Elasticsearch graceful restart without rebalancing shards

Hello, I wanted to know if it is possible to restart a node of an Elasticsearch cluster in a graceful way?
When I restart a node in the cluster, searches that use that node fail (shards are replicated).
Is it possible to blacklist a node for searches before restarting it?
Currently the only solution I have found: re-allocate the shards of the node to other nodes. But I can't do this because my nodes have several terabytes of storage.

Thanks in advance, have a nice day.

You can basically follow the same procedure as explained at:

Just don't do the upgrade :smirk:

1 Like

Hello David and thank you for your response. I tried to do the trick (without upgrade :innocent:) but it seams that you need to do the request on a coordinatating node to do a graceful restart on the other nodes. Am i right ?

You need to do this procedure on all nodes.

1 Like

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.