How to bind Elasticsearch 2.0 on both Loopback and Non-Loopback interfaces?

Question

Starting from version 2.0 Elasticsearch binds only on the loopback interface by default (_local_ in terms of configuration).

The documentation says that there is a way to switch to another network, for example, _non_loopback_ binds to the first non-loopback interface. It works fine.

But I cannot figure out how do I combine these settings so that Elasticsearch binds on both loopback and non-loopback interfaces simultaneously?

PS. My reason is that I use Logstash on each Elasticsearch instance that connects to it via localhost, but I also want other Elasticsearch instances to see each other to form the cluster...

This might be doable using [transport profiles](https://www.elastic.co/guide/en/elasticsearch/reference/current/modules-transport.html#_tcp_transport_profiles) but I haven't tried yet. — Val, Oct 29 '15 at 11:13
@Val, thank you for the suggestion, but according to the reference transport profiles are only for node-to-node communication... Of course, I can still use non-localhost IP address for Logstash configuration, but it would make configuration maintenance much worse than it is used to be in 1.x. — Funbit, Oct 30 '15 at 03:25
@AndreiStefan Using this: hosts => ["127.0.0.1:9200"] . As I said, I could include ethernet IP here, but I would have to do that for each instance separately.. since Logstash does not support environment/dynamic variables in its configuration. — Funbit, Oct 30 '15 at 09:24
@Funbit what protocol are you using with LS? Assuming `http` (since port 9200) — Andrei Stefan, Oct 30 '15 at 09:32
@AndreiStefan Not sure how it is related to the problem, but: I'm using "tcp" input, and "elasticsearch" output. LS 2.0 (which I'm trying to use with ES 2.0) supports only "http" protocol. — Funbit, Oct 30 '15 at 09:55
Ok. Then try in your ES something like the following. Remove any `network.bind*` and `network.publish*` settings from your `elasticsearch.yml` file and add `transport.host: _non_loopback_` and `http.host: _local_`. — Andrei Stefan, Oct 30 '15 at 09:59
@AndreiStefan This way I will lose ability to connect to ES instances from Kibana or Browsers... I dont want to install reverse proxy just for that.. — Funbit, Oct 30 '15 at 10:55
Try another config and see how it goes: `network.bind_host: 0` — Andrei Stefan, Oct 30 '15 at 14:01
@AndreiStefan The magic worked! Thank you! Please post an answer I'll mark it as the right one. — Funbit, Oct 30 '15 at 14:28

score 40 · Accepted Answer · answered Oct 30 '15 at 14:43

40

For 2.0 you would need to use

network.bind_host: 0

answered Oct 30 '15 at 14:43

Andrei Stefan

51,654
6
98
89

This solved my question too: http://stackoverflow.com/questions/33691858/elasticsearch-nonodeavailableexception – Carlos Vega Nov 23 '15 at 16:18
What would be the solution for 2.3.2 ? I did the same as mentioned above but still the problem persists. – som Mar 20 '17 at 17:00
Worked for ES5! – kgf3JfUtW Apr 06 '17 at 18:40
Per https://stackoverflow.com/a/31677628/154461 I think you must use network.host instead of network.bind_host for version >= 2.3 – harschware Jul 12 '17 at 20:30

score 8 · Answer 2 · answered May 21 '19 at 12:18

As of ElasticSearch 7.x, this configuration has changed yet again. for a simple single node cluster bound to loopback, local and external IPs, you essentially do this:

network.host: [_local_, _site_, _global_]
cluster.initial_master_nodes: node-1

The cluster node setting is explained here while the network host setting is in the documentation here, although it doesn't say how you would assign multiple values to network.host.

score 2 · Answer 3 · answered Aug 29 '16 at 07:01

2

Go to

'<path_to_elasticsearch>/elasticsearch-2.3.4/config'

Open elasticsearch.yml

Add network.host: 0.0.0.0

Now check which port elasticsearch is using (9200 is default), go to firewall inbound rules and add those ports.

answered Aug 29 '16 at 07:01

ketankk

2,578
1
29
27

How to bind Elasticsearch 2.0 on both Loopback and Non-Loopback interfaces?

3 Answers3

Linked