ""

echo "Check Weblogic status"

echo ""

echo "--jdbcpool url username password domainname servername poolname" echo " Check Weblogic JDBC Pool"

echo "--server url username password domainname servername"

echo " Check Weblogic Server"

}

if [[ -z "$JAVA_HOME" ]]

then

echo "Please set JAVA_HOME!"

exit $STATE_UNKNOWN

fi

if [[ -z "$CLASSPATH" ]]

then

echo "Please set CLASSPATH!"

exit $STATE_UNKNOWN

else

echo $CLASSPATH | grep "weblogic.jar" | wc -l | read N

if [[ "$N" = "0" ]]

then

echo "Please add weblogic.jar to CLASSPATH!"

exit $STATE_UNKNOWN

fi

fi

PATH=$JAVA_HOME/bin:$PATH

export PATH

JDBC_TYPE="JDBCConnectionPoolRuntime"

SERVER_TYPE="ServerRuntime"

cmd="$1"

# Information options

case "$cmd" in

--help)

print_help

exit $STATE_OK

;;

-h)

print_help

exit $STATE_OK

;;

--version)

print_revision $PROGNAME $REVISIONexit $STATE_OK

;;

-V)

print_revision $PROGNAME $REVISION

exit $STATE_OK

;;

esac

case "$cmd" in

--server)

URL=${2}

USER_NAME=${3}

PASS_WORD=${4}

DOMAIN_NAME=${5}

SERVER_NAME=${6}

SERVER_INFO="${DOMAIN_NAME}:${SERVER_NAME}"

RE=`java weblogic.Admin -url ${URL} -username ${USER_NAME} -password ${PASS_WORD} get -pretty \

-mbean "${DOMAIN_NAME}:Location=${SERVER_NAME},Name=${SERVER_NAME},Type=$ {SERVER_TYPE}"`

printf "${RE}" | grep ^"-" | wc -l | read N

if [[ "$N" -lt "1" ]]

then

#error

printf "${RE}" | awk '{ printf $0 }' | read ERR_INFO

echo "CRITICAL - ${ERR_INFO}"

exit $STATE_CRITICAL

fi

if [[ "$N" -ge "1" ]]

then

HEALTH_STATE=""

RUN_STATE=""

#HealthState State

printf "${RE}" | while read NAME VALUE

do

#PoolState WaitingForConnectionCurrentCount State

#echo "NAME:${NAME} VALUE:${VALUE}"

case "${NAME}" in

HealthState:)

HEALTH_STATE=${VALUE}

;;

State:)

RUN_STATE=${VALUE}

;;

esac

done

#echo "HEALTH_STATE:${HEALTH_STATE}"

#echo "RUN_STATE:${RUN_STATE}"HEALTH_STATE_INFO=${HEALTH_STATE}

echo ${HEALTH_STATE_INFO} | awk -F, '{ print $1 }' | awk -F: '{ print $2 }' | read HEALTH_STATE

#echo "HEALTH_STATE:${HEALTH_STATE}"

#HEALTH_OK HEALTH_WARN HEALTH_CRITICAL HEALTH_FAILED

if [[ "${RUN_STATE}" != "RUNNING" ]]

then

echo "CRITICAL - ${SERVER_INFO} State is ${RUN_STATE}"

exit $STATE_CRITICAL

fi

case "${HEALTH_STATE}" in

EALTH_OK)

;;

HEALTH_WARN)

echo "WARN - ${SERVER_INFO} HealthState is ${HEALTH_STATE_INFO}"

exit $STATE_WARNING

;;

HEALTH_CRITICAL)

echo "CRITICAL - ${SERVER_INFO} HealthState is ${HEALTH_STATE_INFO}"

exit $STATE_CRITICAL

;;

HEALTH_FAILED)

echo "FAILED - ${SERVER_INFO} HealthState is ${HEALTH_STATE_INFO}"

exit $STATE_CRITICAL

;;

esac

fi

echo "OK - ${SERVER_INFO} State is ${RUN_STATE},HealthState is ${HEALTH_STATE_INFO}"

exit $STATE_OK

;;

--jdbcpool)

URL=${2}

USER_NAME=${3}

PASS_WORD=${4}

DOMAIN_NAME=${5}

SERVER_NAME=${6}

POOL_NAME=${7}

POOL_INFO="${DOMAIN_NAME}:${SERVER_NAME}:${POOL_NAME}"

RE=`java weblogic.Admin -url ${URL} -username ${USER_NAME} -password ${PASS_WORD} GET

-pretty \

-mbean "${DOMAIN_NAME}:Location=${SERVER_NAME},Name=${POOL_NAME},ServerRuntime=$ {SERVER_NAME},Type=${JDBC_TYPE}"`

printf "${RE}" | grep ^"-" | wc -l | read N

if [[ "$N" -lt "1" ]]then

#error

printf "${RE}" | awk '{ printf $0 }' | read ERR_INFO

echo "CRITICAL - ${ERR_INFO}"

exit $STATE_CRITICAL

fi

if [[ "$N" -ge "1" ]]

then

POOL_STATE=""

WAIT_CNT=""

RUN_STATE=""

printf "${RE}" | while read NAME VALUE

do

#PoolState WaitingForConnectionCurrentCount State

#echo "NAME:${NAME} VALUE:${VALUE}"

case "${NAME}" in

PoolState:)

POOL_STATE=${VALUE}

;;

WaitingForConnectionCurrentCount:)

WAIT_CNT=${VALUE}

;;

State:)

RUN_STATE=${VALUE}

;;

esac

done

#echo "POOL_STATE:${POOL_STATE}"

#echo "WAIT_CNT:${WAIT_CNT}"

#echo "RUN_STATE:${RUN_STATE}"

if [[ "${POOL_STATE}" != "true" ]]

then

echo "CRITICAL - ${POOL_INFO} PoolState is ${POOL_STATE}"

exit $STATE_CRITICAL

fi

if [[ "${RUN_STATE}" != "Running" ]]

then

echo "CRITICAL - ${POOL_INFO} State is ${RUN_STATE}"

exit $STATE_CRITICAL

fi

if [[ "${WAIT_CNT}" -gt "0" ]]

then

echo "WARNING - ${POOL_INFO} WaitingForConnectionCurrentCount is ${WAIT_CNT}" exit $STATE_WARNING

fi

else

#error

printf "${RE}" | awk '{ printf $0 }' | read ERR_INFO

echo "CRITICAL - ${ERR_INFO}"

exit $STATE_CRITICAL

fi

echo "OK - ${POOL_INFO} State is ${RUN_STATE},PoolState is $

{POOL_STATE},WaitingForConnectionCurrentCount is ${WAIT_CNT}"

exit $STATE_OK

;;

*)

print_usage

exit $STATE_UNKNOWN

;;

esac

5.配置Weblogic监控

将check_wls.sh上传到Nagios软件的libexec目录下，并创建一个ln文件check_wls。

$ ln -s ./check_wls.sh ./check_wls

在nrpe的配置文件中增加相关的命令定义。

Weblogic的具体配置信息如下，

${URL}t3://172.17.1.2:7001

${USER_NAME}weblogic

${PASS_WORD}weblogic

${DOMAIN_NAME}mydomain

${SERVER_NAME}myserver

${POOL_NAME}mypool

编辑nrpe.cfg文件，增加如下内容，

在nrpe的启动脚本中添加环境变量（CLASSPATH、JAVA_HOME）

编辑监控主机的nagios.cfg文件，添加如下内容。

验证配置是否正确。

重启监控主机上的nagios服务以及远程主机上的nrpe服务。

通过IE观察监控情况。

图5.1

就此配置工作完成。

6.结语本文介绍了一种通过Nagios监控Weblogic应用的实现方式，按照Nagios Plugin API 规则编写自己的Shell脚本实现该功能，并简单的描述了配置过程，提供了Shell源码。希望大家指正。