[Netarchivesuite-users] Deployment on more machines part2

[aponb at gmx.at]

Next try to deploy the Suite on more than one machine.
This time I am using three machines, generating all files with 
DeployApplication using the following conf.xml

<?xml version="1.0" encoding="UTF-8"?>
<root xmlns="http://www.netarkivet.dk/schemas/deploy" 
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
    <installtype>test</installtype>
    <jmxMonitorRolePassword>test</jmxMonitorRolePassword>
    <largeIndexTimeout>432000000</largeIndexTimeout>
    <location name="test">
        <installdir>/home/netarchive</installdir>
        <admin>
            <host name="wc02">
                <jmxport>8100</jmxport><!-- arcrep -->
                <jmxport>8101</jmxport><!-- bamonsb -->
                <jmxport>8102</jmxport><!-- harvestdef -->
                <remotefilePort>8300</remotefilePort><!-- arcrep -->
                <remotefilePort>8301</remotefilePort><!-- bamonsb -->
                <remotefilePort>8302</remotefilePort><!-- harvestdef -->
            </host>
        </admin>
        <bitarchive>
            <host name="wc03">
                <jmxport>8100</jmxport><!-- bitarchive -->
                <remotefilePort>8300</remotefilePort><!-- bitarchive -->
                <filedir>/mnt/webarchiv</filedir>
            </host>
        </bitarchive>
        <harvesters>
            <host name="wc04">
                <lowpriport>8081</lowpriport>
                <highpriport>8082</highpriport>
                <jmxport>8100</jmxport><!-- harvester -->
                <jmxport>8101</jmxport><!-- sidekick -->
                <jmxport>8102</jmxport><!-- harvester -->
                <jmxport>8103</jmxport><!-- sidekick -->
                <remotefilePort>8300</remotefilePort><!-- harvester 8081 -->
                <remotefilePort>8301</remotefilePort><!-- sidekick  8081 -->
                <remotefilePort>8302</remotefilePort><!-- harvester 8082 -->
                <remotefilePort>8303</remotefilePort><!-- sidekick  8082 -->
                <heritrixGuiPort>8090</heritrixGuiPort>
                <heritrixJmxPort>8091</heritrixJmxPort>
                <heritrixGuiPort>8092</heritrixGuiPort>
                <heritrixJmxPort>8093</heritrixJmxPort>
            </host>
        </harvesters>
        <access>
            <host name="wc02">
                <port>8077</port>
                <jmxport>8103</jmxport><!-- viewerproxy -->
                <jmxport>8104</jmxport><!-- indexserver -->
                <remotefilePort>8303</remotefilePort><!-- viewerproxy -->
                <remotefilePort>8304</remotefilePort><!-- indexserver -->
            </host>
        </access>
        <jms>
            <host name="wc02">
                
<class>dk.netarkivet.common.distribute.JMSConnectionSunMQ</class>
            </host>
        </jms>
        <mail>
            <host name="examplesmtpserversb.netarkivet.dk">
                <sender>example at netarkivet.dk</sender>
                <receiver>example at netarkivet.dk</receiver>
            </host>
        </mail>
        <application type="indexserver" host="wc02"/>
    </location>
</root>



When I start the application on all machines, everything works fine and 
the System state doesn't report any problem.
But when a Job is created, the crawler is not starting due to an 
exception of the HarcestControllerServer:

18.06.2008 11:43:48 
dk.netarkivet.harvester.harvesting.JMXHeritrixController <init>
INFO: Starting Heritrix for job 312 of harvest 2 in 
/home/netarchive/TEST/harvester_8082/312_1213782227823
18.06.2008 11:44:20 
dk.netarkivet.harvester.harvesting.JMXHeritrixController cleanup
SCHWERWIEGEND: JMX error while cleaning up Heritrix controller
dk.netarkivet.common.exceptions.IOFailure: Failed to connect to URL 
service:jmx:rmi:///jndi/rmi://localhost:8091/jmxrmi after 14 attempts
        at 
dk.netarkivet.common.jmx.JMXUtils.getJMXConnector(JMXUtils.java:351)
        at 
dk.netarkivet.harvester.harvesting.JMXHeritrixController.getHeritrixJMXConnector(JMXHeritrixController.java:823)
        at 
dk.netarkivet.harvester.harvesting.JMXHeritrixController.executeHeritrixCommand(JMXHeritrixController.java:747)
        at 
dk.netarkivet.harvester.harvesting.JMXHeritrixController.cleanup(JMXHeritrixController.java:380)
        at 
dk.netarkivet.harvester.harvesting.HeritrixLauncher.doCrawl(HeritrixLauncher.java:215)
        at 
dk.netarkivet.harvester.harvesting.HarvestController.runHarvest(HarvestController.java:210)
        at 
dk.netarkivet.harvester.harvesting.distribute.HarvestControllerServer$HarvesterThread.run(HarvestControllerServer.java:632)
Caused by: java.io.IOException: Failed to retrieve RMIServer stub: 
javax.naming.ServiceUnavailableException [Root exception is 
java.rmi.ConnectException: Connection refused to host: localhost; nested 
exception is:
        java.net.ConnectException: Connection refused]
        at 
javax.management.remote.rmi.RMIConnector.connect(RMIConnector.java:323)
        at 
javax.management.remote.JMXConnectorFactory.connect(JMXConnectorFactory.java:248)
        at 
dk.netarkivet.common.jmx.JMXUtils.getJMXConnector(JMXUtils.java:338)
        ... 6 more

When I use  netstat -nlp to see which program is blocking port 8091 the 
port doesn't appear. All ports on all three machines are open and no 
firewall is running.
Do you have any idea about that problem?