Sharing index files between multiple JVMs and replication
This post has NOT been accepted by the mailing list yet.
Is there a way to share index files amongst my multiple Solr web-apps, by configuring only one of the JVMs as an indexer, and the remaining, as read-only searchers?
I'd like to configure in such a way that on startup of the read-only searchers, missing cores/indexes are not created, and updates are not handled.
If I can get around the files being locked by the read-only instances, I should be able to scale wider in a given environment, as well as have less replicated copies of my master index (Solr 1.4 Java Replication).
Then once the commit is issued to the slave, I can fire off a RELOAD script for each of my read-only cores.