2 Replies Latest reply on Oct 30, 2015 7:21 AM by Manu Forensic

    Problem with tcp-cluster in Wildfly 8.2.1.Final

    Manu Forensic Newbie

      Hi guys:


      I´ve some problems with a cluster tcp configuration with wildfly 8.2.1.Final. This is my standalone-ha.xml:



      standalone-ha.xml node1(
          <subsystem xmlns="urn:jboss:domain:jgroups:2.0" default-stack="tcpping">
              <stack name="udp">
                  <transport type="UDP" socket-binding="jgroups-udp"/>
                  <protocol type="PING"/>
                  <protocol type="MERGE3"/>
                  <protocol type="FD_SOCK" socket-binding="jgroups-udp-fd"/>
                  <protocol type="FD_ALL"/>
                  <protocol type="VERIFY_SUSPECT"/>
                  <protocol type="pbcast.NAKACK2"/>
                  <protocol type="UNICAST3"/>
                  <protocol type="pbcast.STABLE"/>
                  <protocol type="pbcast.GMS"/>
                  <protocol type="UFC"/>
                  <protocol type="MFC"/>
                  <protocol type="FRAG2"/>
                  <protocol type="RSVP"/>
              <stack name="tcpping">
                  <transport type="TCP" socket-binding="jgroups-tcp"/>
                  <protocol type="TCPPING">
                      <property name="initial_hosts">[7600],[7600]</property>
                      <property name="num_initial_members">0</property>
                      <property name="port_range">0</property>
                      <property name="timeout">2000</property>
                  <protocol type="MPING" socket-binding="jgroups-mping"/>
                  <protocol type="MERGE2"/>
                  <protocol type="FD_SOCK" socket-binding="jgroups-tcp-fd"/>
                  <protocol type="FD"/>
                  <protocol type="VERIFY_SUSPECT"/>
                  <protocol type="pbcast.NAKACK2"/>
                  <protocol type="UNICAST3"/>
                  <protocol type="pbcast.STABLE"/>
                  <protocol type="pbcast.GMS"/>
                  <protocol type="MFC"/>
                  <protocol type="FRAG2"/>
                  <protocol type="RSVP"/>



      Now when I try to start server in standalone mode:# ./standalone.sh -c standalone-ha.xml  -b &


      Server sttart

      11:08:35,004 INFO  [org.jboss.modules] (main) JBoss Modules version 1.3.3.Final

      11:08:35,642 INFO  [org.jboss.msc] (main) JBoss MSC version 1.2.2.Final

      11:08:35,795 INFO  [org.jboss.as] (MSC service thread 1-2) JBAS015899: WildFly 8.2.1.Final "Tweek" starting

      11:08:40,269 INFO  [org.jboss.as.server] (Controller Boot Thread) JBAS015888: Creating http management service using socket-binding (management-http)

      11:08:40,321 INFO  [org.xnio] (MSC service thread 1-2) XNIO version 3.3.0.Final

      11:08:40,349 INFO  [org.xnio.nio] (MSC service thread 1-2) XNIO NIO Implementation Version 3.3.0.Final

      11:08:40,415 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 34) JBAS010280: Activating Infinispan subsystem.

      11:08:40,523 INFO  [org.jboss.as.naming] (ServerService Thread Pool -- 44) JBAS011800: Activating Naming Subsystem

      11:08:40,529 WARN  [org.jboss.as.txn] (ServerService Thread Pool -- 50) JBAS010153: Node identifier property is set to the default value. Please make sure it is unique.

      11:08:40,525 INFO  [org.jboss.as.security] (ServerService Thread Pool -- 49) JBAS013171: Activating Security Subsystem

      11:08:40,521 INFO  [org.jboss.as.jsf] (ServerService Thread Pool -- 41) JBAS012615: Activated the following JSF Implementations: [main]

      11:08:40,634 INFO  [org.wildfly.extension.undertow] (ServerService Thread Pool -- 51) JBAS017502: Undertow 1.1.8.Final starting

      11:08:40,635 INFO  [org.jboss.as.clustering.jgroups] (ServerService Thread Pool -- 38) JBAS010260: Activating JGroups subsystem.

      11:08:40,644 INFO  [org.wildfly.extension.io] (ServerService Thread Pool -- 33) WFLYIO001: Worker 'default' has auto-configured to 4 core threads with 32 task threads based on your 2 available processors

      11:08:40,665 INFO  [org.jboss.as.webservices] (ServerService Thread Pool -- 52) JBAS015537: Activating WebServices Extension

      11:08:40,818 INFO  [org.jboss.remoting] (MSC service thread 1-2) JBoss Remoting version 4.0.7.Final

      11:08:40,845 INFO  [org.wildfly.extension.undertow] (MSC service thread 1-3) JBAS017502: Undertow 1.1.8.Final starting

      11:08:41,387 INFO  [org.jboss.as.security] (MSC service thread 1-1) JBAS013170: Current PicketBox version=4.0.21.Final

      11:08:41,586 INFO  [org.jboss.as.mail.extension] (MSC service thread 1-1) JBAS015400: Bound mail session [java:jboss/mail/Default]

      11:08:41,586 INFO  [org.jboss.as.naming] (MSC service thread 1-4) JBAS011802: Starting Naming Service

      11:08:41,650 INFO  [org.jboss.as.connector.logging] (MSC service thread 1-4) JBAS010408: Starting JCA Subsystem (IronJacamar 1.1.9.Final)

      11:08:41,882 INFO  [org.jboss.as.connector.subsystems.datasources] (ServerService Thread Pool -- 29) JBAS010403: Deploying JDBC-compliant driver class org.h2.Driver (version 1.3)

      11:08:41,947 INFO  [org.wildfly.extension.undertow] (ServerService Thread Pool -- 51) JBAS017527: Creating file handler for path /opt/jcnodo11-8PRE/welcome-content

      11:08:41,954 INFO  [org.jboss.as.connector.deployers.jdbc] (MSC service thread 1-3) JBAS010417: Started Driver service with driver-name = h2

      11:08:42,431 INFO  [org.wildfly.extension.undertow] (MSC service thread 1-3) JBAS017525: Started server default-server.

      11:08:42,772 INFO  [org.wildfly.extension.undertow] (MSC service thread 1-3) JBAS017519: Undertow AJP listener ajp listening on /

      11:08:42,801 INFO  [org.wildfly.extension.undertow] (MSC service thread 1-2) JBAS017519: Undertow HTTP listener default listening on /

      11:08:42,818 INFO  [org.jboss.modcluster] (ServerService Thread Pool -- 54) MODCLUSTER000001: Initializing mod_cluster version 1.3.0.Final

      11:08:42,828 INFO  [org.wildfly.extension.undertow] (MSC service thread 1-1) JBAS017531: Host default-host starting

      11:08:42,885 INFO  [org.jboss.modcluster] (ServerService Thread Pool -- 54) MODCLUSTER000032: Listening to proxy advertisements on /

      11:08:43,482 INFO  [org.jboss.as.connector.subsystems.datasources] (MSC service thread 1-3) JBAS010400: Bound data source [java:jboss/datasources/ExampleDS]

      11:08:44,032 INFO  [org.jboss.as.server.deployment.scanner] (MSC service thread 1-1) JBAS015012: Started FileSystemDeploymentService for directory /opt/jcnodo11-8PRE/standalone/deployments

      11:08:44,418 INFO  [org.jboss.ws.common.management] (MSC service thread 1-2) JBWS022052: Starting JBoss Web Services - Stack CXF Server 4.3.2.Final

      11:08:44,681 INFO  [org.jboss.as] (Controller Boot Thread) JBAS015961: Http management interface listening on

      11:08:44,682 INFO  [org.jboss.as] (Controller Boot Thread) JBAS015951: Admin console listening on

      11:08:44,683 INFO  [org.jboss.as] (Controller Boot Thread) JBAS015874: WildFly 8.2.1.Final "Tweek" started in 10497ms - Started 195 of 325 services (165 services are lazy, passive or on-demand)



      It seems not starting a TCP socket in port 7600:



      # netstat -nlpt|grep java

      tcp        0      0    *                   LISTEN      30348/java

      tcp        0      0 *                   LISTEN      30348/java

      tcp        0      0 *                   LISTEN      30348/java



      However there is multicast in UDP:


      # netstat -nlpu|grep java

      udp        0      0 *                               30348/java



      Could someone help me with this issue?



      Best regards.

        • 1. Re: Problem with tcp-cluster in Wildfly 8.2.1.Final
          Jay SenSharma Master

          Please comment the MPING protocol section


               <!--<protocol type="MPING" socket-binding="jgroups-mping"/>-->


          The MPING, Uses its own IP multicast socket to send and receive discovery requests/responses.

          MPING (=Multicast PING) uses IP multicast to discover the initial membership.  TCP usually requires TCPPING, which has to list all cluster members explicitly, but MPING doesn't have this requirement.  MPING uses its own multicast socket for discovery.



          Couple of things to check :

          1). Are you running ha or full-ha profiles which has clustering ?

          2). Clustering services are by default made "Lazy", Means until you deploy a web application with web.xml  including  <distributable/>


               Else make is EAGER


          • 2. Re: Problem with tcp-cluster in Wildfly 8.2.1.Final
            Manu Forensic Newbie

            Thank you very much.


            It solved my problem.