Question

Currently I have a zookeeper instance controlling replication on 3 physical servers. It is the solr integrated zookeeper. 1 shard, 1 collection.

I have a new requirement in which I will need a new static solr instance (1 new collection, no replication). Same schema as previous collection. A copy of this instance will also be placed on the 3 physical servers mentioned above. A caveat is that I need to perform distributed searches across the 2 collections and have the results blended.

Thanks to javacreed I now know that sharding is not in my solution. Previous questions answers here and here.

In my current setup I run the following command on the server running zookeeper -

java -Dbootstrap_confdir=solr/myApp/conf -Dcollection.configName=myConfig -DzkRun -DnumShards=1 -jar start.jar

Am I correct in saying that this will not change and I will now also manually start the non replicated collection. I really only need to change my search queries to include the 'collection' parameter? Something like -

http://localhost:8983/solr/collection1/select?collection=collection1,collection2

This example is from Solr documentation. I am slightly confused as to whether it should be ...solr/collection1/select?... or ...solr/collection2/select?... or if it even matters?

Thanks

Was it helpful?

Solution

Thanks for your kind word stewart.You can search it directly on solr as

http://localhost:8983/solr/select?collection=collection1,collection2

There is no need to mention any collection path since you are defining them in the collection parameters.

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top