You could always write code that executes a scan/scroll query from one
cluster and bulk loads them into the other. Perhaps one of the existing
reindex plugins can help:
--
Ivan
On Mon, Jul 15, 2013 at 7:13 PM, Raymond Wang rmn190@gmail.com wrote:
I have two ES instance deployed, and the two instance have some data in
common. I wanna to merge the two database into one.
You can start up remote nodes and connect them (directly or via IP
tunnels) to form a cluster so the nodes span data centers. With
awareness, you can configure ES to take the node distribution into
account, for shard allocation.
But heck no, there is no elasticsearch river. A river is primarily meant
being a one-way mechanism to fetch data from foreign sources. I think I
can plug code from my JDBC river and knapsack plugin together to achieve
this in a very naive style.
For ES 1.0, there were some feature announcements. One is the full
fledged backup/restore feature (maybe even on hot clusters) from which I
hope this feature can span clusters/data centers. No details yet, but
I'm excited. Maybe also with the feature of keeping two ES clusters in
sync on a slow connection.
Jörg
Am 16.07.13 09:42, schrieb Raymond Wang:
there are rivers for mysql,mongodb, so is there a river for
elasticsearch it self?
On Tuesday, July 16, 2013 10:13:49 AM UTC+8, Raymond Wang wrote:
I have two ES instance deployed, and the two instance have some
data in common. I wanna to merge the two database into one.
So, is there a tool that can help me?
any help would be appreciated.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.