It is difficult to see which data the indexserver cannot retrieve when generating the deduplication index

Install the netarchivesuite Go the machine (Machine X)where the IndexserverApplication is deployed. Configure the IndexServerApplication to use a localArcRepositoryClient Copy a few (10) metadata-arc-files to the storage used by localArcRepositoryClient Send a message to the IndexServerApplication asking for a dedup-index for 14 jobs, where only the 10 are available in storage. (Specific java-class are available that does this in the tests part of netarchivesuite trunk (r2177: trunk/tests/SendDedupIndexRequestToIndexserver.java) copy SendDedupIndexRequestToIndexserver.java to machine X echo on machine X export INSTALLDIR=/home/test/???? In file $INSTALLDIR/conf/settings_IndexServerApplication.xml insert the following <settings><common> ... <arcrepositoryClient> < class& gt;dk.netarkivet.common.distribute.arcrepository.LocalArcRepositoryClient</ class& gt; <fileDir>/home/test/metadataArkiv/</fileDir> </arcrepositoryClient> ... </common> </settings> javac -cp $INSTALLDIR/lib/dk.netarkivet.archive.jar SendDedupIndexRequestToIndexserver.java Call program using settings of viewerproxy, so the correct JMSbroker is used export CLASSPATH=$INSTALLDIR/lib/dk.netarkivet.archive.jar:/home/test/ java -Ddk.netarkivet.settings.file=$INSTALLDIR/conf/settings_ViewerProxyApplication.xml SendDedupIndexRequestToIndexserver <file-with-jobids>