# Suggested value for vm.max\_map\_count on a multi-node machine

**URL:** <https://discuss.elastic.co/t/suggested-value-for-vm-max-map-count-on-a-multi-node-machine/35021>\
**Category:** Elasticsearch\
**Created:** [November 19, 2015, 7:44am UTC](https://discuss.elastic.co/t/suggested-value-for-vm-max-map-count-on-a-multi-node-machine/35021 "2015-11-19T07:44:30Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Chris\_Fraschetti](https://avatars.discourse-cdn.com/v4/letter/c/9d8465/32.png) [@Chris\_Fraschetti](https://discuss.elastic.co/u/Chris_Fraschetti)\
**Post date:** [November 19, 2015, 7:44am UTC](https://discuss.elastic.co/t/suggested-value-for-vm-max-map-count-on-a-multi-node-machine/35021/1 "2015-11-19T07:44:30Z")

</div>

> **[Configuration | Elasticsearch Guide \[1.7\] | Elastic](https://www.elastic.co/guide/en/elasticsearch/reference/1.7/setup-configuration.html)**

> Elasticsearch also uses a mix of NioFS and MMapFS for the various files. Ensure that you configure the maximum map count so that there is ample virtual memory available for mmapped files. This can be set temporarily:

> sysctl -w vm.max\_map\_count=262144  
> Or you can set it permanently by modifying vm.max\_map\_count setting in your /etc/sysctl.conf.

Assuming the machine has sufficient memory (256GB total, 4 ES nodes w/ 30G each), would it be a fair assumption that 262144 \* 4 (1048576) would be a decent value to use here?

I know this setting itself doesn't actually imply any memory usage but rather ES's ability to allocate more mmap areas.

Input is welcomed.

Thanks!

---

<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 5, 2017, 11:37pm UTC](https://discuss.elastic.co/t/suggested-value-for-vm-max-map-count-on-a-multi-node-machine/35021/2 "2017-07-05T23:37:23Z")

</div>


