# How to avoid cluster re balance when node face GC Old gen

**URL:** <https://discuss.elastic.co/t/how-to-avoid-cluster-re-balance-when-node-face-gc-old-gen/91094>\
**Category:** Elasticsearch\
**Created:** [June 28, 2017, 11:48am UTC](https://discuss.elastic.co/t/how-to-avoid-cluster-re-balance-when-node-face-gc-old-gen/91094 "2017-06-28T11:48:54Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Eli\_Revach](https://avatars.discourse-cdn.com/v4/letter/e/e9c0ed/32.png) [@Eli\_Revach](https://discuss.elastic.co/u/Eli_Revach)\
**Post date:** [June 28, 2017, 11:48am UTC](https://discuss.elastic.co/t/how-to-avoid-cluster-re-balance-when-node-face-gc-old-gen/91094/1 "2017-06-28T11:48:54Z")

</div>

Hi ,  
ES=1.7.5  
From time to we keep getting GC Old gen (we are working on ways to minimize it ) , ES trigger long cluster re balance . Anyway to avoid cluster re balance on such event ?

---

<div class="post-metadata">

**Author:** ![Julien](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/julien/32/19688_2.png) [@Julien](https://discuss.elastic.co/u/Julien)\
**Post date:** [June 29, 2017, 10:17am UTC](https://discuss.elastic.co/t/how-to-avoid-cluster-re-balance-when-node-face-gc-old-gen/91094/2 "2017-06-29T10:17:11Z")

</div>

Please take a look at this documentation which covers this question:  
[https://www.elastic.co/guide/en/elasticsearch/reference/1.7/delayed-allocation.html](https://www.elastic.co/guide/en/elasticsearch/reference/1.7/delayed-allocation.html)

And the real solution is as you mentioned to minimise long GCs as much as possible ([https://www.elastic.co/guide/en/elasticsearch/guide/current/\_monitoring\_individual\_nodes.html](https://www.elastic.co/guide/en/elasticsearch/guide/current/_monitoring_individual_nodes.html))

---

<div class="post-metadata">

**Author:** ![Eli\_Revach](https://avatars.discourse-cdn.com/v4/letter/e/e9c0ed/32.png) [@Eli\_Revach](https://discuss.elastic.co/u/Eli_Revach)\
**Post date:** [June 29, 2017, 1:06pm UTC](https://discuss.elastic.co/t/how-to-avoid-cluster-re-balance-when-node-face-gc-old-gen/91094/3 "2017-06-29T13:06:29Z")

</div>

Hi ,  
I already try :  
curl -XPUT 'localhost:9200/\_all/\_settings' -d '{  
"settings": {  
"index.unassigned.node\_left.delayed\_timeout": "5m"  
}  
}'  
And still seeing long re balance , i consider setting cluster.routing.allocation.balance.threshold" : "100.0f , but it will stop the balancer

---

<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:** [July 27, 2017, 1:06pm UTC](https://discuss.elastic.co/t/how-to-avoid-cluster-re-balance-when-node-face-gc-old-gen/91094/4 "2017-07-27T13:06:31Z")

</div>

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