BlogJava-冉翔的技术专栏-文章分类-05.Web/App Serverhttp://www.blogjava.net/ranxiang/category/4385.html关注于J2SE，J2EE以及开源方面的技术。zh-cnTue, 27 Feb 2007 22:38:56 GMTTue, 27 Feb 2007 22:38:56 GMT60Apache与Tomcat的整合【mod_jk篇】http://www.blogjava.net/ranxiang/articles/23144.html冉翔冉翔Fri, 09 Dec 2005 07:19:00 GMThttp://www.blogjava.net/ranxiang/articles/23144.htmlhttp://www.blogjava.net/ranxiang/comments/23144.htmlhttp://www.blogjava.net/ranxiang/articles/23144.html#Feedback0http://www.blogjava.net/ranxiang/comments/commentRss/23144.htmlhttp://www.blogjava.net/ranxiang/services/trackbacks/23144.htmlApache与Tomcat的整合

作者：冉翔

 

环境：

windows2003

jdk1.5

Apache 2.0.53

tomcat5.5.9

 

相关软件下载地址：

apache 2.0.53：

http://archive.apache.org/dist/httpd/binaries/win32/

tomcat5.5.9：

http://archive.apache.org/dist/jakarta/tomcat-5/v5.5.9/bin/jakarta-tomcat-5.5.9.zip

jdk1.5：

http://java.sun.com/j2se/1.5.0/download.jsp

mod_jk-1.2.12-apache-2.0.54.so：

http://www.apache.org/dist/jakarta/tomcat-connectors/jk/binaries/win32/jk-1.2.12/mod_jk-1.2.12-apache-2.0.54.so

 

步骤，

 

前戏：先在D:/Java/tomcat/webapps/下面建立一个文件夹test。

里面放上一个jsp和html文件，叫做index.jsp和test.html好了：）

 

首先，将mod_jk-1.2.12-apache-2.0.54.so

放入apache/modules文件夹下面。

然后在apache/conf文件夹下，新建文件workers.properties

 

内容如下：

# 这现行根据安装目录做修改

workers.tomcat_home=D:/Java/tomcat

workers.java_home=C:/jdk1.5

ps=\

# worker.list=ajp13

worker.list=ajp12,ajp13

 

worker.ajp12.port=8007

worker.ajp12.host=localhost

worker.ajp12.type=ajp12

worker.ajp12.lbfactor=1

           

worker.ajp13.port=8009       

worker.ajp13.host=localhost    

worker.ajp13.type=ajp13

worker.ajp13.lbfactor=1

 

worker.loadbalancer.type=lb

 

worker.loadbalancer.balanced_workers=ajp12, ajp13

worker.inprocess.type=jni

worker.inprocess.class_path=$(workers.tomcat_home)$(ps)classes

worker.inprocess.class_path=$(workers.tomcat_home)$(ps)lib$(ps)jaxp.jar

 

worker.inprocess.class_path=$(workers.tomcat_home)$(ps)lib$(ps)parser.jar

 

worker.inprocess.class_path=$(workers.tomcat_home)$(ps)common$(ps)lib$(ps)jasper.jar

 

worker.inprocess.class_path=$(workers.tomcat_home)$(ps)common$(ps)lib$(ps)servlet.jar

 

worker.inprocess.class_path=$(workers.tomcat_home)$(ps)common$(ps)lib$(ps)webserver.jar

 

worker.inprocess.class_path=$(workers.java_home)$(ps)lib$(ps)tools.jar

 

worker.inprocess.cmd_line=-config

 

worker.inprocess.cmd_line=$(workers.tomcat_home)/conf/jni_server.xml

 

worker.inprocess.cmd_line=-home

 

worker.inprocess.cmd_line=$(workers.tomcat_home)

 

worker.inprocess.jvm_lib=$(workers.java_home)$(ps)jre$(ps)bin$(ps)classic$(ps)jvm.dll

 

 

worker.inprocess.stdout=$(workers.tomcat_home)$(ps)inprocess.stdout

 

worker.inprocess.stderr=$(workers.tomcat_home)$(ps)inprocess.stderr

 

worker.inprocess.sysprops=tomcat.home=$(workers.tomcat_home)

 

编辑apache/conf文件夹内的httpd.conf

添加如下字段：

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

LoadModule jk_module modules/mod_jk-1.2.12-apache-2.0.54.so

#禁止访问WEB_INF和META_INF目录。

 <Directory ~ "/WEB-INF/">

     Order allow,deny

     Deny from all

    </Directory>

     <Directory ~ "/META-INF/">

     Order allow,deny

     Deny from all

    </Directory>

 

#添加映射，使可以使用http://localhost/test/来访问

Alias /test/ "D:/Java/tomcat/webapps/test/"

    <Directory "D:/Java/tomcat/webapps/test/">

        Options Indexes MultiViews

        AllowOverride None

        Order allow,deny

        Allow from all

        DirectoryIndex index.jsp

    </Directory>

#添加映射，使可以使用http://localhost/test来访问，注意最后没有那个/

Alias /test "D:/Java/tomcat/webapps/test/"

    <Directory "D:/Java/tomcat/webapps/test/">

        Options Indexes MultiViews

        AllowOverride None

        Order allow,deny

        Allow from all

        DirectoryIndex index.jsp

    </Directory>

 

#加载配置文件

JkWorkersFile "D:/Java/xampp/apache/conf/workers.properties"

#对URL进行规则匹配

JkMount /test/ ajp13

JkMount /test/servlet/* ajp13

JkMount /test/*.jsp ajp13

JkMount /test/*.do ajp13

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

 

查看tomcat/conf/下的server.xml

添加如下字段，当然如果已经有了就不用添加了，或者有注释，去掉就可以了。

<Connector port="8009"

               enableLookups="false" redirectPort="8443" protocol="AJP/1.3" URIEncoding="GBK" />

 

至此启动tomcat，启动apache。

访问

http://localhost，出现apache欢迎界面

http://localhost:8080，出现tomcat欢迎界面

http://localhost:8080/waityou，出现index.jsp的页面，去掉:8080，正常。

http://lcaolhost:8080/waityou/test.html，出现test.html的页面，去掉8080，正常。

 

阅读材料：

http://community.csdn.net/Expert/topic/3902/3902301.xml?temp=.441662

http://jakarta.apache.org/tomcat/connectors-doc/

《Tomcat权威指南》英文版

 

感谢：PIGO。在QQ上对我进行热心的讲解和辅导。解除了我的很多疑问。教会了我很多东西。



冉翔 2005-12-09 15:19 发表评论
]]>【转载】tomcat的server.xml（中文版）http://www.blogjava.net/ranxiang/articles/23145.html冉翔冉翔Fri, 09 Dec 2005 07:19:00 GMThttp://www.blogjava.net/ranxiang/articles/23145.htmlhttp://www.blogjava.net/ranxiang/comments/23145.htmlhttp://www.blogjava.net/ranxiang/articles/23145.html#Feedback0http://www.blogjava.net/ranxiang/comments/commentRss/23145.htmlhttp://www.blogjava.net/ranxiang/services/trackbacks/23145.html <!-- Tomcat服务器配置示例文件 -->

<!-- Note that component elements are nested corresponding to their
     parent-child relationships with each other -->
<!-- 注意，这些组件的构成是根据相互之间的父子关系进行嵌套的。-->

<!-- A "Server" is a singleton element that represents the entire JVM,
     which may contain one or more "Service" instances. The Server
     listens for a shutdown command on the indicated port.

     Note: A "Server" is not itself a "Container", so you may not
     define subcomponents such as "Valves" or "Loggers" at this level.
 -->
<!-- 一个“Server”是一个提供完整的JVM的独立组件，它可以包含一个或多个
     “Service”实例。服务器在指定的端口上监听shutdown命令。
     
     注意：一个“Server”自身不是一个“Container”（容器），因此在这里你
     不可以定义诸如“Valves”或者“Loggers”子组件
-->

<Server port="8005" shutdown="SHUTDOWN" debug="0">


  <!-- Comment these entries out to disable JMX MBeans support -->
  <!-- 这些指令组件关闭JMX MBeans支持 -->
 
  <!-- You may also configure custom components (e.g. Valves/Realms) by
       including your own mbean-descriptor file(s), and setting the
       "descriptors" attribute to point to a ';' seperated list of paths
       (in the ClassLoader sense) of files to add to the default list.
       e.g. descriptors="/com/myfirm/mypackage/mbean-descriptor.xml"
  -->
  <!-- 你也可以通过包含你自己的mbean描述文件配置自定义的组件，然后设置
       “descriptors”属性为以“；”为分隔的文件名列表严将它添加到默认列表
       中，例如：descriptors="/com/myfirm/mypackage/mbean-descriptor.xml"。
 -->
       
  <Listener className="org.apache.catalina.mbeans.ServerLifecycleListener"
            debug="0"/>
  <Listener className="org.apache.catalina.mbeans.GlobalResourcesLifecycleListener"
            debug="0"/>

  <!-- Global JNDI resources -->
  <!-- 全局JNDI资源 -->
  <GlobalNamingResources>

    <!-- Test entry for demonstration purposes -->
    <!-- 出于示例目的的测试入口 -->
    <Environment name="simpleValue" type="java.lang.Integer" value="30"/>

    <!-- Editable user database that can also be used by
         UserDatabaseRealm to authenticate users -->
    <!-- 可编辑的，用来通过UserDatabaseRealm认证用户的用户数据库 -->
    
    <Resource name="UserDatabase" auth="Container"
              type="org.apache.catalina.UserDatabase"
       description="User database that can be updated and saved">
    </Resource>
    <ResourceParams name="UserDatabase">
      <parameter>
        <name>factory</name>
        <value>org.apache.catalina.users.MemoryUserDatabaseFactory</value>
      </parameter>
      <parameter>
        <name>pathname</name>
        <value>conf/tomcat-users.xml</value>
      </parameter>
    </ResourceParams>

  </GlobalNamingResources>

  <!-- A "Service" is a collection of one or more "Connectors" that share
       a single "Container" (and therefore the web applications visible
       within that Container). Normally, that Container is an "Engine",
       but this is not required.

       Note: A "Service" is not itself a "Container", so you may not
       define subcomponents such as "Valves" or "Loggers" at this level.
   -->
  <!-- 一个“Service”是一个或多个共用一个单独“Container”(容器)的“Connectors”
       组合（因此，应用程序在容器中可见）。通常，这个容器是一个“Engine”
       （引擎），但这不是必须的。
       
       注意：一个“Service”自身不是一个容器，因此，在这个级别上你不可定义
       诸如“Valves”或“Loggers”子组件。
  -->

  <!-- Define the Tomcat Stand-Alone Service -->
  <!-- 定义Tomcat的标准独立服务 -->
  <Service name="Catalina">

    <!-- A "Connector" represents an endpoint by which requests are received
         and responses are returned. Each Connector passes requests on to the
         associated "Container" (normally an Engine) for processing.

         By default, a non-SSL HTTP/1.1 Connector is established on port 8080.
         You can also enable an SSL HTTP/1.1 Connector on port 8443 by
         following the instructions below and uncommenting the second Connector
         entry. SSL support requires the following steps (see the SSL Config
         HOWTO in the Tomcat 5 documentation bundle for more detailed
         instructions):
         * If your JDK version 1.3 or prior, download and install JSSE 1.0.2 or
           later, and put the JAR files into "$JAVA_HOME/jre/lib/ext".
         * Execute:
             %JAVA_HOME%\bin\keytool -genkey -alias tomcat -keyalg RSA (Windows)
             $JAVA_HOME/bin/keytool -genkey -alias tomcat -keyalg RSA (Unix)
           with a password value of "changeit" for both the certificate and
           the keystore itself.

         By default, DNS lookups are enabled when a web application calls
         request.getRemoteHost(). This can have an adverse impact on
         performance, so you can disable it by setting the
         "enableLookups" attribute to "false". When DNS lookups are disabled,
         request.getRemoteHost() will return the String version of the
         IP address of the remote client.
    -->
    <!-- 一个“Connector”（连接器）代表一个请求被接收和应答所需要的端点。每个连
         接器通过相关的“Container”（容器）处理请求。
         
         默认情况下，一个非SSL的HTTP/1.1的连接器被绑定在端口8080。你也可以通过
         根据后面的使用说明并取消第二个连接器入口的注释，在端口8443上建立一个
         SSL HTTP/1.1的连接器。开放SSL支持需要下面几步（参见Tomcat 5文档中怎样
         配置SSL的说明以取得更多的详细信息）：
         * 如果你的JDK是1.3或1.3以前的版本，下载安装JSSE 1.0.2或以后版本，并放
           置JAR文件到“$JAVA_HOME/jre/lib/ext”目录下。
         * 带一个“changeit”的口令值执行：
             %JAVA_HOME%\bin\keytool -genkey -alias tomcat -keyalg RSA (Windows)
             $JAVA_HOME/bin/keytool -genkey -alias tomcat -keyalg RSA (UNIX)
           来生成它自己的证书私钥。
           
         默认情况下，当一个web应用程序调用请求时，DNS查询是可行的。这将对性能造
         成一些不利的影响，因此，你可以将“enableLookups”设置为“false”来关闭DNS
         查询。当DNS查询被关闭时，request.getRemoteHost()将返回包含远程客户IP地
         址的字符串。
  -->

    <!-- Define a non-SSL Coyote HTTP/1.1 Connector on port 8080 -->
    <!-- 在8080端口上定义一个非SSL的HTTP/1.1连接器 -->
    <Connector port="8080"
               maxThreads="150" minSpareThreads="25" maxSpareThreads="75"
               enableLookups="false" redirectPort="8443" acceptCount="100"
               debug="0" connectionTimeout="20000"
               disableUploadTimeout="true" />
    <!-- Note : To disable connection timeouts, set connectionTimeout value
     to 0 -->
    <!-- 注意，要关闭连接超时，将connectionTimeout设置为0 -->

<!-- Note : To use gzip compression you could set the following properties :

compression="on"
compressionMinSize="2048"
noCompressionUserAgents="gozilla, traviata"
compressableMimeType="text/html,text/xml"
-->
<!-- 注意：要使用gzip压缩需要设置如下属性：

compression="on"
compressionMinSize="2048"
noCompressionUserAgents="gozilla,traviata"
compressableMimeType="text/html,text/xml"
-->

    <!-- Define a SSL Coyote HTTP/1.1 Connector on port 8443 -->
    <!-- 在端口8443上定义一个SSL的HTTP/1.1的连接器 -->
    <!--
    <Connector port="8443"
               maxThreads="150" minSpareThreads="25" maxSpareThreads="75"
               enableLookups="false" disableUploadTimeout="true"
               acceptCount="100" debug="0" scheme="https" secure="true"
               clientAuth="false" sslProtocol="TLS" />
    -->

    <!-- Define a Coyote/JK2 AJP 1.3 Connector on port 8009 -->
    <!-- 在端口8009上定义一个Coyote/JK2 AJP 1.3连接器 -->
    <Connector port="8009"
               enableLookups="false" redirectPort="8443" debug="0"
               protocol="AJP/1.3" />

    <!-- Define a Proxied HTTP/1.1 Connector on port 8082 -->
    <!-- 在8082端口上定义一个代理HTTP/1.1连接器 -->
    <!-- See proxy documentation for more information about using this. -->
    <!-- 参见代理文档以取得这里使用属性的更多的信息。 -->
    <!--
    <Connector port="8082"
               maxThreads="150" minSpareThreads="25" maxSpareThreads="75"
               enableLookups="false"
               acceptCount="100" debug="0" connectionTimeout="20000"
               proxyPort="80" disableUploadTimeout="true" />
    -->

    <!-- An Engine represents the entry point (within Catalina) that processes
         every request. The Engine implementation for Tomcat stand alone
         analyzes the HTTP headers included with the request, and passes them
         on to the appropriate Host (virtual host). -->
    <!-- 一个“Engine”（引擎）代表处理每个请求的入口点（在Catalina内）。这个Tomcat
         的标准独立引擎实现分析包含在请求中的HTTP头信息，并将请求传送到适当的主机
         或虚拟主机上。-->

    <!-- You should set jvmRoute to support load-balancing via JK/JK2 ie :
    <Engine name="Standalone" defaultHost="localhost" debug="0" jvmRoute="jvm1">
    -->
    <!-- 你应该这样设置jvmRoute以便在JK/JK2通道上支持负载平衡：
    <Engine name="Standalone" defaultHost="localhost" debug="0" jvmRoute="jvm1">
    -->
         
    <!-- Define the top level container in our container hierarchy -->
    <!-- 在我们的容器体系中定义最高级别的容器 -->
    <Engine name="Catalina" defaultHost="localhost" debug="0">

      <!-- The request dumper valve dumps useful debugging information about
           the request headers and cookies that were received, and the response
           headers and cookies that were sent, for all requests received by
           this instance of Tomcat. If you care only about requests to a
           particular virtual host, or a particular application, nest this
           element inside the corresponding <Host> or <Context> entry instead.

           For a similar mechanism that is portable to all Servlet 2.4
           containers, check out the "RequestDumperFilter" Filter in the
           example application (the source for this filter may be found in
           "$CATALINA_HOME/webapps/examples/WEB-INF/classes/filters").

           Request dumping is disabled by default. Uncomment the following
           element to enable it. -->
      <!-- 请求转储器值转储非常有用的通过Tomcat这个实例接收到的所有请求的，关于
           接收到的请求头和cookies，以及发送出的回应头和cookies的调试信息。如果
           你只在意一个特定的虚拟主机或者一个特定应用程序的请求，在对应的<Host>
           或<Context>入口中套用这个元素。
           
           对于所有的Servlet（小服务程序）2.4来所，在示例程序中检验
           “RequestDumperFilter”过滤是非常容易的（这个过滤器源代码可以在
           “$CATALINA_HOME/webapps/examples/WEB-INF/classes/filters”下找到。
           
           请求转储默认情况下是关闭的，如果要使用它，取消下面行的注释符。-->
      <!--
      <Valve className="org.apache.catalina.valves.RequestDumperValve"/>
      -->

      <!-- Global logger unless overridden at lower levels -->
      <!-- 全局日志设置，除非在更低的级别上覆盖它的设置 -->
      <Logger className="org.apache.catalina.logger.FileLogger"
              prefix="catalina_log." suffix=".txt"
              timestamp="true"/>

      <!-- Because this Realm is here, an instance will be shared globally -->
      <!-- 因为这个Realm（区域）在这儿，因此这个实例将被全局共享。 -->

      <!-- This Realm uses the UserDatabase configured in the global JNDI
           resources under the key "UserDatabase". Any edits
           that are performed against this UserDatabase are immediately
           available for use by the Realm. -->
      <!-- 这个Realm（区域）使用在全局JNDI资源下，由“UserDatabase”配置的用户
           数据库。通过使用Realm将使对于这个用户数据库的任何编辑都将会立即可用
      <Realm className="org.apache.catalina.realm.UserDatabaseRealm"
                 debug="0" resourceName="UserDatabase"/>

      <!-- Comment out the old realm but leave here for now in case we
           need to go back quickly -->
      <!-- 注释掉旧有的Realm以便我们能很快速的恢复旧有配置。 -->
      <!--
      <Realm className="org.apache.catalina.realm.MemoryRealm" />
      -->

      <!-- Replace the above Realm with one of the following to get a Realm
           stored in a database and accessed via JDBC -->
      <!-- 下面的属性用于从储存在数据库或从JDBC通道中得到的Realm以取代上面的
           Realm配置。-->

      <!--
      <Realm className="org.apache.catalina.realm.JDBCRealm" debug="99"
             driverName="org.gjt.mm.mysql.Driver"
          connectionURL="jdbc:mysql://localhost/authority"
         connectionName="test" connectionPassword="test"
              userTable="users" userNameCol="user_name" userCredCol="user_pass"
          userRoleTable="user_roles" roleNameCol="role_name" />
      -->

      <!--
      <Realm className="org.apache.catalina.realm.JDBCRealm" debug="99"
             driverName="oracle.jdbc.driver.OracleDriver"
          connectionURL="jdbc:oracle:thin:@ntserver:1521:ORCL"
         connectionName="scott" connectionPassword="tiger"
              userTable="users" userNameCol="user_name" userCredCol="user_pass"
          userRoleTable="user_roles" roleNameCol="role_name" />
      -->

      <!--
      <Realm className="org.apache.catalina.realm.JDBCRealm" debug="99"
             driverName="sun.jdbc.odbc.JdbcOdbcDriver"
          connectionURL="jdbc:odbc:CATALINA"
              userTable="users" userNameCol="user_name" userCredCol="user_pass"
          userRoleTable="user_roles" roleNameCol="role_name" />
      -->

      <!-- Define the default virtual host
           Note: XML Schema validation will not work with Xerces 2.2.
       -->
      <!-- 定义默认的虚拟主机
           注意：XML模式确认将不能与Xerces 2.2同工作。
      -->
      <Host name="localhost" debug="0" appBase="webapps"
       unpackWARs="true" autoDeploy="true"
       xmlValidation="false" xmlNamespaceAware="false">

        <!-- Defines a cluster for this node,
             By defining this element, means that every manager will be changed.
             So when running a cluster, only make sure that you have webapps in there
             that need to be clustered and remove the other ones.
             A cluster has the following parameters:
             为这个结点定义一个群集，通过定义这个元素，意味着每个管理员都将被改变。
             因此，当运行一个群集时，只有确认在必须被聚集和移除其他的地方有你的
             web软件。
             一个群集有下面这些参数：

             className = the fully qualified name of the cluster class
             className = 这个群集类的全限定名

             name = a descriptive name for your cluster, can be anything
             name = 你的群集的描述名，可以是任何描述

             debug = the debug level, higher means more output
             debug = 调试级别，更高意味着更多的输出

             mcastAddr = the multicast address, has to be the same for all the nodes
             mcastAddr = 多播地址，所有的节点都必须有同样的多播地址

             mcastPort = the multicast port, has to be the same for all the nodes
             mcastPort = 多播端口，对所有的节点都必须有同样的多播端口
             
             mcastBindAddr = bind the multicast socket to a specific address
             mcastBindAddr = 绑定多播套接字到一个指定的地址
             
             mcastTTL = the multicast TTL if you want to limit your broadcast
             mcastTTL = 多播生存期，如果你想要限制你的广播的话
             
             mcastSoTimeout = the multicast readtimeout
             mcastSoTimeout = 多播读取超时

             mcastFrequency = the number of milliseconds in between sending a "I'm alive" heartbeat
             mcastFrequency = 发送“I'm alive”（我还活着）信息的间隔毫秒数

             mcastDropTime = the number a milliseconds before a node is considered "dead" if no heartbeat is received
             mcastDropTime = 多长时间没有收到“I'm alive”就将这个节点标识为死节点（单位：毫秒）

             tcpThreadCount = the number of threads to handle incoming replication requests, optimal would be the same amount of threads as nodes
             tcpThreadCount = 用于处理重复引入请求的线程数，最好是每个节点都有相同的线程数

             tcpListenAddress = the listen address (bind address) for TCP cluster request on this host,
                                in case of multiple ethernet cards.
                                auto means that address becomes
                                InetAddress.getLocalHost().getHostAddress()
             tcpListenAddress = 在有多块以太网卡的主机上的监听TCP群集请求的地址（绑定地址），
                                auto意味着地址由InetAddress.getLocalHost().getHostAddress()取得。

             tcpListenPort = the tcp listen port
             tcpListenPort = tcp监听端口

             tcpSelectorTimeout = the timeout (ms) for the Selector.select() method in case the OS
                                  has a wakup bug in java.nio. Set to 0 for no timeout
             tcpSelectorTimeout = 在操作系统中有java.nio唤醒错的情况下，使用Selector.select()的超时
                                  毫秒数。设为0则没有超时限制

             printToScreen = true means that managers will also print to std.out
             printToScreen = true意味着管理员的相关信息也将打印输出到std.out

             expireSessionsOnShutdown = true means that
             expireSessionsOnShutdown = true 意味着?????

             useDirtyFlag = true means that we only replicate a session after setAttribute,removeAttribute has been called.
                            false means to replicate the session after each request.
                            false means that replication would work for the following piece of code:
                            <%
                            HashMap map = (HashMap)session.getAttribute("map");
                            map.put("key","value");
                            %>
             useDirtyFlag = true意味着我们只能在调用setAttribute，removeAttribute后才能复制一个会话；
                            false意味着在每个请求后复制会话。
                            false意味着复制将以下列代码方式工作：
                            <%
                            HashMap map = (HashMap)session.getAttribute("map");
                            map.put("key","value");
                            %>
                            
             replicationMode = can be either 'pooled', 'synchronous' or 'asynchronous'.
                               * Pooled means that the replication happens using several sockets in a synchronous way. Ie, the data gets replicated, then the request return. This is the same as the 'synchronous' setting except it uses a pool of sockets, hence it is multithreaded. This is the fastest and safest configuration. To use this, also increase the nr of tcp threads that you have dealing with replication.
                               * Synchronous means that the thread that executes the request, is also the
                               thread the replicates the data to the other nodes, and will not return until all
                               nodes have received the information.
                               * Asynchronous means that there is a specific 'sender' thread for each cluster node,
                               so the request thread will queue the replication request into a "smart" queue,
                               and then return to the client.
                               The "smart" queue is a queue where when a session is added to the queue, and the same session
                               already exists in the queue from a previous request, that session will be replaced
                               in the queue instead of replicating two requests. This almost never happens, unless there is a
                               large network delay.
             replicationMode = 可以是“pooled”、“synchronous”或者“asynchronous”
                               * Pooled意味着在多个套接字上同步进行复制。例如，先复制数据，然后请求返回。
                               这与“synchronous”相同，除了它使用一个套接字池，因此，它是多线程的。它是
                               最快和最高级的设置，使用这个配置也增加了处理复制的tcp线程的域。
                               * Synchronous指执行请求的线程，它也是向其他节点复制数据的线程，并且直到
                               所有的节点都己经收到信息后才返回。
                               * Asynchronous，对每个群集节点而言，这是一个特别的“sender”（发送者）线程，
                               因此请求线程将复制请求排队到一个小队列中，然后，返回给用户。
                               这个小队列是一个当会话被添加到队列中，而从先前的请求中，同一个会话己经存在
                               ，这个会话将被放置在队列中以代替重复的两个请求。这几乎从不会发生，除非存在
                               较大的网络延迟。
        -->
        <!--
            When configuring for clustering, you also add in a valve to catch all the requests
            coming in, at the end of the request, the session may or may not be replicated.
            A session is replicated if and only if all the conditions are met:
            1. useDirtyFlag is true or setAttribute or removeAttribute has been called AND
            2. a session exists (has been created)
            3. the request is not trapped by the "filter" attribute

            The filter attribute is to filter out requests that could not modify the session,
            hence we don't replicate the session after the end of this request.
            The filter is negative, ie, anything you put in the filter, you mean to filter out,
            ie, no replication will be done on requests that match one of the filters.
            The filter attribute is delimited by ;, so you can't escape out ; even if you wanted to.

            filter=".*\.gif;.*\.js;" means that we will not replicate the session after requests with the URI
            ending with .gif and .js are intercepted.
        -->
        <!-- 当配置群集时，你也添加一个值来捕获所有传入的请求，在请求结束时，这个会话可能被子复制，
             也可能不被复制。一个会话是否复制取决于下列条件是否发生：
             1. useDirtyFlag为真，或者setAttribute或者removeAttribute己经被调用。
             2. 会话己经存在（己经被创建了）。
             3. 请求没有被“filter”属性所捕获。
             
             这个filter（过滤器）属性用于过滤那些不能修改会话的请求，因此，我们在这个请求结束后不复制
             会主。这个过滤器是消极的，例如，你向过滤器发送了一些东西，然后被过滤出去，在那个匹配的过
             滤器上不会发生复制。
             过滤器属性以分号为分隔，因此你不能换行，即使你想这么做。
             
             filter=".*\.gif;.*\.js;"意味着在进行以.gif和.js结束的URI请求后不复制会话。
        
        <!--
        <Cluster className="org.apache.catalina.cluster.tcp.SimpleTcpCluster"
                 managerClassName="org.apache.catalina.cluster.session.DeltaManager"
                 expireSessionsOnShutdown="false"
                 useDirtyFlag="true">

            <Membership
                className="org.apache.catalina.cluster.mcast.McastService"
                mcastAddr="228.0.0.4"
                mcastPort="45564"
                mcastFrequency="500"
                mcastDropTime="3000"/>

            <Receiver
                className="org.apache.catalina.cluster.tcp.ReplicationListener"
                tcpListenAddress="auto"
                tcpListenPort="4001"
                tcpSelectorTimeout="100"
                tcpThreadCount="6"/>

            <Sender
                className="org.apache.catalina.cluster.tcp.ReplicationTransmitter"
                replicationMode="pooled"/>

            <Valve className="org.apache.catalina.cluster.tcp.ReplicationValve"
                   filter=".*\.gif;.*\.js;.*\.jpg;.*\.htm;.*\.html;.*\.txt;"/>
        </Cluster>
        -->



        <!-- Normally, users must authenticate themselves to each web app
             individually. Uncomment the following entry if you would like
             a user to be authenticated the first time they encounter a
             resource protected by a security constraint, and then have that
             user identity maintained across *all* web applications contained
             in this virtual host. -->
        <!-- 通常，对每一个应用程序，用户必须逐个的认证他们自己。当他们遇到被
             某种安全约束所保护的资源时，如果你想让用户只在第一次被认证，那么，
             取消下面入口行的注释，然后在这个虚拟主机中包含这个用户的，横跨所有
             web应用程序的用户身份识别。-->
             
        <!--
        <Valve className="org.apache.catalina.authenticator.SingleSignOn"
                   debug="0"/>
        -->

        <!-- Access log processes all requests for this virtual host. By
             default, log files are created in the "logs" directory relative to
             $CATALINA_HOME. If you wish, you can specify a different
             directory with the "directory" attribute. Specify either a relative
             (to $CATALINA_HOME) or absolute path to the desired directory.
        -->
        <!-- 这个虚拟主机的所有请求的访问日志过程。默认情况下，日志文件被创建在相对
             于$CATALINA_HOME目录下的logs目录下。通过directory属性，你也可以指定一个
             不同的目录，如果你想的话。可以使用相对路径（相对于$CATALINA_HOME），也
             可以使用绝对路径。-->
        <!--
        <Valve className="org.apache.catalina.valves.AccessLogValve"
                 directory="logs" prefix="localhost_access_log." suffix=".txt"
                 pattern="common" resolveHosts="false"/>
        -->

        <!-- Logger shared by all Contexts related to this virtual host. By
             default (when using FileLogger), log files are created in the "logs"
             directory relative to $CATALINA_HOME. If you wish, you can specify
             a different directory with the "directory" attribute. Specify either a
             relative (to $CATALINA_HOME) or absolute path to the desired
             directory.-->
        <!-- 与这个虚拟主机有关系所有上下文所共享的日志。默认情况下（使用FileLogger时），
             日志文件被创建在相对于$CATALINA_HOME目录下的logs目录下。如果你希望，你也可
             以给directory指定一个不同的路径。可以使用相对于$CATALINA_HOME的相对路径，也
             可以使用绝对路径。
        <Logger className="org.apache.catalina.logger.FileLogger"
                 directory="logs" prefix="localhost_log." suffix=".txt"
            timestamp="true"/>

      </Host>

    </Engine>

  </Service>

</Server>

冉翔 2005-12-09 15:19 发表评论
]]>