# Why our cluster some data nodes leaving cluster by itself?and why we have a node cpu/load so high?

**URL:** <https://discuss.elastic.co/t/why-our-cluster-some-data-nodes-leaving-cluster-by-itself-and-why-we-have-a-node-cpu-load-so-high/116466>\
**Category:** Elasticsearch\
**Created:** [January 22, 2018, 10:19am UTC](https://discuss.elastic.co/t/why-our-cluster-some-data-nodes-leaving-cluster-by-itself-and-why-we-have-a-node-cpu-load-so-high/116466 "2018-01-22T10:19:53Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![beaman](https://avatars.discourse-cdn.com/v4/letter/b/b5e925/32.png) [@beaman](https://discuss.elastic.co/u/beaman)\
**Post date:** [January 22, 2018, 10:19am UTC](https://discuss.elastic.co/t/why-our-cluster-some-data-nodes-leaving-cluster-by-itself-and-why-we-have-a-node-cpu-load-so-high/116466/1 "2018-01-22T10:19:53Z")

</div>

our cluster informations:  
nodes:14  
master nodes:3  
data nodes:32C/60G(m4.4xlarge)  
master nodes:4C/8G(c4.xlarge)  
disk:6TB SSD (10000IOPS)

first question:  
1, one of our cluster data nodes left cluster by itself, and we did not found any question from log, after  
several minutes ,it join cluster by itself, but all shards will be reassign.  
Second question:  
1, one of our cluster always have high cpu and load, we check the log, the log like this

"Caused by: java.lang.IllegalStateException: Can't get text on a START\_OBJECT at 1:620  
at org.elasticsearch.common.xcontent.json.JsonXContentParser.text(JsonXContentParser.java:88) ~[elasticsearch-5.3.1.jar:5.3.1]  
at org.elasticsearch.common.xcontent.support.AbstractXContentParser.textOrNull(AbstractXContentParser.java:221) ~[elasticsearch-5.3.1.jar:5.3.1]  
at org.elasticsearch.index.mapper.TextFieldMapper.parseCreateField(TextFieldMapper.java:380) ~[elasticsearch-5.3.1.jar:5.3.1]  
at org.elasticsearch.index.mapper.FieldMapper.parse(FieldMapper.java:287) ~[elasticsearch-5.3.1.jar:5.3.1]"

at last ,could you help me?

---

<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:** [February 19, 2018, 10:20am UTC](https://discuss.elastic.co/t/why-our-cluster-some-data-nodes-leaving-cluster-by-itself-and-why-we-have-a-node-cpu-load-so-high/116466/2 "2018-02-19T10:20:33Z")

</div>

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