You received this message because you are subscribed to the Google Groups "elasticsearch" group.
For more options, visit https://groups.google.com/groups/opt_out.
Dibyendu, please consider making an abstraction for the indexer piece, so people can plug other indexing engines, such as Solr(Cloud) in there. Oh, I guess this assumes you will open-source this, but I'm not sure if you are planning on doing that or not?
I have the same question as dancer. How could I deal with failures in inserting data to ES or HBase. Retry or RollBack? As we all know, RollBack in HBase or ES dose not easily especially that will happen in concurrent environment.
Re: I am speaking @HBase Conference 2013 on "Using HBase Coprocessors to Index Columns in an Elasticsearch Cluster"
We do Elasticsearch consulting all the time (see sig) and some of of clients have multi-billion document indices with high index and/or query rates.
It's stable and works well most of the time. Like anything else, it has its quirks and requires some learning. We've seen our ES customers having issues with split brain and there are a couple of oldish issues open where others are reporting seeing the same. I imagine/hope ES is working on fixing this.