# Repurposing master node as data node

**URL:** <https://discuss.elastic.co/t/repurposing-master-node-as-data-node/284993>\
**Category:** Elasticsearch\
**Created:** [September 23, 2021, 10:58am UTC](https://discuss.elastic.co/t/repurposing-master-node-as-data-node/284993 "2021-09-23T10:58:54Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![elastic-learner](https://avatars.discourse-cdn.com/v4/letter/e/b5ac83/32.png) [@elastic-learner](https://discuss.elastic.co/u/elastic-learner)\
**Post date:** [September 23, 2021, 10:58am UTC](https://discuss.elastic.co/t/repurposing-master-node-as-data-node/284993/1 "2021-09-23T10:58:54Z")

</div>

I have a 3 node es cluster with all nodes having all roles. I want to convert these nodes to dedicated master nodes and add extra dedicated data nodes. For this the best approach is to evacuate all the shards from the node using allocation filtering and then setting **node.data: false**. Unfortunately, in my cluster there is an index with **auto\_expand\_replicas** set to _0-all_ and as a result one shard of that index is remaining in the node even after evacuation.

As an alternative I was considering **Elasticsearch-node repurpose** , but according to documentation([elasticsearch-node | Elasticsearch Guide [7.15] | Elastic](https://www.elastic.co/guide/en/elasticsearch/reference/current/node-tool.html)) this operation is unsafe.

Is there a way out of this situation?

---

<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:** [October 21, 2021, 10:59am UTC](https://discuss.elastic.co/t/repurposing-master-node-as-data-node/284993/2 "2021-10-21T10:59:04Z")

</div>

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