The consensus here seems to be NOT to do that. Have 2 SolrClouds, one per DC and feed them the same data. That way each cloud has its own quorum of ZKs, and as long as both clouds are up, they should be in sync. You have true DR in that each site is separate and not dependent on any data from the other.
You will need to handle the case when 1 DC goes down, and then comes back up, you need to re-sync the data it has "missed", but presumably whatever is sending to the cloud can keep track of whether both clouds accepted the data or not with some kind of checkpointing. On 8 February 2013 12:45, mizayah <miza...@gmail.com> wrote: > I dont think its so simple. > > First I need to have at last 3 zoo to keep failver for one server. > Second after one zoo die, i need restart of all solrs. > > Maybe i define simply question. > Two data centers. > How to replicate two solr claster between two datacenters? > In no SolrClaud there is repeater, if i connect all SolrCloud nodes in one > claster between dc i will make lots of trafiick between them. > I dont mention about that i will get leader elected in wrong datacanter > eventually. > > How can i have two claster of solr and replicate them between two > datacenters? > > > > -- > View this message in context: > http://lucene.472066.n3.nabble.com/SolrCloud-new-zookeper-node-on-different-ip-replicate-between-two-clasters-tp4039101p4039217.html > Sent from the Solr - User mailing list archive at Nabble.com. >
