Then you'll have to re-index probably after you set up your new collections.
If you have stored _all_ your original fields you could query from your 4x and index to your SolrCloud, but it'd be best if you could just reindex from the original source. Best, Erick On Tue, Dec 12, 2017 at 10:02 PM, Amin Raeiszadeh <amin24march1...@gmail.com> wrote: > thanks for your guides Erick > before this i use solr v4.10.3 in a cluster of 40 searcher (solr instance) > each with about 40 cores. > now i want to migrate to solr cloud v6.5.1 i design my own custom policy > for document routing(default mode of routing) between shards an i have > about 1000 shards at start time. > i think by your way it is not possible because of routing method for update > documents in future solr could not find old document to update it with new > document. > > On Tue, Dec 12, 2017 at 7:24 PM, Erick Erickson <erickerick...@gmail.com> > wrote: > >> What you haven't told us is whether you need to shard or not. But >> there's no difference between the index built for stand-alone Solr and >> SolrCloud. >> >> So just create your SolrCloud instance, probably single shard, 1 >> replica with the appropriate configset. Then shut that down and copy >> the index from your stand-alone Solr to the replica's index directory. >> Then start Solr back up. You should be fine. >> >> Now do whatever you need to do to build out your SorlCloud instance. >> If you need more shards use SPLITSHARD. If you need more replicas use >> ADDREPLICA etc. >> >> Best, >> Erick >> >> On Tue, Dec 12, 2017 at 2:37 AM, Amin Raeiszadeh >> <amin24march1...@gmail.com> wrote: >> > i have a lucene index that some fields of docs are indexed with custom >> > incremental gaps and all fields are stored too(not only indexed). >> > i need to import this docs to solr cloud. >> > is there any way to automatically rebuild this docs for importing in >> solr >> > with costum gaps by some thing likes converter. >>
