用nagios來監(jiān)控網絡服務器和網絡服務

Nagiosnagios可以對服務器進行全面的監(jiān)控，包括服務(apache、mysql、ntp、dns、disk、qmail和sshd等等)的狀態(tài)，服務器的狀態(tài)(up、down等等)。它是一個完全GPL協(xié)議的開源軟件包，包含有nagios主程序和它的各個插件，配置非常靈活，可以監(jiān)視的項目很多，可以自定義shell腳本進行監(jiān)控服務，非常適合大型網絡。nagios的包含主動監(jiān)控和被動監(jiān)控。主動檢查是通過監(jiān)控中心的主機發(fā)出請求，讓運行在遠程主機上的nrpe守護進程收集信息，然后報告它，它通過web接口把數(shù)據(jù)顯示在頁面上。它的工作原理如下：被動監(jiān)控是當遠程被監(jiān)控主機處于防火墻之內的時候，只有遠程主機可以訪問到監(jiān)控中心，防火墻之內可以設置另外一個監(jiān)控中心，遠程監(jiān)控中心的nagios收集服務器信息以后，和nsca報告，由naca客戶端報告naca的服務器端，然后報告監(jiān)控中心的nagios，通過web接口顯示監(jiān)控結果。nagios的功能非常強大，/是它的窩，只有e文、法文和日文，沒有中文，可惜啊。我現(xiàn)在引用它的一段文字進行總結一下到底什么是nagios：WhatIsThis?什么是nagios？Nagios®isasystemandnetworkmonitoringapplication.Itwatcheshostsandservicesthatyouspecify,alertingyouwhenthingsgobadandwhentheygetbetter.NagioswasoriginallydesignedtorununderLinux,althoughitshouldworkundermostotherunicesaswell.SomeofthemanyfeaturesofNagios®include:Monitoringofnetworkservices(SMTP,POP3,HTTP,NNTP,PING,etc.)Monitoringofhostresources(processorload,diskusage,etc.)SimpleplugindesignthatallowsuserstoeasilydeveloptheirownservicechecksParallelizedservicechecksAbilitytodefinenetworkhosthierarchyusing"parent"hosts,allowingdetectionofanddistinctionbetweenhoststhataredownandthosethatareunreachableContactnotificationswhenserviceorhostproblemsoccurandgetresolved(viaemail,pager,oruser-definedmethod)AbilitytodefineeventhandlerstoberunduringserviceorhosteventsforproactiveproblemresolutionAutomaticlogfilerotationSupportforimplementingredundantmonitoringhostsOptionalwebinterfaceforviewingcurrentnetworkstatus,notificationandproblemhistory,logfile,etc.Nagios是一個監(jiān)視系統(tǒng)和網絡的應用程序。它監(jiān)視你所指定主機和服務，當監(jiān)視的內容變好或者變壞時發(fā)出警告。Nagios最初是被設計在Linux平臺上運行的，然而現(xiàn)在在其他平臺上也運行良好。Nagios的特性包括：監(jiān)視網絡服務（SMTP,POP3,HTTP,NNTP,PING等等）監(jiān)視主機資源（處理器負載、磁盤空間等）容許用戶開發(fā)自己的插件去檢查自定義的項目；通過使用“父主機”，定義網絡主機的分層，容許探測主機down掉或者不可到達。可以定義在主機或服務運行期間，事件發(fā)生以后如何處理和解決方式；自動記錄錯誤日志；支持冗余監(jiān)視；可選web接口，通過web頁面查看當前網絡狀態(tài)，提示和報告故障歷史，日志文件等；Nagios的系統(tǒng)要求：Linux、Unix等apacheGD庫（1.63以上）zlibpnglibjpeglibbasicicons等,其中apache的安裝在blog中已經有相關的文章，搜索一下就行;gd、zlib、pnglib和jpeglib安裝比較簡單，步驟：下載tarballtarzxvfxxx.tar.gzcdxxx./configuremake&&makeinstallNagios的安裝過程(FreeBSD)nagios的安裝比較簡單，復雜的是設置和配置參數(shù)的設定。不過你要放松一點，畢竟我們要搞定它，不是嗎？那就開始吧：1：獲得最新的安裝包，/download2：以root身份登錄服務器，目前最新的版本是2.5：nagios，版本2.5：fetch/sourceforge/nagios/nagios-2.5.tar.gzorwget/sourceforge/nagios/nagios-2.5.tar.gz獲得nagios插件，版本1.4.3：/sourceforge/nagiosplug/nagios-plugins-1.4.3.tar.gz3）獲得圖庫文件：/nagios/imagepak-base.tar.gzNRPE，版本2.5.2/sourceforge/nagios/nrpe-2.5.2.tar.gzNSCA，版本2.6/sourceforge/nagios/nsca-2.6.tar.gz3：切換到root用戶：sudosu4：解壓縮tarzxvfnagios-2.5.tar.gz5：建立運行nagios的用戶：addusernagios6：建立安裝nagios的文件夾，并使這個文件夾的所有者為nagios:nagiosmkdir/usr/local/nagioschownnagios.nagios/usr/local/nagios7：確認web服務器的用戶可能會通過web接口執(zhí)行一些命令，必須確定web服務器以哪個用戶運行的，通常為apache：grep"AUser"/usr/local/apache2/conf/httpd.conf8：建立命令文件組這個新的組會包括apache的用戶和nagios的用戶pwgroupaddnagcmdpwusermodapache-Gnagcmdpwusermodnagios-Gnagcmdcat/etc/groupnagcmd:*:9007:apache,nagios8：運行配置腳本并安裝nagioscdnagios-2.5./configure--prefix=/usr/local/nagios--with-gd-lib=/usr/local/lib--with-gd-inc=/usr/local/include***Configurationsummaryfornagios2.507-13-2006***GeneralOptions:Nagiosexecutable:nagiosNagiosuser/group:nagios,nagiosCommanduser/group:nagios,nagiosEmbeddedPerl:noEventBroker:yesInstall${prefix}:/usr/local/nagiosLockfile:${prefix}/var/nagios.lockInitdirectory:/usr/local/etc/rc.dHostOS:freebsd6.0WebInterfaceOptions:HTMLURL:http://localhost/nagios/CGIURL:http://localhost/nagios/cgi-bin/Traceroute(usedbyWAP):/usr/sbin/tracerouteReviewtheoptionsaboveforaccuracy.Iftheylookokay,type'makeall'tocompilethemainprogramandCGIs.makeallmakeinstallmakeinstall-initmakeinstall-commandmodemakeinstall-config9：安裝nagios-pluginstarzxvfnagios-plugins-1.4.3.tar.gzcdnagios-plugins-1.4.3./configure--prefix=/usr/local/nagios-pluginsmakeallmakeinstall安裝完成以后在/usr/local/nagios-plugins-plugins會產生一個libexec的目錄，將該目錄全部移動到/usr/local/nagios目錄下即可。mv/usr/local/nagios-plugins-plugins/libexec//usr/local/nagios/10：imagepak-base.tar.gz的安裝tar-vzfimagepak-base.tar.gz解壓以后是base目錄mvbase//usr/local/nagios/share/images/logos/現(xiàn)在開始配置：1：配置web接口假設你已經運行了apache，如果沒有，請參考：http://localhost/upload/blog.php?do-showone-tid-18.htmlvi/usr/local/apache2/conf/httpd.conf添加如下內容：ScriptAlias/nagios/cgi-bin/usr/local/nagios/sbin<Directory"/usr/local/nagios/sbin">OptionsExecCGIAllowOverrideNoneOrderallow,denyAllowfromallAuthName"NagiosAccess"AuthTypeBasicAuthUserFile/usr/local/nagios/etc/htpasswd.usersRequirevalid-user</Directory>Alias/nagios/usr/local/nagios/share<Directory"/usr/local/nagios/share">OptionsNoneAllowOverrideNoneOrderallow,denyAllowfromallAuthName"NagiosAccess"AuthTypeBasicAuthUserFile/usr/local/nagios/etc/htpasswd.usersRequirevalid-user</Directory>修改完畢，保存文件，并重啟apache：/usr/local/apahce2/bin/apachectlrestart2：配置apache的BASIC認證：生成認證密碼：/usr/local/apache2/bin/htpasswd-/usr/local/nagios/etc/htpasswd.usersnagiosnagiosapache接口配置完成。開始配置nagios：cd/usr/local/nagios/etc/在/usr/local/nagios/etc下是nagios的配置模板文件-sample,把.cfg-sample文件全部拷貝成.cfg例如:cpnagios.cfg-samplenagios.cfg全部拷貝完成即可.viminimal.cfg注釋所有command：注釋的方法是在每一個定義語句前面添加”#“修改cgi.cfg修改use_authentication=l為use_authentication=0,即不用驗證.不然有一些頁面不會顯示?，F(xiàn)在檢查配置文件是否有語法錯誤：/usr/local/nagios/bin/nagios-v/usr/local/nagios/etc/nagios.cfg如果正確，會顯示以下結果：TotalWarnings:0TotalErrors:0否則，需要根據(jù)提示進行修改配置文件。配置文件等會再弄。現(xiàn)在啟動nagios/usr/local/nagios/bin/nagios-d/usr/local/nagios/etc/nagios.cfg為了使nagios異常中斷，我們使用daemontools啟動：安裝daemontool：mkdir-p/packagechmodl755/packagecd/packagefetchhttp://cr.yp.to/daemontools/daemontools-0.76.tar.gzcdadmin/daemontools-0.76/package/install檢查svscan進程是否啟動：psaux|grepsvscanroot 3760.00.0l6360con-IW- 0:00.00/bin/sh/command/svscanbootroot 4ll0.00.0l224208con-S8Jul060:42.50svscan/serviceok,啟動正常了。cd/servicemkdirnagioschmodl755nagiostouch./runchmod755./runvirunPATH=/usr/local/bin:/usr/bin:/binexportPATHexecenv-PATH=$PATH\/usr/local/nagios/bin/nagios/usr/local/nagios/etc/nagios.cfgmkdirlogcdlogtouch./runchmod755./runvi./run#!/bin/shexecsetuidgidlogadminmultilogts1000000n100./mainmkdirmainchmod777mainchownnagios.nagiosmaintouchstatuschownnagios.nagiosstatussvc-u/service/nagios/svstat/service/nagios/root@##psauxww|grepnagiosroot232760.00.11176488??I5:00PM0:01.71supervisenagiosnagios342510.00.323161552??S6:06PM0:00.10/usr/local/nagios/bin/nagios/usr/local/nagios/etc/nagios.cfgroot@##ok,現(xiàn)在把nagios服務做成自動啟動的服務了。通過svc命令可以啟動或者停止服務。svcoptsservicesoptsisaseriesofgetopt-styleoptions.servicesconsistsofanynumberofarguments,eachargumentnamingadirectoryusedbysupervise.-u:Up.Iftheserviceisnotrunning,startit.Iftheservicestops,restartit.-d:Down.Iftheserviceisrunning,senditaTERMsignalandthenaCONTsignal.Afteritstops,donotrestartit.-o:Once.Iftheserviceisnotrunning,startit.Donotrestartitifitstops.-p:Pause.SendtheserviceaSTOPsignal.-c:Continue.SendtheserviceaCONTsignal.-h:Hangup.SendtheserviceaHUPsignal.-a:Alarm.SendtheserviceanALRMsignal.-i:Interrupt.SendtheserviceanINTsignal.-t:Terminate.SendtheserviceaTERMsignal.-k:Kill.SendtheserviceaKILLsignal.-x:Exit.supervisewillexitassoonastheserviceisdown.Ifyouusethisoptiononastablesystem,you'redoingsomethingwrong;superviseisdesignedtorunforever.比如：停止nagios－－svc-d/service/nagios/重啟nagios－－svc-t/service/nagios/啟動nagios－－svc-u/service/nagios/當然，你也可以使用inited的方式進行：/usr/local/etc/rc.d/nagiosstart/stop好了，反正daemontools很強大，以后慢慢熟悉，轉入正題?，F(xiàn)在打開網頁：http://localhost/nagios/一定會讓你大吃一驚，呵呵，我的服務器和服務狀態(tài)都清楚的看到了?，F(xiàn)在我們的nagios中只有一個，那就是它自己，loca山ost,呵呵，等會我們添加別的主機和主機服務，ok,我們認識一下nagios的廬山真面目：配置nagios：1） 為主機添加服務2） 添加主機并添加服務3） 停止一個服務4） 刪除一臺主機和服務5） 查看所有主機的故障6） 查看一臺特定的主機狀態(tài)7） 改變報警的時間間隔8） 改變發(fā)現(xiàn)故障的重試次數(shù)9） 如何在nagios中使用外部命令1）為主機添加一個服務為localhost主機添加qmail服務的監(jiān)控，方法如下：viminimal.cfgdefineservice{use generic-service;Nameofservicetemplatetousehost_name localhostTOC\o"1-5"\h\zservice_descriptionqmail_smtpis_volatile 0check_period 24x7max_check_attempts1normal_check_interval1retry_check_interval 1contact_groups adminsnotification_options w,u,c,rnotification_interval960notification_period 24x7check_command check_smtp!20%!10%!/

可以直接拷貝原有的進行修改，我這個就是拷貝的原有的check」ocal_disk進行的。修改host_name，service_description，check_command等defineservice{use generic-service;Nameofservicetemplatetousehost_name localhostservice_descriptionqmail_pop3TOC\o"1-5"\h\zis_volatile 0check_period 24x7max_check_attempts 1normal_check_interval1retry_check_interval 1contact_groupsnotification_optionsnotification_intervalnotification_periodcheck_command}contact_groupsnotification_optionsnotification_intervalnotification_periodcheck_command}照貓畫虎的進行修改adminsw,u,c,r96024x7check_pop!20%!10%!/然后去修改：vicheckcommands.cfg#'check_qmail'commanddefinitiondefinecommand{command_namecheck_qmailcommand_line$USER1$/check_smtp-H}definecommand{command_namecheck_pop3command_line$USER1$/check_pop-H}保存，然后檢查配置文件：/usr/local/nagios/bin/nagios-v/usr/local/nagios/etc/nagios.cfg如果沒有錯誤會顯示：TotalWarnings:0TotalErrors:0如果有錯誤，請根據(jù)提示進行錯誤的修正。重啟nagiossvc-d/service/nagios/&&svc-u/service/nagios/通過web頁面檢查nagios的結果：53/nagios/點擊“ServiceDetail"會出現(xiàn)：2）添加主機并添加服務我們會監(jiān)控這臺主機的負載、磁盤等一些沒有通過端口方式啟動的服務器狀態(tài)，以及它的服

務，比如：apache、mysql、qmail和ntp等等吧。那么沒有端口的nagios直接能監(jiān)控到嗎？答案是不行。所以我們必須在兩臺主機上安裝nrpe，nrpe可以啟動5666端口，把檢測的信息源源不斷的傳給監(jiān)控中心的主機。ok,我們把apache、mysql、qmail和ntp先加上，這回我們把監(jiān)控的主機和服務新建一個文件：cd/usr/local/nagios/etc/touch10_5_1_156.cfgvinagios.cfgcfg_file=/usr/local/nagios/etc/10_5_1_156.cfgvi10_5_1_156.cfg定義一個主機：definehost{use generic-host ;Nameofhosttemplatetousehost_nametest_nrpealias clientaddress56check_commandcheck-host-alivemax_check_attempts1check_period24x7notification_interval120notification_period24x7notification_optionsd,rcontact_groupsadmins}定義主機需要檢查的服務：defineservice{use generic-service ;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionPINGTOC\o"1-5"\h\zis_volatile 0check_period 24x7max_check_attempts1normal_check_interval1retry_check_interval 1contact_groups adminsnotification_options w,u,c,rnotification_interval960notification_periodcheck_command}notification_periodcheck_command}check_ping!100.0,20%!500.0,60%defineservice{

usegeneric-service;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionapacheis_volatile 0check_period24x7max_check_attempts1normal_check_interval1retry_check_interval1contact_groupsadminsnotification_optionsw,u,c,rnotification_interval960notification_periodcheck_commandnotification_periodcheck_command}check_http!100.0,20%!500.0,60%defineservice{usegeneric-service;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionmysqlis_volatile 0check_period24x7max_check_attempts1normal_check_interval1retry_check_interval1contact_groupsadminsnotification_optionsw,u,c,rnotification_interval960notification_periodcheck_commandnotification_periodcheck_command}check_mysql!100.0,20%!500.0,60%defineservice{usegeneric-service ;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionntpis_volatile 0check_period24x7max_check_attempts1normal_check_interval1retry_check_interval1contact_groupsadminsnotification_optionsw,u,c,rnotification_interval960notification_period24x7

check_commandcheck_ntp!100.0,20%!500.0,60%}defineservice{usegeneric-service ;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionqmail_smtpis_volatile 0check_period24x7max_check_attempts1normal_check_interval1retry_check_interval1contact_groupsadminsnotification_optionsw,u,c,rnotification_interval960notification_periodcheck_commandnotification_periodcheck_command}check_smtp!100.0,20%!500.0,60%defineservice{usegeneric-service ;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionqmail_pop3is_volatile 0check_period24x7max_check_attempts1normal_check_interval1retry_check_interval1contact_groupsadminsnotification_optionsw,u,c,rnotification_interval960notification_periodcheck_commandnotification_periodcheck_commandcheck_pop!100.0,20%!500.0,60%}現(xiàn)在我們象上次一樣把服務也定義完了

}現(xiàn)在我們象上次一樣把服務也定義完了new_host_serviceliu1084回復于：2006-08-1418:29:02此時是不是多了一個主機和它下面的服務呢?那是肯定的，添加主機和服務可能出現(xiàn)的問題有如下情況：1：配置參數(shù)出現(xiàn)問題，如果你沒有檢查配置就啟動nagios，可能會啟動成功，但是顯示會不正常；解決方法：調整配置參數(shù)2：Connectionrefused當出現(xiàn)這個問題的時候，我開始以為是ssh的無密碼登錄沒有成功，但是其實我的服務器沒有啟動該服務造成的，啟動服務即可。但是這些是有端口的服務，沒有使用端口的狀態(tài)任何檢測？使用nrpe，ok,我們現(xiàn)在在服務器上安裝nrpe：一、遠程主機的配置1、安裝nrpe與配置fetch/sourceforge/nagios/nrpe-2.5.2.tar.gztarzxvfnrpe-2.5.2.tar.gzcdnrpe-2.5.2./configure--enable-ssl--enable-command-argsmakeallmkdir-p/usr/local/nagios/etcmkdir/usr/local/nagios/binmkdir/usr/local/nagios/libexecpwaddgroupnagiospwuseraddnagios-gnagios-d/usr/local/nagios/-s/sbin/nologinchown-Rnagios:nagios/usr/local/nagioscp./sample-config/nrpe.cfg/usr/local/nagios/etccpsrc/nrpe/usr/local/nagios/bin2、啟動nrpe,端口為5666/usr/local/nagios/bin/nrpe-c/usr/local/nagios/etc/nrpe.cfg-dnetstat-ant|grep5666tcp400*.5666 *.*LISTEN二、監(jiān)控服務器上的配置1、 安裝nrpe（主要是使用check_nrpe模塊）fetch/sourceforge/nagios/nrpe-2.5.2.tar.gztarzxvfnrpe-2.5.2.tar.gzcdnrpe-2.5.2./configure--enable-ssl--enable-command-argsmakeallcpsrc/check_nrpe/usr/local/nagios/libexec2、 nagios文件的配置vicheckcommands.cfg定義check_nrpe命令#'check_nrep'commanddefinitiondefinecommand{command_namecheck_nrpecommand_line/usr/local/nagios/libexec/check_nrpe-H$HOSTADDRESS$-c$ARG1$}三、上面我們已經配置了一部分參數(shù),下面是配置的最終結果：definehost{use generic-host ;Nameofhosttemplatetousehost_nametest_nrpealias clientaddress56check_commandcheck-host-alivemax_check_attempts1check_period24x7

notification_interval120notification_period24x7notification_optionsd,rcontact_groupsadmins}#'check_load'commanddefinitiondefinecommand{command_namecheck_loadcommand_line$USER1$/check_load-w$ARG1$-c$ARG2$}#'check_load'commanddefinitiondefinecommand{command_namecheck_diskcommand_line$USER1$/check_disk-w$ARG1$-c$ARG2$}defineservice{usegeneric-service ;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionPINGis_volatile 0check_period24x7max_check_attempts1normal_check_interval1retry_check_interval1contact_groupsadminsnotification_optionsw,u,c,rnotification_interval960notification_periodcheck_commandnotification_periodcheck_command}check_ping!100.0,20%!500.0,60%defineservice{usegeneric-service ;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionapacheis_volatile 0check_period24x7max_check_attempts1normal_check_interval1retry_check_interval1contact_groupsadminsnotification_optionsw,u,c,r

notification_intervalnotification_periodcheck_command96024x7check_http!100.0,20%!500.0,60%}defineservice{usegeneric-service;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionmysqlis_volatile 0check_period24x7max_check_attempts1normal_check_interval1retry_check_interval1contact_groupsadminsnotification_optionsw,u,c,rnotification_interval960notification_periodcheck_commandnotification_periodcheck_command}check_mysql!100.0,20%!500.0,60%defineservice{usegeneric-service;Nameofservicetemplatetousehost_name test_nrpeservice_descriptionntpis_volatile 0check_period24x7max_check_attempts1normal_check_interval1retry_check_interval1contact_groupsadminsnotification_optionsw,u,c,rnotification_interval960notification_periodcheck_commandnotification_periodcheck_command}check_ntp!100.0,20%!500.0,60%defin