Hi, You can have multiple nodes as long as you make sure that your collection has only one shard, then the joins will work.
-- Jan Høydahl, search solution architect Cominvent AS - www.cominvent.com > 30. aug. 2018 kl. 19:51 skrev Steve Pruitt <bpru...@opentext.com>: > > Shawn, > > You are correct. I created another setup. This time with 1 node, 1 shard, 2 > replicas and the join worked! > Running with the example SolrCloud setup doesn't work for join queries. > > Thanks. > > -S > > > -----Original Message----- > From: Steve Pruitt <bpru...@opentext.com> > Sent: Thursday, August 30, 2018 12:25 PM > To: solr-user@lucene.apache.org > Subject: RE: [EXTERNAL] - Re: join works with a core, doesn't work with a > collection > > Gosh, really? This is not mentioned anywhere in the documentation that I can > find. There are node to HW considerations if you are joining across > different Collections. > But, the same Collection? Tell me this is not so. > > -S > > -----Original Message----- > From: Shawn Heisey <apa...@elyograg.org> > Sent: Thursday, August 30, 2018 12:11 PM > To: solr-user@lucene.apache.org > Subject: Re: [EXTERNAL] - Re: join works with a core, doesn't work with a > collection > > On 8/30/2018 9:49 AM, Steve Pruitt wrote: >> If you mean another running Solr server running, then no. > > I mean multiple Solr processes. > > The cloud example (started with bin/solr -e cloud) starts two Solr instances > if you give it the defaults. They are both running on the same machine, but > if part of the data is on the instance running on port > 8983 and part of the data is on the instance running on port 7574, I don't > think you can do a join. > > Thanks, > Shawn >
