See particularly the ADDREPLICA command and the "node" parameter. You might not even need the "node" parameter since when you add a replica Solr does its best to put the new replica on an underutilized node.
Best, Erick On Thu, Jun 18, 2015 at 2:58 PM, Shawn Heisey <apa...@elyograg.org> wrote: > On 6/18/2015 3:23 PM, Jim.Musil wrote: >> Let's say I have a zookeeper ensemble with several Solr nodes connected to >> it. I've created a collection successfully and all is well. >> >> What happens when I want to add another solr node? >> >> I've tried spinning one up and connecting it to zookeeper, but the new node >> doesn't "join" the collection. What's the expected next step? >> >> This is Solr 5.1. > > The new node will be part of the cloud as soon as it starts, but until > you take action with the Collections API, it will not have any indexes > on it. SolrCloud does not automatically create replicas except in a > very specific set of circumstances that I do not think are very common. > > You'll need to either create a new collection or take steps to modify > your current collection(s) so that one or more shard replicas are > located on the new node. > > https://cwiki.apache.org/confluence/display/solr/Collections+API > > Thanks, > Shawn >
