Kindly help me to understand the behaviour of ES nodes in Cluster when
nodes have different Index mappings.
I have 2 ES nodes both are currently having same Index versions. Now I want
to upgrade both the nodes with the new index mapping.
Scenario 1 : Without keeping the node down, start mapping changes on Node1.
During the mapping changes, if any request comes & suppose it is handled by
Node1 only.
Now, How & when this Node#1 & Node#2 will synchronise?
Scenario 2 : Without keeping the node down, start mapping changes on Node1.
During the mapping changes, if any request comes & suppose it is handled by
Node2.
Now, How & when this Node#1 & Node#2 will synchronise?
Want to know, whose data will be available after the complete mappings
completion on both nodes.
Mapping is applied at cluster level, and existing index wont get the new mapping. You will need to reindex your data, aka create a new index after you apply the new mapping
Yes, I am creating new index and then migrating the data from older index
to new index. So, when this migration is going on, if any request comes,
then what would be the behaviour?
On Friday, June 13, 2014 3:11:52 PM UTC+5:30, Luis García Acosta wrote:
Mapping is applied at cluster level, and existing index wont get the new
mapping. You will need to reindex your data, aka create a new index after
you apply the new mapping
Yes, I am creating new index and then migrating the data from older index
to new index. So, when this migration is going on, if any request comes,
then what would be the behaviour?
On Friday, June 13, 2014 3:11:52 PM UTC+5:30, Luis García Acosta wrote:
Mapping is applied at cluster level, and existing index wont get the new
mapping. You will need to reindex your data, aka create a new index after
you apply the new mapping
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.