# Indexing and Recovery of Elasticsearch Nodes

**URL:** <https://discuss.elastic.co/t/indexing-and-recovery-of-elasticsearch-nodes/120396>\
**Category:** Elasticsearch\
**Created:** [February 19, 2018, 5:17am UTC](https://discuss.elastic.co/t/indexing-and-recovery-of-elasticsearch-nodes/120396 "2018-02-19T05:17:28Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![Ong](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/ong/32/40766_2.png) [@Ong](https://discuss.elastic.co/u/Ong)\
**Post date:** [February 19, 2018, 5:17am UTC](https://discuss.elastic.co/t/indexing-and-recovery-of-elasticsearch-nodes/120396/1 "2018-02-19T05:17:29Z")

</div>

I understand that any node in an Elasticsearch cluster can be indexed into and the data will be sent to other cluster nodes.

In the case where 1 node goes down, is it the responsibility of the application to switch to another node in the cluster to index data to?  
i.e. the application has to maintain a list of IPs or Hostnames of the cluster nodes and it needs to index the next node in the list in case the current node is unreachable.

Regarding the recovery of a node, for example, if node1 in a 3 node cluster goes down and during this time, nodes 2 and 3 continue to receive new index writes, when node1 recovers, will node1 automatically receive the new index writes from nodes 2 and 3?

---

<div class="post-metadata">

**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\
**Post date:** [February 19, 2018, 5:18am UTC](https://discuss.elastic.co/t/indexing-and-recovery-of-elasticsearch-nodes/120396/2 "2018-02-19T05:18:17Z")

</div>

> [@Ong](#):
>
> is it the responsibility of the application to switch to another node in the cluster to index data to?

Yes, if you use any of the clients we provide this functionality.

> [@Ong](#):
>
> when node1 recovers, will node1 automatically receive the new index writes from nodes 2 and 3?

Yep.

---

<div class="post-metadata">

**Author:** ![Ong](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/ong/32/40766_2.png) [@Ong](https://discuss.elastic.co/u/Ong)\
**Post date:** [February 19, 2018, 5:30am UTC](https://discuss.elastic.co/t/indexing-and-recovery-of-elasticsearch-nodes/120396/3 "2018-02-19T05:30:54Z")

</div>

Thanks for the reply Mark. I would like to clarify the case where if the current node goes down, the user application is responsible for handling the switching of IPs or does the Elastic REST API automatically handle it for the application, i.e. the switch is transparent to the user application?

---

<div class="post-metadata">

**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\
**Post date:** [February 19, 2018, 5:31am UTC](https://discuss.elastic.co/t/indexing-and-recovery-of-elasticsearch-nodes/120396/4 "2018-02-19T05:31:52Z")

</div>

Elasticsearch does not handle it, you need the client to do it.

---

<div class="post-metadata">

**Author:** ![Ong](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/ong/32/40766_2.png) [@Ong](https://discuss.elastic.co/u/Ong)\
**Post date:** [February 19, 2018, 5:32am UTC](https://discuss.elastic.co/t/indexing-and-recovery-of-elasticsearch-nodes/120396/5 "2018-02-19T05:32:23Z")

</div>

Thanks for the clarification

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\
**Post date:** [March 19, 2018, 5:32am UTC](https://discuss.elastic.co/t/indexing-and-recovery-of-elasticsearch-nodes/120396/6 "2018-03-19T05:32:45Z")

</div>

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.