# Elasticsearch cluster **URL:** \ **Category:** Elasticsearch\ **Created:** [January 24, 2013, 8:14pm UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499 "2013-01-24T20:14:41Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![Jimi\_Parekh](https://avatars.discourse-cdn.com/v4/letter/j/7feea3/32.png) [@Jimi\_Parekh](https://discuss.elastic.co/u/Jimi_Parekh)\ **Post date:** [January 24, 2013, 8:14pm UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499/1 "2013-01-24T20:14:41Z")
To create multiple nodes in elasticsearh cluster, do we need to start the multiple elasticsearch instance? I have a one elasticsearch storage on one server. How can I create cluster with muliple nodes which reads the same storage? -Jimi. -- ---
**Author:** ![dadoonet](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/dadoonet/32/137187_2.png) [@dadoonet](https://discuss.elastic.co/u/dadoonet)\ **Post date:** [January 24, 2013, 8:37pm UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499/2 "2013-01-24T20:37:52Z")
Each node uses its own storage. But you can use SharedFS: [http://www.elasticsearch.org/guide/reference/modules/gateway/fs.html](http://www.elasticsearch.org/guide/reference/modules/gateway/fs.html) But, why do you want to do that? -- David 😉 Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs Le 24 janv. 2013 à 21:14, Jimi Parekh [jimiparekh@gmail.com](mailto:jimiparekh@gmail.com) a écrit : To create multiple nodes in elasticsearh cluster, do we need to start the multiple elasticsearch instance? I have a one elasticsearch storage on one server. How can I create cluster with muliple nodes which reads the same storage? ## -Jimi. -- ---
**Author:** ![drewr](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/drewr/32/7803_2.png) [@drewr](https://discuss.elastic.co/u/drewr)\ **Post date:** [January 24, 2013, 8:41pm UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499/3 "2013-01-24T20:41:17Z")
Jimi Parekh wrote: > To create multiple nodes in elasticsearh cluster, do we need to > start the multiple elasticsearch instance? I have a one > elasticsearch storage on one server. How can I create cluster with > muliple nodes which reads the same storage? Start each node with the path.data parameter. A simple example would be: bin/elasticsearch -Des.path.data=/data/es -f You can also set it in elasticsearch.yml config file. path.data: /data/es -Drew -- ---
**Author:** ![Jimi\_Parekh](https://avatars.discourse-cdn.com/v4/letter/j/7feea3/32.png) [@Jimi\_Parekh](https://discuss.elastic.co/u/Jimi_Parekh)\ **Post date:** [January 24, 2013, 8:56pm UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499/4 "2013-01-24T20:56:13Z")
I have one elasticsearch index stored on server-1. Now I want to make sure if server-1 is down, still users can query the index and get the result. I am not sure whats the best possible way to accomplish it. I thought by adding new nodes i can do it. On Thursday, January 24, 2013 3:37:52 PM UTC-5, David Pilato wrote: > Each node uses its own storage. > But you can use SharedFS: > [Elasticsearch Platform — Find real-time answers at scale | Elastic](http://www.elasticsearch.org/guide/reference/modules/gateway/fs.html) > > But, why do you want to do that? > > -- > David 😉 > Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs > > Le 24 janv. 2013 à 21:14, Jimi Parekh \<[jimip...@gmail.com](mailto:jimip...@gmail.com) \\> > a écrit : > > To create multiple nodes in elasticsearh cluster, do we need to start the > multiple elasticsearch instance? > I have a one elasticsearch storage on one server. How can I create cluster > with muliple nodes which reads the same storage? > > -Jimi. > > -- -- ---
**Author:** ![dadoonet](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/dadoonet/32/137187_2.png) [@dadoonet](https://discuss.elastic.co/u/dadoonet)\ **Post date:** [January 24, 2013, 9:02pm UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499/5 "2013-01-24T21:02:36Z")
So just launch a new node and let Elasticsearch manage that for you. All data are replicated on nodes by Elasticsearch so you don't have to share anything here. HTH -- David 😉 Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs Le 24 janv. 2013 à 21:56, Jimi Parekh [jimiparekh@gmail.com](mailto:jimiparekh@gmail.com) a écrit : > I have one elasticsearch index stored on server-1. Now I want to make sure if server-1 is down, still users can query the index and get the result. > I am not sure whats the best possible way to accomplish it. I thought by adding new nodes i can do it. > > On Thursday, January 24, 2013 3:37:52 PM UTC-5, David Pilato wrote: > > > Each node uses its own storage. > > But you can use SharedFS: [Elasticsearch Platform — Find real-time answers at scale | Elastic](http://www.elasticsearch.org/guide/reference/modules/gateway/fs.html) > > > > But, why do you want to do that? > > > > -- > > David 😉 > > Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs > > > > Le 24 janv. 2013 à 21:14, Jimi Parekh [jimip...@gmail.com](mailto:jimip...@gmail.com) a écrit : > > > > To create multiple nodes in elasticsearh cluster, do we need to start the multiple elasticsearch instance? > > I have a one elasticsearch storage on one server. How can I create cluster with muliple nodes which reads the same storage? > > > > ## -Jimi. > > -- -- ---
**Author:** ![Jimi\_Parekh](https://avatars.discourse-cdn.com/v4/letter/j/7feea3/32.png) [@Jimi\_Parekh](https://discuss.elastic.co/u/Jimi_Parekh)\ **Post date:** [January 24, 2013, 9:46pm UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499/6 "2013-01-24T21:46:35Z")
Thats what my confusion. Suppose I start new node on the same sever (server-1) and server-1 is down than none of the node will be availble,right? So I want to add a new node which runs on another server say ,server-2 but still access the same index as server-1. How to do that ? On Thursday, January 24, 2013 4:02:36 PM UTC-5, David Pilato wrote: > So just launch a new node and let Elasticsearch manage that for you. > > All data are replicated on nodes by Elasticsearch so you don't have to > share anything here. > > HTH > > -- > David 😉 > Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs > > Le 24 janv. 2013 à 21:56, Jimi Parekh \<[jimip...@gmail.com](mailto:jimip...@gmail.com) \\> > a écrit : > > I have one elasticsearch index stored on server-1. Now I want to make sure > if server-1 is down, still users can query the index and get the result. > I am not sure whats the best possible way to accomplish it. I thought by > adding new nodes i can do it. > > On Thursday, January 24, 2013 3:37:52 PM UTC-5, David Pilato wrote: > > > Each node uses its own storage. > > But you can use SharedFS: > > [Elasticsearch Platform — Find real-time answers at scale | Elastic](http://www.elasticsearch.org/guide/reference/modules/gateway/fs.html) > > > > But, why do you want to do that? > > > > -- > > David 😉 > > Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs > > > > Le 24 janv. 2013 à 21:14, Jimi Parekh [jimip...@gmail.com](mailto:jimip...@gmail.com) a écrit : > > > > To create multiple nodes in elasticsearh cluster, do we need to start the > > multiple elasticsearch instance? > > I have a one elasticsearch storage on one server. How can I create > > cluster with muliple nodes which reads the same storage? > > > > -Jimi. > > > > -- > > -- -- ---
**Author:** ![polyfractal](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/polyfractal/32/48162_2.png) [@polyfractal](https://discuss.elastic.co/u/polyfractal)\ **Post date:** [January 25, 2013, 12:58am UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499/7 "2013-01-25T00:58:02Z")
If the nodes are configured to use the same cluster name, and you have enough replicas provisioned (e.g. \> 0) then Elasticsearch will automatically migrate a replica to the new node and rebalance everything. There isn't a good reason to start two nodes on the same server, other than testing how ES works. On Thursday, January 24, 2013 4:46:35 PM UTC-5, Jimi Parekh wrote: > Thats what my confusion. Suppose I start new node on the same sever > (server-1) and server-1 is down than none of the node will be > availble,right? > So I want to add a new node which runs on another server say ,server-2 but > still access the same index as server-1. How to do that ? > > On Thursday, January 24, 2013 4:02:36 PM UTC-5, David Pilato wrote: > > > So just launch a new node and let Elasticsearch manage that for you. > > > > All data are replicated on nodes by Elasticsearch so you don't have to > > share anything here. > > > > HTH > > > > -- > > David 😉 > > Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs > > > > Le 24 janv. 2013 à 21:56, Jimi Parekh [jimip...@gmail.com](mailto:jimip...@gmail.com) a écrit : > > > > I have one elasticsearch index stored on server-1. Now I want to make > > sure if server-1 is down, still users can query the index and get the > > result. > > I am not sure whats the best possible way to accomplish it. I thought by > > adding new nodes i can do it. > > > > On Thursday, January 24, 2013 3:37:52 PM UTC-5, David Pilato wrote: > > > > > Each node uses its own storage. > > > But you can use SharedFS: > > > [Elasticsearch Platform — Find real-time answers at scale | Elastic](http://www.elasticsearch.org/guide/reference/modules/gateway/fs.html) > > > > > > But, why do you want to do that? > > > > > > -- > > > David 😉 > > > Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs > > > > > > Le 24 janv. 2013 à 21:14, Jimi Parekh [jimip...@gmail.com](mailto:jimip...@gmail.com) a écrit : > > > > > > To create multiple nodes in elasticsearh cluster, do we need to start > > > the multiple elasticsearch instance? > > > I have a one elasticsearch storage on one server. How can I create > > > cluster with muliple nodes which reads the same storage? > > > > > > -Jimi. > > > > > > -- > > > > -- ---
**Author:** ![pounraj\_manikandan](https://avatars.discourse-cdn.com/v4/letter/p/9de0a6/32.png) [@pounraj\_manikandan](https://discuss.elastic.co/u/pounraj_manikandan)\ **Post date:** [January 25, 2013, 7:52am UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499/8 "2013-01-25T07:52:28Z")
Hello Jimi, As David says Elasticsearch starts allocating shards according our setup whenever a new node has been added to the same cluster (if your index is not corrupted). I believe below link will give you more detailed: [http://www.elasticsearch.org/guide/reference/modules/cluster.html ](http://www.elasticsearch.org/guide/reference/modules/cluster.html) If you start a new instance **in different machine** , copy the data directory from failed machine (if index is not crashed & the directory is not shard to new machine) with similar cluster settings. Elasticsearch makes the shards available for you. If you start a new instance **in same machine** pointing to same data directory & clustername you are done (also depends on other cluster configurations). Elasticsearch maintains its storage in below structure (by default) for any new instance started clustername/nodes/0/indices/indexname/SHARDS if you start one more instances clustername/nodes/0/indices/indexname/SHARDS clustername/nodes/1/indices/indexname/SHARDS (moves some from node0 .. if you dont have any copy setup) Pounraj Manikandan ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 6, 2017, 2:54am UTC](https://discuss.elastic.co/t/elasticsearch-cluster/10499/9 "2017-07-06T02:54:39Z")