If I have an elasticsearch installation at, say, version 0.90.3, and I want
to later upgrade to 0.90.2 or 0.91 or even 1.0, can that be done one node
at a time? So I can keep the cluster available during the upgrade?
major version upgrades (as from 0.90 to 1.0) require a cluster restart.
Minor versions can be done in a rolling upgrade, one node at a time.
You might want to disable shard allocation during that. See
If I have an elasticsearch installation at, say, version 0.90.3, and I
want to later upgrade to 0.90.2 or 0.91 or even 1.0, can that be done one
node at a time? So I can keep the cluster available during the upgrade?
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.