# Allocation awareness with many VMs per host

**URL:** <https://discuss.elastic.co/t/allocation-awareness-with-many-vms-per-host/117064>\
**Category:** Elasticsearch\
**Created:** [January 25, 2018, 3:23pm UTC](https://discuss.elastic.co/t/allocation-awareness-with-many-vms-per-host/117064 "2018-01-25T15:23:43Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![janhoy](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/janhoy/32/14384_2.png) [@janhoy](https://discuss.elastic.co/u/janhoy)\
**Post date:** [January 25, 2018, 3:23pm UTC](https://discuss.elastic.co/t/allocation-awareness-with-many-vms-per-host/117064/1 "2018-01-25T15:23:43Z")

</div>

Hi,

Deploying ES across four VMWare ESX hosts, with about 3-4 ES VMs per host. I plan to configure allocation awareness based on physical host ID, but since this virtual rig is configured to automatically start a dead VM on one of the other ESX hosts, that means the host ID is not static for each ES node, so we cannot hardcode it in the start script.

Further the hosting provider tells me that is is impossible from within the guest VM (RHEL7) to dynamically obtain a stable host identifier. I tell them not to move VMs around, but the answer is that is is automatic and would cause extra manual work to stop VMs from moving during e.g. planned maintenance on one physical host, so they like to avoid that.

I'm sure we're not the first to have such a setup. How have others solved this?

Jan

---

<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 22, 2018, 3:24pm UTC](https://discuss.elastic.co/t/allocation-awareness-with-many-vms-per-host/117064/2 "2018-02-22T15:24:07Z")

</div>

Closed on February 22, 2018, 3:24pm UTC

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