We need to clearly distinguish between losing a _shard_ and losing a _replica_. You have RF=2 so you have two replicas in each shard.
If you stop a Solr node hosting one of the two replicas you'll see that the leader function will switch to the running replica. Now if you nuke the data directory then restart the node, then the index will be replicated from the leader and no data is lost. It's all automatic, just sit back and watch. If you nuke the data directory for _all_ replicas that make up a particular shard (2 in this case), there's nothing you can do to get it back except re-index. Best, Erick On Fri, May 19, 2017 at 4:51 AM, Sudershan Madhavan < smadha...@digitalriver.com> wrote: > Hi, > > Good day. I have a solrcloud cluster which has a collection with RF=2 and > NumShards=3 on 6 Nodes. We want to test how to recover from unexpected > situations like shard loss. So we will probably execute an *rm -rf *on > the solr data directory on one of the replica or master. Now the question > is, how will this shredded node recover from the shard loss? Are manual > steps required(if yes, then what needs to be done), or will it > automatically recover from the replica? > > > > > > > > > > Regards, > > > > SUDERSHAN MADHAVAN *|* Digital River World Payments *|* Senior Database > Administrator > m: +46 733 312525 *|* smadha...@digitalriver.com *|* digitalriver.com > <http://www.digitalriver.com/> > > Managing Director: Michael Roos *|* Registered Office: Stockholm *|* > Company Number: 556548-0026 > > Textilgatan 31, SE-120 30 Stockholm, Sweden > > *[image: In-2C-14px]* <https://www.linkedin.com/company-beta/2991085/> > *Follow > us on* LinkedIn <https://www.linkedin.com/company-beta/2991085/> > > >
