jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dominique Pfister" <dominique.pfis...@day.com>
Subject Re: jackrabbit cluster
Date Mon, 19 May 2008 11:46:17 GMT
Hi Rocco,

I noticed that the JDBC URLs on both machine 1 and machine 2 contain
"localhost" as host information which probably denotes separate MySQL
database servers running on machine 1 and machine 2, respectively.
Please note, that clustering requires a "shared" database, i.e. both
cluster instances should connect to the *same* database server. The
informational message, indicating that the content is not available on
machine 2, hints at the same problem.

Kind regards
Dominique

On 5/19/08, Rocco Russo <rocco.russo@eng.it> wrote:
> Hi all,
>
>  I set up a  cluster configuration (2 tomcats on 2 different machines) with a
>  SHARED repository on MySQL db.
>
>
>
>  My environment is:
>
>
>
>  on Machine 1
>
>
>
>  jdk: 1.5.0_14
>
>  server1: tomcat 5.5.26
>
>  app: jackrabbit-webapp-1.4
>
>  repository.xml in cluster configuration :
>
>
>
>             <Cluster id="node_1">
>
>                         <Journal
>  class="org.apache.jackrabbit.core.journal.DatabaseJournal">
>
>                                    <!-local revision file -->
>
>  <param name="revision" value="${rep.home}/revision"/>
>
>                                    <param name="driver"
>  value="com.mysql.jdbc.Driver"/>
>
>                                    <param name="url"
>  value="jdbc:mysql://localhost:3306/jcr_repo?useUnicode=true&amp;characterEnc
>  oding=UTF-8" />
>
>                                    <param name="user" value="root" />
>
>                                    <param name="password" value="root" />
>
>                                    <param name="schema" value="mysql"/>
>
>                                    <param name="schemaObjectPrefix"
>  value="J_C_"/>
>
>                         </Journal>
>
>             </Cluster>
>
>
>
>
>
>  on Machine 2
>
>
>
>  jdk: 1.5.0_14
>
>  server2: tomcat 5.5.26
>
>  app: jackrabbit-webapp-1.4
>
>  repository.xml in cluster configuration :
>
>
>
>             <Cluster id="node_2" >
>
>                         <Journal
>  class="org.apache.jackrabbit.core.journal.DatabaseJournal">
>
>                                    <!-local revision file -->
>
>  <param name="revision" value="${rep.home}/revision"/>
>
>                                    <param name="driver"
>  value="com.mysql.jdbc.Driver"/>
>
>                                    <param name="url"
>  value="jdbc:mysql://localhost:3306/jcr_repo?useUnicode=true&amp;characterEnc
>  oding=UTF-8" />
>
>                                    <param name="user" value="root" />
>
>                                    <param name="password" value="root" />
>
>                                    <param name="schema" value="mysql"/>
>
>                                    <param name="schemaObjectPrefix"
>  value="J_C_"/>
>
>                         </Journal>
>
>             </Cluster>
>
>
>
>  When I start Tomcat 1 and Tomcat 2 for the first time I can see all nodes
>  stored in the shared repository browsing  from the server 1 ad also from the
>  server 2.
>
>  If I upload a file from server 1, I can see the new file only from the
>  server 1, but I cannot see the new file browsing server 2.
>
>
>
>  In the logs I can see this situation during file upload:
>
>
>
>  1)       The Server 1 appends revision correctly:
>
>  19.05.2008 12:15:03 *INFO * ClusterNode: Appended revision: 1
>  (ClusterNode.java, line 616)
>
>
>
>  2)       A row in the journal table is added
>
>
>
>  3)       The Server 2  synchronizes the appended revision
>
>  19.05.2008 12:14:58 *INFO * ClusterNode: Processing revision: 1
>  (ClusterNode.java, line 963)
>
>  19.05.2008 12:14:58 *INFO * SearchManager: Node no longer available
>  98a6e6ba-0965-4bd5-a531-ed8e302cf6f4, skipped. (SearchManager.java, line
>  467)
>
>  19.05.2008 12:14:58 *INFO * SearchManager: Node no longer available
>  ae22c5ac-ba10-469c-a9ed-48fd4866fd4e, skipped. (SearchManager.java, line
>  467)
>
>  19.05.2008 12:14:58 *INFO * AbstractJournal: Synchronized to revision: 1
>  (AbstractJournal.java, line 213)
>
>
>
>  But the result is that if I browse the node tree form server 2 I cannot see
>  the new uploaded file.
>
>  I have noticed that during synchronization there are these log info:
>
>  19.05.2008 12:14:58 *INFO * SearchManager: Node no longer available
>  98a6e6ba-0965-4bd5-a531-ed8e302cf6f4, skipped. (SearchManager.java, line
>  467)
>
>  19.05.2008 12:14:58 *INFO * SearchManager: Node no longer available
>  ae22c5ac-ba10-469c-a9ed-48fd4866fd4e, skipped. (SearchManager.java, line
>  467)
>
>  Is it a problem when these logs appear? What do these logs mean?  Am I doing
>  something wrong in the configuration of cluster? Have I missed  something?
>
>
>
>  Any suggestion is appreciated.
>
>  Thanks in advanced
>
>
>
>
>
>
>
>
>
>

Mime
View raw message