I had some questions on the Nodes in Elasticsearch Cluster -
Can i add extra nodes to an EXISTING cluster without restarting ANY OF
the nodes in the cluster?
If so, how can i do that.
Which settings & mapping of an indice, if i change it - i will need to
restart the elasticsearch cluster?
Is there any way by which i don't set the UNICAST field in the
elasticsearch.yml config file, so that the Nodes when they come up they
automatically detect the CLUSTER and join it.
SINCE I DON'T WANT TO SET the UNICAST discovery.zen.ping.unicast.hosts
field.
On Friday, September 20, 2013 2:45:36 PM UTC-4, pranav amin wrote:
Hi all,
I had some questions on the Nodes in Elasticsearch Cluster -
Can i add extra nodes to an EXISTING cluster without restarting ANY OF
the nodes in the cluster?
If so, how can i do that.
Which settings & mapping of an indice, if i change it - i will need to
restart the elasticsearch cluster?
Is there any way by which i don't set the UNICAST field in the
elasticsearch.yml config file, so that the Nodes when they come up they
automatically detect the CLUSTER and join it.
SINCE I DON'T WANT TO SET the UNICAST discovery.zen.ping.unicast.hosts
field.
Can i just designate A PARTICULAR NODE TO ONLY contain REPLICA DATA? So
it will not have ACTIVE SHARD DATA, BUT JUST REPLICA.
Can i FORCE my Reporting application to ALWAYS GO TO NODE CONTAINING
REPLICA DATA AND NOT TO ACTIVE SHARD? does it offer any performance
benefit if i do so?
Can i add extra nodes to an EXISTING cluster without restarting ANY OF the nodes in the cluster?
If so, how can i do that.
YES
bin/elasticseach will start a new node
Which settings & mapping of an indice, if i change it - i will need to restart the elasticsearch cluster?
You don't have to stop a cluster when adding/deleting/modifying an index or a mapping.
Some index settings can not be changed on an existing index, even with stopping the cluster.
Is there any way by which i don't set the UNICAST field in the elasticsearch.yml config file, so that the Nodes when they come up they automatically detect the CLUSTER and join it.
SINCE I DON'T WANT TO SET the UNICAST discovery.zen.ping.unicast.hosts field.
Use defaults: multicast.
Can i just designate A PARTICULAR NODE TO ONLY contain REPLICA DATA? So it will not have ACTIVE SHARD DATA, BUT JUST REPLICA.
Can i FORCE my Reporting application to ALWAYS GO TO NODE CONTAINING REPLICA DATA AND NOT TO ACTIVE SHARD? does it offer any performance benefit if i do so?
Why do you want to do that? I don't see any benefit.
You have to know that elasticsearch by default will act as a load balancer to distribute the load on shards (primaries and replicas).
For the 1st, if i bring up a node with the same cluster name would it be
detected automatically and join the cluster? are there any settings i need
to do.
I used to set the Unicast field in config file, but i was just limited to
the IP address of nodes i know. In case there is a New Node, i will need to
modify this field to make sure it now knows about the New node. Then i will
need to restart it.
Is there any other way, where automatically it is detected without any
config changes?
Thanks
On Friday, September 20, 2013 2:52:25 PM UTC-4, David Pilato wrote:
Can i add extra nodes to an EXISTING cluster without restarting ANY
OF the nodes in the cluster?
If so, how can i do that.
YES
bin/elasticseach will start a new node
Which settings & mapping of an indice, if i change it - i will need
to restart the elasticsearch cluster?
You don't have to stop a cluster when adding/deleting/modifying an index
or a mapping.
Some index settings can not be changed on an existing index, even with
stopping the cluster.
Is there any way by which i don't set the UNICAST field in the
elasticsearch.yml config file, so that the Nodes when they come up they
automatically detect the CLUSTER and join it.
SINCE I DON'T WANT TO SET the UNICAST discovery.zen.ping.unicast.hosts
field.
Use defaults: multicast.
For the 1st, if i bring up a node with the same cluster name would it be detected automatically and join the cluster? are there any settings i need to do.
I used to set the Unicast field in config file, but i was just limited to the IP address of nodes i know. In case there is a New Node, i will need to modify this field to make sure it now knows about the New node. Then i will need to restart it.
Is there any other way, where automatically it is detected without any config changes?
Thanks
On Friday, September 20, 2013 2:52:25 PM UTC-4, David Pilato wrote:
Can i add extra nodes to an EXISTING cluster without restarting ANY OF the nodes in the cluster?
If so, how can i do that.
YES
bin/elasticseach will start a new node
Which settings & mapping of an indice, if i change it - i will need to restart the elasticsearch cluster?
You don't have to stop a cluster when adding/deleting/modifying an index or a mapping.
Some index settings can not be changed on an existing index, even with stopping the cluster.
Is there any way by which i don't set the UNICAST field in the elasticsearch.yml config file, so that the Nodes when they come up they automatically detect the CLUSTER and join it.
SINCE I DON'T WANT TO SET the UNICAST discovery.zen.ping.unicast.hosts field.
Use defaults: multicast.
I said: use default settings. That means: don't disable multicast.
If you disable multicast, just add a node which have other nodes in the unicast node list.
It will "declare" itself to the cluster.
--
David
Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs
For the 1st, if i bring up a node with the same cluster name would it be detected automatically and join the cluster? are there any settings i need to do.
I used to set the Unicast field in config file, but i was just limited to the IP address of nodes i know. In case there is a New Node, i will need to modify this field to make sure it now knows about the New node. Then i will need to restart it.
Is there any other way, where automatically it is detected without any config changes?
Thanks
On Friday, September 20, 2013 2:52:25 PM UTC-4, David Pilato wrote:
Can i add extra nodes to an EXISTING cluster without restarting ANY OF the nodes in the cluster?
If so, how can i do that.
YES
bin/elasticseach will start a new node
Which settings & mapping of an indice, if i change it - i will need to restart the elasticsearch cluster?
You don't have to stop a cluster when adding/deleting/modifying an index or a mapping.
Some index settings can not be changed on an existing index, even with stopping the cluster.
Is there any way by which i don't set the UNICAST field in the elasticsearch.yml config file, so that the Nodes when they come up they automatically detect the CLUSTER and join it.
SINCE I DON'T WANT TO SET the UNICAST discovery.zen.ping.unicast.hosts field.
Use defaults: multicast.
On Friday, September 20, 2013 5:00:10 PM UTC-4, David Pilato wrote:
I said: use default settings. That means: don't disable multicast.
If you disable multicast, just add a node which have other nodes in the
unicast node list.
It will "declare" itself to the cluster.
--
David
Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs
Le 20 sept. 2013 à 21:49, pranav amin <parulp...@gmail.com <javascript:>>
a écrit :
thanks for the quick turn around.
For the 1st, if i bring up a node with the same cluster name would it be
detected automatically and join the cluster? are there any settings i need
to do.
I used to set the Unicast field in config file, but i was just limited to
the IP address of nodes i know. In case there is a New Node, i will need to
modify this field to make sure it now knows about the New node. Then i will
need to restart it.
Is there any other way, where automatically it is detected without any
config changes?
Thanks
On Friday, September 20, 2013 2:52:25 PM UTC-4, David Pilato wrote:
Can i add extra nodes to an EXISTING cluster without restarting ANY
OF the nodes in the cluster?
If so, how can i do that.
YES
bin/elasticseach will start a new node
Which settings & mapping of an indice, if i change it - i will need
to restart the elasticsearch cluster?
You don't have to stop a cluster when adding/deleting/modifying an index
or a mapping.
Some index settings can not be changed on an existing index, even with
stopping the cluster.
Is there any way by which i don't set the UNICAST field in the
elasticsearch.yml config file, so that the Nodes when they come up they
automatically detect the CLUSTER and join it.
SINCE I DON'T WANT TO SET the UNICAST discovery.zen.ping.unicast.hosts
field.
Use defaults: multicast.
HTH
David
--
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 elasticsearc...@googlegroups.com <javascript:>.
For more options, visit https://groups.google.com/groups/opt_out.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.
