jboss - Web 应用程序不会加入 Infinispan 集群

Question

我最近一直在玩 Infinispan，之前没有使用过 Infinispan。

我遇到了一个有趣的问题，我想知道是否有人能够阐明它。

我有一个独立的 Java 应用程序 (GridGrabber.jar)，它捆绑了 Infinispan jar ，并有一个类可以从网格中添加/删除和列出项目。在应用程序中，我创建了 CacheManager，如下所示:

DefaultCacheManager m = new DefaultCacheManager("cluster.xml");

该应用程序使用 java -jar GridGrabber.jar 在命令行上运行并且工作正常，当我在终端启动两个实例时，它们都聚集在一起，我可以看到来自两个节点的网格:

Sam@macbook $ java -jar GridGrabber.jar -Djava.net.preferIPv4Stack=true -Djgroups.bind_addr=127.0.0.1
Apr 16, 2012 9:50:04 PM org.infinispan.remoting.transport.jgroups.JGroupsTransport start
INFO: ISPN000078: Starting JGroups Channel
Apr 16, 2012 9:50:04 PM org.infinispan.remoting.transport.jgroups.JGroupsTransport buildChannel
INFO: ISPN000088: Unable to use any JGroups configuration mechanisms provided in properties {}. Using default JGroups configuration!
Apr 16, 2012 9:50:05 PM org.jgroups.logging.JDKLogImpl warn
WARNING: receive buffer of socket java.net.DatagramSocket@bc794cc was set to 20MB, but the OS only allocated 65.51KB. This might lead to performance problems. Please set your max receive buffer in the OS correctly (e.g. net.core.rmem_max on Linux)
Apr 16, 2012 9:50:05 PM org.jgroups.logging.JDKLogImpl warn
WARNING: receive buffer of socket java.net.MulticastSocket@731de9b was set to 25MB, but the OS only allocated 65.51KB. This might lead to performance problems. Please set your max receive buffer in the OS correctly (e.g. net.core.rmem_max on Linux)
Apr 16, 2012 9:50:06 PM org.infinispan.remoting.transport.jgroups.JGroupsTransport viewAccepted
INFO: ISPN000094: Received new cluster view: [Samuel-Weavers-MacBook-Pro-62704|1] [Samuel-Weavers-MacBook-Pro-62704, Samuel-Weavers-MacBook-Pro-3360]
Apr 16, 2012 9:50:06 PM org.infinispan.remoting.transport.jgroups.JGroupsTransport startJGroupsChannelIfNeeded
INFO: ISPN000079: Cache local address is Samuel-Weavers-MacBook-Pro-3360, physical addresses are [fe80:0:0:0:e2f8:47ff:fe06:109a%5:51422]
Apr 16, 2012 9:50:06 PM org.jgroups.logging.JDKLogImpl warn
WARNING: Samuel-Weavers-MacBook-Pro-3360: not member of view [Samuel-Weavers-MacBook-Pro-62704|2]; discarding it
Apr 16, 2012 9:50:06 PM org.infinispan.factories.GlobalComponentRegistry start
INFO: ISPN000128: Infinispan version: Infinispan 'Brahma' 5.1.2.FINAL
Apr 16, 2012 9:50:06 PM org.infinispan.jmx.CacheJmxRegistration start
INFO: ISPN000031: MBeans were successfully registered to the platform mbean server.
Apr 16, 2012 9:50:12 PM org.infinispan.remoting.transport.jgroups.JGroupsTransport viewAccepted
INFO: ISPN000093: Received new, MERGED cluster view: MergeView::[Samuel-Weavers-MacBook-Pro-3360|3] [Samuel-Weavers-MacBook-Pro-3360, Samuel-Weavers-MacBook-Pro-62704], subgroups=[Samuel-Weavers-MacBook-Pro-62704|1] [Samuel-Weavers-MacBook-Pro-3360], [Samuel-Weavers-MacBook-Pro-62704|2] [Samuel-Weavers-MacBook-Pro-62704]

现在，这里是有趣的地方。我还有一个部署在 AS7 上的 Web 应用程序(servlet + JSP)，其目的是提供进入网格的数据的实时更新网页。 servlet 调用与 GridGrabber 基本相同的代码并使用相同的 cluster.xml。

当它启动时，一切都部署正常，但由于某种原因它没有选择网格中的其他节点，尽管它们都在同一台机器上运行并且相互看到没有问题。部署消息是:

    17:06:16,889 INFO  [org.jboss.modules] JBoss Modules version 1.1.1.GA
17:06:17,107 INFO  [org.jboss.msc] JBoss MSC version 1.0.2.GA
17:06:17,155 INFO  [org.jboss.as] JBAS015899: JBoss AS 7.1.1.Final "Brontes" starting
17:06:17,972 INFO  [org.xnio] XNIO Version 3.0.3.GA
17:06:17,972 INFO  [org.jboss.as.server] JBAS015888: Creating http management service using socket-binding (management-http)
17:06:17,981 INFO  [org.xnio.nio] XNIO NIO Implementation Version 3.0.3.GA
17:06:17,990 INFO  [org.jboss.remoting] JBoss Remoting version 3.2.3.GA
17:06:18,006 INFO  [org.jboss.as.logging] JBAS011502: Removing bootstrap log handlers
17:06:18,013 INFO  [org.jboss.as.configadmin] (ServerService Thread Pool -- 26) JBAS016200: Activating ConfigAdmin Subsystem
17:06:18,017 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 31) JBAS010280: Activating Infinispan subsystem.
17:06:18,038 INFO  [org.jboss.as.naming] (ServerService Thread Pool -- 38) JBAS011800: Activating Naming Subsystem
17:06:18,062 INFO  [org.jboss.as.osgi] (ServerService Thread Pool -- 39) JBAS011940: Activating OSGi Subsystem
17:06:18,070 INFO  [org.jboss.as.security] (ServerService Thread Pool -- 44) JBAS013101: Activating Security Subsystem
17:06:18,077 INFO  [org.jboss.as.webservices] (ServerService Thread Pool -- 48) JBAS015537: Activating WebServices Extension
17:06:18,103 INFO  [org.jboss.as.security] (MSC service thread 1-8) JBAS013100: Current PicketBox version=4.0.7.Final
17:06:18,121 INFO  [org.jboss.as.connector] (MSC service thread 1-3) JBAS010408: Starting JCA Subsystem (JBoss IronJacamar 1.0.9.Final)
17:06:18,134 INFO  [org.jboss.as.naming] (MSC service thread 1-4) JBAS011802: Starting Naming Service
17:06:18,173 INFO  [org.jboss.as.mail.extension] (MSC service thread 1-4) JBAS015400: Bound mail session [java:jboss/mail/Default]
17:06:18,241 INFO  [org.jboss.ws.common.management.AbstractServerConfig] (MSC service thread 1-7) JBoss Web Services - Stack CXF Server 4.0.2.GA
17:06:18,290 INFO  [org.jboss.as.connector.subsystems.datasources] (ServerService Thread Pool -- 27) JBAS010403: Deploying JDBC-compliant driver class org.h2.Driver (version 1.3)
17:06:18,407 INFO  [org.apache.coyote.http11.Http11Protocol] (MSC service thread 1-7) Starting Coyote HTTP/1.1 on http-localhost-127.0.0.1-8080
17:06:18,802 INFO  [org.jboss.as.connector.subsystems.datasources] (MSC service thread 1-7) JBAS010400: Bound data source [java:jboss/datasources/ExampleDS]
17:06:18,885 INFO  [org.jboss.as.server.deployment.scanner] (MSC service thread 1-7) JBAS015012: Started FileSystemDeploymentService for directory /Users/Sam/Downloads/jboss-as-7.1.1.Final/standalone/deployments
17:06:18,893 INFO  [org.jboss.as.server.deployment.scanner] (DeploymentScanner-threads - 1) JBAS015003: Found TwitterWebApp.war in deployment directory. To trigger deployment create a file called TwitterWebApp.war.dodeploy
17:06:18,894 INFO  [org.jboss.as.server.deployment.scanner] (DeploymentScanner-threads - 1) JBAS015003: Found SBTwitterStreamGui.war in deployment directory. To trigger deployment create a file called SBTwitterStreamGui.war.dodeploy
17:06:18,920 INFO  [org.jboss.as.remoting] (MSC service thread 1-5) JBAS017100: Listening on localhost/127.0.0.1:4447
17:06:18,920 INFO  [org.jboss.as.remoting] (MSC service thread 1-2) JBAS017100: Listening on /127.0.0.1:9999
17:06:19,018 INFO  [org.jboss.as] (Controller Boot Thread) JBAS015951: Admin console listening on http://127.0.0.1:9990
17:06:19,019 INFO  [org.jboss.as] (Controller Boot Thread) JBAS015874: JBoss AS 7.1.1.Final "Brontes" started in 2479ms - Started 133 of 208 services (74 services are passive or on-demand)
17:06:19,031 INFO  [org.jboss.as.server.deployment] (MSC service thread 1-5) JBAS015876: Starting deployment of "TwitterWebApp.war"
17:06:19,031 INFO  [org.jboss.as.server.deployment] (MSC service thread 1-4) JBAS015876: Starting deployment of "SBTwitterStreamGui.war"
17:06:19,426 INFO  [org.jboss.web] (MSC service thread 1-7) JBAS018210: Registering web context: /SBTwitterStreamGui
17:06:20,267 INFO  [org.jboss.web] (MSC service thread 1-3) JBAS018210: Registering web context: /TwitterWebApp
17:06:20,313 INFO  [org.jboss.as.server] (DeploymentScanner-threads - 2) JBAS018559: Deployed "TwitterWebApp.war"
17:06:20,313 INFO  [org.jboss.as.server] (DeploymentScanner-threads - 2) JBAS018559: Deployed "SBTwitterStreamGui.war"
17:06:23,399 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (http-localhost-127.0.0.1-8080-1) ISPN000078: Starting JGroups Channel
17:06:23,400 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (http-localhost-127.0.0.1-8080-1) ISPN000088: Unable to use any JGroups configuration mechanisms provided in properties {}. Using default JGroups configuration!
17:06:23,805 WARN  [org.jgroups.protocols.UDP] (http-localhost-127.0.0.1-8080-1) receive buffer of socket java.net.DatagramSocket@4ad6be89 was set to 20MB, but the OS only allocated 65.51KB. This might lead to performance problems. Please set your max receive buffer in the OS correctly (e.g. net.core.rmem_max on Linux)
17:06:23,807 WARN  [org.jgroups.protocols.UDP] (http-localhost-127.0.0.1-8080-1) receive buffer of socket java.net.MulticastSocket@7bb28246 was set to 25MB, but the OS only allocated 65.51KB. This might lead to performance problems. Please set your max receive buffer in the OS correctly (e.g. net.core.rmem_max on Linux)
17:06:26,824 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (http-localhost-127.0.0.1-8080-1) ISPN000094: Received new cluster view: [Samuel-Weavers-MacBook-Pro-102|0] [Samuel-Weavers-MacBook-Pro-102]
17:06:26,946 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (http-localhost-127.0.0.1-8080-1) ISPN000079: Cache local address is Samuel-Weavers-MacBook-Pro-102, physical addresses are [10.36.5.166:64952]
17:06:26,955 INFO  [org.infinispan.factories.GlobalComponentRegistry] (http-localhost-127.0.0.1-8080-1) ISPN000128: Infinispan version: Infinispan 'Brahma' 5.1.2.FINAL
17:06:26,956 WARN  [org.infinispan.config.TimeoutConfigurationValidatingVisitor] (http-localhost-127.0.0.1-8080-1) ISPN000148: Invalid <transport>: distributedSyncTimout value of 240000. It can not be higher than <stateRetrieval>:timeout which is 120000
17:06:27,045 INFO  [org.infinispan.jmx.CacheJmxRegistration] (http-localhost-127.0.0.1-8080-1) ISPN000031: MBeans were successfully registered to the platform mbean server.
17:06:27,061 INFO  [stdout] (http-localhost-127.0.0.1-8080-1) Cache Size: 0
17:06:27,062 INFO  [stdout] (http-localhost-127.0.0.1-8080-1) Cluster Name: demoCluster
17:06:27,063 INFO  [stdout] (http-localhost-127.0.0.1-8080-1) Cluster Size: 1
17:06:27,063 INFO  [stdout] (http-localhost-127.0.0.1-8080-1) Cluster Members: [Samuel-Weavers-MacBook-Pro-102]
17:06:27,064 INFO  [stdout] (http-localhost-127.0.0.1-8080-1) Cache Manager: org.infinispan.manager.DefaultCacheManager@3c818c4@Address:Samuel-Weavers-MacBook-Pro-102

有谁知道 AS7 部署看不到网格中的其他节点是否有原因？在我看来，它应该可以工作，因为它们都在寻找相同的集群名称 (demoCluster)，并且它们都使用相同的 cluster.xml。 cluster.xml如下:

<infinispan
      xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
      xsi:schemaLocation="urn:infinispan:config:5.1 http://www.infinispan.org/schemas/infinispan-config-5.1.xsd"
      xmlns="urn:infinispan:config:5.1">

   <global>
      <transport clusterName="demoCluster"/>
      <globalJmxStatistics enabled="true"/>
   </global>

   <default>
      <jmxStatistics enabled="true"/>
      <clustering mode="distribution">
         <hash numOwners="2" rehashRpcTimeout="120000"/>
         <sync/>
      </clustering>
   </default>
</infinispan>

我从部署消息中注意到的一件事是 Web 应用程序似乎使用 ipv4 地址，而 jar 似乎使用 ipv6 地址与 JGroups，但我不确定这是否是根本原因。任何帮助表示赞赏。

Answer 1

我似乎已经通过执行以下操作解决了这个问题:

关于 JGroups ipv4/ipv6 地址，我是正确的，所以我通过提供以下 JVM 参数强制独立应用程序使用 IPv4:-Djava.net.preferIPv4Stack=true

这里要注意一件事——我从终端运行独立的 jar，并且 Web 应用程序是从 Eclipse 中部署到 AS7 上的。从终端运行 jar 时传递 JVM 参数不成功。但是，也从 Eclipse 内部运行应用程序，并在运行配置“运行”菜单中为其提供 JVM 参数 -> “运行配置” -> 在 Java 应用程序下添加新的(或编辑现有应用程序) -> 参数 -> VM 参数-> 在框中粘贴 preferIPv4stack 参数。

现在，当从 Eclipse 中同时运行独立应用程序和 Web 应用程序时，它们应该加入同一个集群并开始相互发送消息。

附带说明 - 在说“消息已丢弃 - 发件人不在窗口中”之后我遇到了 JGroups 错误。经过进一步调查，事实证明连接到(公司)VPN 导致消息被丢弃。我一关闭 VPN，它就开始正常工作。

希望这可以帮助遇到同样问题的人。