[jboss-cvs] JBoss Messaging SVN: r6812 - trunk/docs/user-manual/en/modules.
jboss-cvs-commits at lists.jboss.org
jboss-cvs-commits at lists.jboss.org
Fri May 15 12:41:02 EDT 2009
Author: timfox
Date: 2009-05-15 12:41:02 -0400 (Fri, 15 May 2009)
New Revision: 6812
Modified:
trunk/docs/user-manual/en/modules/clusters.xml
Log:
completed chapter
Modified: trunk/docs/user-manual/en/modules/clusters.xml
===================================================================
--- trunk/docs/user-manual/en/modules/clusters.xml 2009-05-15 16:35:26 UTC (rev 6811)
+++ trunk/docs/user-manual/en/modules/clusters.xml 2009-05-15 16:41:02 UTC (rev 6812)
@@ -498,7 +498,7 @@
pairs, these pairs are then used by the client connection load balancing policy
on the client side when creating connections to the cluster.</para>
<para>If you're using JMS but you're not using JNDI then you can also specify the
- list of [connector, backup connector] pairs directly when instantiating the
+ list of [connector, backup connector] pairs directly when instantiating the
<literal>JBossConnectionFactory</literal>, here's an
example:<programlisting>List<Pair<TransportConfiguration, TransportConfiguration>> serverList = new ArrayList<Pair<TransportConfiguration, TransportConfiguration>>();
@@ -521,13 +521,12 @@
this factory will create connections according to the client connection load
balancing policy applied to that list of servers.</para>
</section>
-
- <section>
- <title>Specifying List of Servers using the Core API</title>
- <para>If you're using the Core API you can also specify the list of servers directly
- when creating the <literal>ClientSessionFactory</literal> instance. Here's an
- example:</para>
- <programlisting>List<Pair<TransportConfiguration, TransportConfiguration>> serverList = new ArrayList<Pair<TransportConfiguration, TransportConfiguration>>();
+ <section>
+ <title>Specifying List of Servers using the Core API</title>
+ <para>If you're using the Core API you can also specify the list of servers directly
+ when creating the <literal>ClientSessionFactory</literal> instance. Here's an
+ example:</para>
+ <programlisting>List<Pair<TransportConfiguration, TransportConfiguration>> serverList = new ArrayList<Pair<TransportConfiguration, TransportConfiguration>>();
serverList.add(new Pair<TransportConfiguration, TransportConfiguration>(liveTC0, backupTC0));
serverList.add(new Pair<TransportConfiguration, TransportConfiguration>(liveTC1, backupTC1));
@@ -538,18 +537,42 @@
ClientSession sesison1 = factory.createClientSession(...);
ClientSession session2 = factory.createClientSession(...);</programlisting>
- <para>In the above snippet we create a list of pairs of <literal
- >TransportConfiguration</literal> objects. Each <literal
- >TransportConfiguration</literal> object contains knowledge of how to make a
- connection to a specific server. For more information on this please see the chapter
- on connectors [LINK].</para>
- <para>A <literal>ClientSessionFactoryImpl</literal> instance is then created passing the
- list of servers in the constructor. Any sessions subsequently created by this
- factory will create sessions according to the client connection load balancing
- policy applied to that list of servers.</para>
+ <para>In the above snippet we create a list of pairs of <literal
+ >TransportConfiguration</literal> objects. Each <literal
+ >TransportConfiguration</literal> object contains knowledge of how to make a
+ connection to a specific server. For more information on this please see the
+ chapter on connectors [LINK].</para>
+ <para>A <literal>ClientSessionFactoryImpl</literal> instance is then created passing
+ the list of servers in the constructor. Any sessions subsequently created by
+ this factory will create sessions according to the client connection load
+ balancing policy applied to that list of servers.</para>
+ </section>
</section>
+ <section>
+ <title>Specifying List of Servers to form a Cluster</title>
+ <para>Let's take a look at an example where each cluster connection is defined for a
+ symmetric cluster, but we're not using discovery for each node to discover its
+ neighbours, instead we'll configure each cluster connection to have explicit
+ knowledge of all the other nodes in the cluster.</para>
+ <para>Here's an example cluster connection definition showing that:</para>
+ <programlisting><cluster-connections>
+ <cluster-connection name="my-explicit-cluster">
+ <address>jms</address>
+ <connector-ref connector-name="my-connector1" backup-connector-name="my-backup-connector1"/>
+ <connector-ref connector-name="my-connector2" backup-connector-name="my-backup-connector2"/>
+ <connector-ref connector-name="my-connector3" backup-connector-name="my-backup-connector3"/>
+ </cluster-connection>
+</cluster-connections></programlisting>
+ <para>The <literal>cluster-connection</literal> element can contain zero or more
+ <literal>connector-ref</literal> elements, each one of which specifies a
+ <literal>connector-name</literal> attribute and an optional <literal
+ >backup-connector-name</literal> attribute. The <literal
+ >connector-name</literal> attribute references a connector defined in <literal
+ >jbm-configuration.xml</literal> which will be used as a live connector. The
+ <literal>backup-connector-name</literal> is optional, and if specified it also
+ references a connector defined in <literal>jbm-configuration.xml</literal>. For more
+ information on connectors please see chapter [LINK].</para>
</section>
-
</section>
<section>
<title>Message Redistribution</title>
More information about the jboss-cvs-commits
mailing list