I’m preparing a new Graylog scenario composed of 2 different sites with good network connectivity (1x Graylog server, 1x ElasticSearch hot node and 1x ElasticSearch wam node per site and in the same cluster).
Assuming all the sources from site 1 point to the Graylog server in the same site. Is it possible to prioritize the traffic from Graylog to ES to be sent to the ElasticSearch hot node in the site 1? (the same would apply for the site 2)
If elasticsearch_replicas is set to 1, I guess the replication traffic will happen only between the data nodes, isn’t it?
Thanks in advance,