I really don't understand this. It seems that not doing a full import and just re-submitting the changed documents is something you should consider.
Or just don't commit. Or consider using two cores, the idea here is that you have your "live" core that serves requests, and indexing to your new core, then swapping the cores after the indexing is complete. Best Erick On Tue, Jan 31, 2012 at 11:54 AM, dprasadx <dpras...@gmail.com> wrote: > I need to index our data 10 times a day due to frequent data changes. > We have placed a mechanism where the data entered by the user in the front > end is submitted into the solr index directly through solrj server. So, if > the solrj commit occurs during the middle of indexing, I lose all the > records in the index and the index contains only the documents during that > point of commit. Once the indexing is completed, I get back all the records. > > -- > View this message in context: > http://lucene.472066.n3.nabble.com/Solrj-commit-affecting-documents-count-tp3703646p3703972.html > Sent from the Solr - User mailing list archive at Nabble.com.
