Hi,
Im using solr 1.4.1 version and I have the following configuration for
replication in master and slave
Solrconfig.xml (master)
<requestHandler name="/replication" class="solr.ReplicationHandler" >
<lst name="master">
<str name="replicateAfter">commit</str>
<str name="replicateAfter">startup</str>
<str name="confFiles">schema.xml,stopwords.txt</str>
</lst>
</requestHandler>
SolrConfig.xml (slave)
<requestHandler name="/replication" class="solr.ReplicationHandler" >
<lst name="slave">
<str
name="masterUrl">http://localhost:8982/solr/corez/replication</str>
</lst>
</requestHandler>
The master is used to index the data and the search requests will be given
to the slave. Tha data can be also pushed into slave from the UI (on need
basis)
At times, we find that the replicated folder in slave is of index.timestamp
format and when this occurs, some segements are not getting replicated to
slave. Due to this, the searches are not returning the proper results.
When I searched regarding this issue I found various answers to it. Can you
tell me which of the below listed issue is the root cause?
1. Can it be due to the folloing issue raised in solr?
https://issues.apache.org/jira/browse/SOLR-1781
2. Can this occur due to the slave having additional data (which is being
pushed from UI)?
3. Can this occur due to solr warming as described in the below forum post?
http://lucene.472066.n3.nabble.com/Solr-warming-when-using-master-slave-replication-td3293838.html
4. Can this be due to the segments.gen file not being copied properly as
given in the below post?
http://lucene.472066.n3.nabble.com/Lucene-FieldCache-memory-requirements-td484731i20.html
Can you please guide me to the root of this problem?
Thanks
--
View this message in context:
http://lucene.472066.n3.nabble.com/Multiple-Versions-getting-formed-while-replicating-in-solr-1-4-1-tp4003986.html
Sent from the Solr - User mailing list archive at Nabble.com.
