# New node not getting added **URL:** \ **Category:** Elasticsearch\ **Created:** [February 8, 2016, 7:14pm UTC](https://discuss.elastic.co/t/new-node-not-getting-added/41216 "2016-02-08T19:14:51Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![abinay](https://avatars.discourse-cdn.com/v4/letter/a/838e76/32.png) [@abinay](https://discuss.elastic.co/u/abinay)\ **Post date:** [February 8, 2016, 7:14pm UTC](https://discuss.elastic.co/t/new-node-not-getting-added/41216/1 "2016-02-08T19:14:51Z")
Hii I am trying to add a new node with elasticsearch version 2.2.0 to my existing cluster that contains only single node with elasticsearch version 2.1.0 . I got to know that its possible to mix these two versions. Now When I do curl -XGET '[http://localhost:9200/\_nodes](http://localhost:9200/_nodes)' I don't see the newly added node . This clearly means my new node has not been added right ? So what can be the reason for this ? ---
**Author:** ![w0lverine](https://avatars.discourse-cdn.com/v4/letter/w/90db22/32.png) [@w0lverine](https://discuss.elastic.co/u/w0lverine)\ **Post date:** [February 9, 2016, 11:44am UTC](https://discuss.elastic.co/t/new-node-not-getting-added/41216/2 "2016-02-09T11:44:15Z")
Is there any particular reason you are not using the same version? It might not be the cause now but as you scale up ELK it could cause problems down the road. Also what is your set-up like? Did you define your "localhost" as the "master node"? ---
**Author:** ![abinay](https://avatars.discourse-cdn.com/v4/letter/a/838e76/32.png) [@abinay](https://discuss.elastic.co/u/abinay)\ **Post date:** [February 9, 2016, 1:13pm UTC](https://discuss.elastic.co/t/new-node-not-getting-added/41216/3 "2016-02-09T13:13:21Z")
@w0lverine I tried to upgrade the node using this link - [https://www.elastic.co/guide/en/elasticsearch/reference/current/rolling-upgrades.html](https://www.elastic.co/guide/en/elasticsearch/reference/current/rolling-upgrades.html) I lost my old logs probably because I did not stop the node during upgradation . Am I right ? I just did step 3 because my cluster has only one node. Am I right in thinking so ?? ---
**Author:** ![w0lverine](https://avatars.discourse-cdn.com/v4/letter/w/90db22/32.png) [@w0lverine](https://discuss.elastic.co/u/w0lverine)\ **Post date:** [February 9, 2016, 11:07pm UTC](https://discuss.elastic.co/t/new-node-not-getting-added/41216/4 "2016-02-09T23:07:22Z")
Yes I wouldn't skip around the documentation and do steps out of order unless you know what you are doing. Start from step 1: disable shard allocation step 2: go ahead and stop non-essential indexing(if you are able too) step 3: stop and upgrade single node ..... If you suddenly stop the node it will not replicate the shards that were being processed during the time of the node shutdown. Hence why disable shard allocation before shutdown of node. ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 5, 2017, 11:17pm UTC](https://discuss.elastic.co/t/new-node-not-getting-added/41216/5 "2017-07-05T23:17:49Z")