Dear Community,
I want to put in place a kind of redundant loadbalancing setup with 2 ElasticSearch nodes + 2 Kibana Nodes with Elasticsearch Coordinated Only node role.
Here a picture of the setup
I will give you my configuration for each nodes :
elastic-01:
cluster.name: mycluster
node.name: elastic-01
network.host: 10.10.68.24
http.port: 9200
discovery.zen.ping.unicast.hosts: ["elastic-01", "elastic-02", "kibana-01", "kibana-02"]
discovery.zen.master_election.ignore_non_master_pings: true
elastic-02:
cluster.name: mycluster
node.name: elastic-02
network.host: 10.10.68.25
http.port: 9200
discovery.zen.ping.unicast.hosts: ["elastic-01", "elastic-02", "kibana-01", "kibana-02"]
discovery.zen.master_election.ignore_non_master_pings: true
kibana-01:
cluster.name: mycluster
node.master: false
node.data: false
node.ingest: false
search.remote.connect: false
network.host: localhost
http.port: 9200
transport.host: 10.10.68.25
transport.tcp.port: 9300
kibana-02:
cluster.name: mycluster
node.master: false
node.data: false
node.ingest: false
search.remote.connect: false
network.host: localhost
http.port: 9200
transport.host: 10.10.68.26
transport.tcp.port: 9300
It seems that when I browse from http://kibana-01:5601/ i have the following error :
plugin:elasticsearch@6.2.3 Service Unavailable
The same from the http://kibana-02:5601.
The elasticsearch logs from the Coordinated Only Nodes :
[2018-04-20T13:36:02,145][WARN ][o.e.d.z.ZenDiscovery ] [zxSDt-c] not enough master nodes discovered during pinging (found [[]], but needed [-1]), pinging again
It seems that the coordinated only nodes are not seing the other elasticsearch nodes..
Maybe you have an idea on what is wrong with my setup.
Best Regards, Edouard.