We have a elastic cluster with n node setup. In our setup, restart of the node will give a new ip address to the node. Every time we restart a node and it joins back the cluster the server is able to recover easily. The issue we are having is with the client. The client is not able to recover from this situation and the Queries hang until we restart our application/services and make it connect again. Does the client cache the IP address of the nodes? We are using the transport client now and moving towards using the high level rest client. Will the high level rest client be able to recover from this situation? Is there any configuration to override this behavior?
I believe that the Rest Client will manage that situation better.
Thanks for your response. Will there be any difference between the low-level rest client and high-level rest client in this regard particularly?
No. That's managed by the low level with the optional Sniffer: https://www.elastic.co/guide/en/elasticsearch/client/java-rest/current/_usage.html
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.