Configurate elasticsearch under cluster with 4 machines and NLB (windows server)

Elastic Stack Elasticsearch
1

Hello everybody!

I'm trying to figure out which are the parameters to set in the
elasticsearch.yml to get working the re-index os objects between 4 machines
that are in the same network. This 4 machines are configured with NLB
responding by the ip 123.123.123.5 (virtual address that make the load
balancing) and each machine with 123.123.123.123.1/2/3/4. Just setting the
same cluster.name and differente node.name they didn't see each other:

http://localhost:9200/_cluster/nodes?pretty
{
"ok" : true,
"cluster_name" : "CLUSTER_NAME",
"nodes" : {
"w6bUFO79S9GHQ_AL20s29Q" : {
"name" : "MACHINE_3",
"transport_address" : "inet[/123.123.123.5:9300]",
"hostname" : "RCJ3",
"version" : "0.90.0",
"http_address" : "inet[/123.123.123.5:9200]"
}
}
}

http://localhost:9200/_cluster/nodes/health?pretty
{
"ok" : true,
"cluster_name" : "CLUSTER_NAME",
"nodes" : { }
}

After I' tried to change the network.publish_host of each machine to your
own ip and still not working:

http://localhost:9200/_cluster/nodes?pretty
{
"ok" : true,
"cluster_name" : "RCJHD_SEARCH_ENGINE",
"nodes" : {
"g7B9xw9yTeW-28YrpIozZw" : {
"name" : "RCJ3",
"transport_address" : "inet[/123.123.123.3:9300]",
"hostname" : "RCJ3",
"version" : "0.90.0",
"http_address" : "inet[/123.123.123.3:9200]"
}
}
}

Any ideias?

Thanks!

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

2

How have you configured NLB?

I have the same setting. I have set up 3 machines. Each machine has got 2
NIC, one specific for the NLB cluster which points to the cluster virtual
ip, and a regular NIC that has got ip for the machine in question.

It all works very well, and it it just a matter of adding another host to
the cluster and starting ES on that host to make them find eachother.

ES has zen discovery, which if the nodes are on the same network the will
find the other nodes (with the same cluster name). Do you get anything in
the log? Do the second node get any warning, or something upon start?

//Tommy

Den fredagen den 14:e juni 2013 kl. 15:17:11 UTC+2 skrev Cassiano Tartari:

Hello everybody!

I'm trying to figure out which are the parameters to set in the
elasticsearch.yml to get working the re-index os objects between 4
machines that are in the same network. This 4 machines are configured with
NLB responding by the ip 123.123.123.5 (virtual address that make the
load balancing) and each machine with 123.123.123.123.1/2/3/4. Just
setting the same cluster.name and differente node.name they didn't see
each other:

http://localhost:9200/_cluster/nodes?pretty
{
"ok" : true,
"cluster_name" : "CLUSTER_NAME",
"nodes" : {
"w6bUFO79S9GHQ_AL20s29Q" : {
"name" : "MACHINE_3",
"transport_address" : "inet[/123.123.123.5:9300]",
"hostname" : "RCJ3",
"version" : "0.90.0",
"http_address" : "inet[/123.123.123.5:9200]"
}
}
}

http://localhost:9200/_cluster/nodes/health?pretty
{
"ok" : true,
"cluster_name" : "CLUSTER_NAME",
"nodes" : { }
}

After I' tried to change the network.publish_host of each machine to your
own ip and still not working:

http://localhost:9200/_cluster/nodes?pretty
{
"ok" : true,
"cluster_name" : "RCJHD_SEARCH_ENGINE",
"nodes" : {
"g7B9xw9yTeW-28YrpIozZw" : {
"name" : "RCJ3",
"transport_address" : "inet[/123.123.123.3:9300]",
"hostname" : "RCJ3",
"version" : "0.90.0",
"http_address" : "inet[/123.123.123.3:9200]"
}
}
}

Any ideias?

Thanks!

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.