I'm having trouble finding any documentation on how to configure
ElasticSearch clustering. In Guide/API, there are some pages on how
to query the cluster state and health, but I can't find anything on
how to assign nodes to a cluster. Where are those docs?
On Sat, 2011-04-16 at 10:41 -0700, tsuraan wrote:
I'm having trouble finding any documentation on how to configure
Elasticsearch clustering. In Guide/API, there are some pages on how
to query the cluster state and health, but I can't find anything on
how to assign nodes to a cluster. Where are those docs?
Typically, you just give each node the same cluster name and start them
up. They discover each other using multicast.
For more, see
- Elasticsearch Platform — Find real-time answers at scale | Elastic
- Elasticsearch Platform — Find real-time answers at scale | Elastic
or for EC2
clint
Typically, you just give each node the same cluster name and start them
up. They discover each other using multicast.For more, see
-Elasticsearch Platform — Find real-time answers at scale | Elastic
-Elasticsearch Platform — Find real-time answers at scale | Elasticor for EC2
-Elasticsearch Platform — Find real-time answers at scale | Elastic
Ok, cool. So without a multicast network in place, I can just have a
config file entry like this:
{ 'discovery' :
{ 'zen' :
{ 'ping' :
{ 'unicast' :
{ 'hosts' : ['host1:port1', 'host2:port2', ... ] }}}}}
Is the port here just the normal Elasticsearch client port (9200 in my
case)? In the multicast docs, it says the port is typically 54328,
but I'm guessing maybe multicast works completely differently.
The port is the internal transport port: 9300. You don't have to set it explicitly, it will automatically use the default port chosen for hte transport, so you can just go with the host.
On Saturday, April 16, 2011 at 9:09 PM, tsuraan wrote:
Typically, you just give each node the same cluster name and start them
up. They discover each other using multicast.
For more, see
-Elasticsearch Platform — Find real-time answers at scale | Elastic
-Elasticsearch Platform — Find real-time answers at scale | Elasticor for EC2
-Elasticsearch Platform — Find real-time answers at scale | ElasticOk, cool. So without a multicast network in place, I can just have a
config file entry like this:{ 'discovery' :
{ 'zen' :
{ 'ping' :
{ 'unicast' :
{ 'hosts' : ['host1:port1', 'host2:port2', ... ] }}}}}Is the port here just the normal Elasticsearch client port (9200 in my
case)? In the multicast docs, it says the port is typically 54328,
but I'm guessing maybe multicast works completely differently.