Hello everybody,
I've searching around for the last two weeks trying to find a solution for the
following set-up. What I would like to achieve is a set-up where 4 JBoss
instances work together in a clustered environment. Two nodes are actually
serving client requests while two other nodes are being hot-standby being
Apache working as software loadbalancer. This works just great and is up and
running. So no difficulties concerning this matter.
Now I would like to extends the configuration by setting up a fifth JBoss
instance running the default configuration that needs to serve like a
centralized, cluster-wide cache. And this is where difficulties appear.
Browsing the Wiki has showed me (due to two articles written by Manik Surtani)
that I need to set-up a TreeCache on this fifth machine together with a
TcpCacheServer. Each node then synchronizes with the global cache by using the
TcpDelegatingCacheLoader. The TreeCache on JBoss instance 5 can then be backed
by a JDBCCacheLoader to persist the cached data.
As described in this article, I have installed a default JBoss instance wich I
extended with jboss-cache.jar and jgroups.jar in it's default lib folder to
provice caching functionality. I'm using JBossCache 1.3.0 together with the
JGroups version provided with JBoss 4.0.2 on a JBoss 4.0.2 configuration. In
fact, all my JBoss nodes are also running 4.0.2. I configured both the
TreeCache and the TcpCacheServer in a HttpCache-service.xml file (located
below) and copied this one to the deploy folder together with a mysql-ds.xml
file because my TreeCache uses the JDBCCacheLoader.
A first strange thing appeared when starting the AS. I got an error message
stating that the datasource was not bounded alltough it is available in the
deploy folder. As far as I know, xxx-ds.xml files are always deployed before
xxx-service.xml files under normal deployment order configuration. Can anyone
explain what the reason for this behaviour might be? Anyway I solved it by
using a deploy.last folder that contains my xxx-service.xml file which solved
the problem.
A second problem I have is with the TcpCacheServer. As soon as this service is
started by the AS, it seems to freeze. The web-console in fact is blocking, so
I dont' have the possibility to do anything with the default JBoss
configuration. Does anyone have the same behaviour in this circumstances? If
so, did anyone found a solution for this one. I know I must be doing something
wrong within this set-up but I can't figure out what. Any help would be greatly
appreciated off course and any assistance for finding a solution would result
in a neat article within the Wiki describing this entire configuration.
Below some of the configuration files I used within my attempt to set-up this
configuration.
The JBoss instance serving as cache has the following service definition file
called HttpCache-service.xml
| <server>
|
| <classpath codebase="./lib" archives="jboss-cache.jar, jgroups.jar"/>
|
| <mbean code="org.jboss.cache.TreeCache"
name="jboss.cache:service=HttpCache">
| <depends>jboss:service=Naming</depends>
| <depends>jboss:service=TransactionManager</depends>
|
| <attribute
name="TransactionManagerLookupClass">org.jboss.cache.JBossTransactionManagerLookup</attribute>
|
| <attribute name="NodeLockingScheme">PESSIMISTIC</attribute>
|
| <attribute name="IsolationLevel">REPEATABLE_READ</attribute>
|
| <attribute name="CacheMode">LOCAL</attribute>
|
| <attribute name="UseInterceptorMbeans">TRUE</attribute>
|
| <attribute name="LockAcquisitionTimeout">5000</attribute>
|
| <attribute name="CacheLoaderConfiguration">
| <config>
| <passivation>TRUE</passivation>
| <preload>/</preload>
| <shared>FALSE</shared>
| <cacheloader>
|
<class>org.jboss.cache.loader.JDBCCacheLoader</class>
| <properties>
|
cache.jdbc.datasource=java:/HttpCacheDS
| </properties>
|
<fetchPersistentState>true</fetchPersistentState>
|
<ignoreModifications>false</ignoreModifications>
| <async>false</async>
| </cacheloader>
| </config>
| </attribute>
| </mbean>
|
| <mbean code="org.jboss.cache.loader.tcp.TcpCacheServer"
| name="jboss.cache:service=TcpCacheServer">
| <depends optional-attribute-name="Cache"
proxy-type="attribute">jboss.cache:service=HttpCache</depends>-->
| <attribute
name="BindAddress">${jboss.bind.address:localhost}</attribute>
| <attribute name="Port">7500</attribute>
| <attribute name="MBeanServerName"></attribute>
| </mbean>
|
| </server>
Each JBoss cluster node (running the all configuration) has a modified
tc5-cluster-service.xml file that should (if I understand everything correct)
use the global cache to synchronize with:
| <server>
|
| <classpath codebase="./lib" archives="jboss-cache.jar, jgroups.jar"/>
|
| <mbean code="org.jboss.cache.TreeCache"
| name="jboss.cache:service=TomcatClusteringCache">
| <depends>jboss:service=Naming</depends>
| <depends>jboss:service=TransactionManager</depends>
|
| <attribute
name="TransactionManagerLookupClass">org.jboss.cache.JBossTransactionManagerLookup</attribute>
|
| <attribute name="IsolationLevel">REPEATABLE_READ</attribute>
|
| <attribute name="CacheMode">LOCAL</attribute>
|
| <attribute name="LockAcquisitionTimeout">10000</attribute>
|
| <attribute name="UseReplQueue">false</attribute>
|
| <attribute name="UseMarshalling">false</attribute>
|
| <attribute name="CacheLoaderConfiguration">
| <config>
| <shared>true</shared>
|
| <cacheloader>
|
<class>org.jboss.cache.loader.tcp.TcpDelegatingCacheLoader</class>
| <properties>
| host=sinfonix
| port=7500
| </properties>
| <async>true</async>
|
<fetchPersistentState>false</fetchPersistentState>
|
<ignoreModifications>false</ignoreModifications>
| </cacheloader>
| </config>
| </attribute>
|
| </mbean>
|
| </server>
|
My datasource configuration file looks like the following
<datasources>
| <local-tx-datasource>
| <jndi-name>HttpCacheDS</jndi-name>
| <connection-url>jdbc:mysql://localhost:3306/jbosscache</connection-url>
| <driver-class>com.mysql.jdbc.Driver</driver-class>
| <user-name>xxxxx</user-name>
| <password>xxxxx</password>
|
<exception-sorter-class-name>org.jboss.resource.adapter.jdbc.vendor.MySQLExceptionSorter</exception-sorter-class-name>
| <!-- sql to call when connection is created
| <new-connection-sql>some arbitrary sql</new-connection-sql>
| -->
| <!-- sql to call on an existing pooled connection when it is obtained
from pool
| <check-valid-connection-sql>some arbitrary
sql</check-valid-connection-sql>
| -->
|
| <!-- corresponding type-mapping in the standardjbosscmp-jdbc.xml
(optional) -->
| <metadata>
| <type-mapping>mySQL</type-mapping>
| </metadata>
| </local-tx-datasource>
| </datasources>
Can anyone tell me whether or not this is the appropriate approach to achieve
the configuration mentionned above or am I completely wrong and is there
another approach that must be followed.
Any reference containing information how to create this kind of set-up would be
greatly appreciated. Many thanks in advance to all the people out there,
certainly the few who found enough energy to read this question ;-)
Cheers,
Kurt
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3955645#3955645
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3955645
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
JBoss-user mailing list
JBoss-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jboss-user
