From the ES documentation here https://www.elastic.co/guide/en/elasticsearch/reference/current/setup-upgrade.html says we require a full cluster restart for an upgrade.
Is there a way to upgrade a cluster without downtime i.e is there a way of a rolling upgrade?
Nope.
But ou can create a fresh new 5.0 cluster and then use reindex from remote to read from 2.x indices.
Of course you will need to stop index operations at some point or to send them to the new cluster instead but it depends on your constraints I guess.
How does a Production cluster with a 50 -100 node deployment run a upgrade like this?
It means you have to run an equivalent 50 - 100 nodes 5.0 cluster in parallel.
May be not possible. In which case you have to stop the service, upgrade, then start again. So with downtime.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.