An actor for load-balancing requests between Solr servers.
An actor for load-balancing requests between Solr servers. Based directly on org.apache.solr.client.solrj.impl.LBHttpSolrServer
so its notes on indexing hold for this implementation. Isn't strictly round robin (and neither is the original, due
to error handling and retries), but tries servers in random order for every request (so roughly equal distribution
over time). The random query sequences always try "zombie" servers after all "live" servers have been tried. Doesn't
provide ability to add/remove servers after creation, but could easily be added. The capability to query all servers
in parallel and return the first successful response isn't currently implemented, but could easily be added.
An actor for load-balancing requests between Solr servers. Based directly on org.apache.solr.client.solrj.impl.LBHttpSolrServer so its notes on indexing hold for this implementation. Isn't strictly round robin (and neither is the original, due to error handling and retries), but tries servers in random order for every request (so roughly equal distribution over time). The random query sequences always try "zombie" servers after all "live" servers have been tried. Doesn't provide ability to add/remove servers after creation, but could easily be added. The capability to query all servers in parallel and return the first successful response isn't currently implemented, but could easily be added.
org.apache.solr.client.solrj.impl.LBHttpSolrServer