My out-of-the-box Solr 4.3.1 does not recognize the trailing /replication at the end of masterUrl in the slave setting.
Here's the relevant part of my solrconfig.xml on the slave:
<requestHandler name="/replication" class="solr.ReplicationHandler">
<lst name="slave">
<str name="masterUrl">http://mymaster:myport/solr/mycorename/replication</str>
<str name="pollInterval">00:00:60</str>
</lst>
</requestHandler>
And here's the relevant part on the master:
<requestHandler name="/replication" class="solr.ReplicationHandler">
<lst name="master">
<str name="replicateAfter">commit</str>
<str name="replicateAfter">startup</str>
<str name="replicateAfter">optimization</str>
<str name="confFiles">schema.xml,stopwords.txt</str>
</lst>
</requestHandler>
And this is the error I'm getting on the slave, every minute (like specified by pollInterval
):
Master at: http://mymaster:myport/solr/collection1 is not available. Index fetch failed. Exception: Host name may not be null
http://wiki.apache.org/solr/SolrReplication says to hit the masterUrl with a browser to make sure it's right; a response status OK is a must. When I hit mymaster:myport/solr/mycorename/replication with a browser, I get the response status OK. When I hit mymaster:myport/solr/mycorename, I get a 404.
An extensive Google search didn't yield anything definitive on how to set a master URL and have Solr not trim the word 'replication'.
Is there something I'm missing in my config(s) that makes Solr strip the trailing /replication
from masterUrl? Is there a way I can configure the master to have a replication URL that doesn't require trailing /replication
?
Thanks