We are using ES 1.3.2.
We have a need to specify custom id and routing values when indexing.
We've been doing this using Java APIs, however we would now like to update
the mapping to specify 'path's for _id and _routing.
The question we have is:
Since this type already has a huge number of documents, can we change
the mapping? When we tried it, we got a '"acknowledged": true' response,
but it doesn't seem to be working when we tried indexing.
In case there is a way to achieve this, will it affect only the new
documents being indexed?
On Monday, October 27, 2014 3:03:16 PM UTC+5:30, Preeti Raj - Buchhada
wrote:
We are using ES 1.3.2.
We have a need to specify custom id and routing values when indexing.
We've been doing this using Java APIs, however we would now like to update
the mapping to specify 'path's for _id and _routing.
The question we have is:
Since this type already has a huge number of documents, can we change
the mapping? When we tried it, we got a '"acknowledged": true' response,
but it doesn't seem to be working when we tried indexing.
In case there is a way to achieve this, will it affect only the new
documents being indexed?
You can update a mapping (adding new fields or new sub fields)
Old documents won't be updated.
A common best practice is to reindex your data.
About specific fields like _routing and _id, I think that updating the mapping for them won't work so it will require to reindex.
Never tested it though.
Le 28 octobre 2014 à 08:13:04, Preeti Raj - Buchhada (pbuchhada@gmail.com) a écrit:
Any ideas on this?
On Monday, October 27, 2014 3:03:16 PM UTC+5:30, Preeti Raj - Buchhada wrote:
We are using ES 1.3.2.
We have a need to specify custom id and routing values when indexing.
We've been doing this using Java APIs, however we would now like to update the mapping to specify 'path's for _id and _routing.
The question we have is:
Since this type already has a huge number of documents, can we change the mapping? When we tried it, we got a '"acknowledged": true' response, but it doesn't seem to be working when we tried indexing.
In case there is a way to achieve this, will it affect only the new documents being indexed?
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.