On Wed, Dec 10, 2008 at 11:00 PM, Rakesh Sinha <[EMAIL PROTECTED]> wrote: > Hi - > I am a new user of Solr tool and came across the introductory > tutorial here - http://lucene.apache.org/solr/tutorial.html . > I am planning to use Solr in one of my projects . I see that the > tutorial mentions about a REST api / interface to add documents and to > query the same. > > I would like to create the indices locally , where the web server (or > pool of servers ) will have access to the database directly , but use > the query REST api to query for the results.
If your data resides in DB consider using DIH. http://wiki.apache.org/solr/DataImportHandler > > I am curious how this could be possible without taking the http rest > api submission to add to indices. (For the sake of simplicity - we can > assume it would be just one node to store the index but multiple > readers / query machines that could potentially connect to the solr > web service and retrieve the query results. Also the index might be > locally present in the same machine as that of the Solr host or at > least accessible through NFS etc. ) I guess you are thinking of using a master/slave setup. see this http://wiki.apache.org/solr/CollectionDistribution or http://wiki.apache.org/solr/SolrReplication > > Thanks for helping out to some starting pointers regarding the same. > -- --Noble Paul
