# Cluster Provision in High Avaliability

**URL:** <https://forums.suse.com/t/cluster-provision-in-high-avaliability/39460>\
**Category:** SUSE Rancher Prime\
**Created:** [November 15, 2022, 3:08pm UTC](https://forums.suse.com/t/cluster-provision-in-high-avaliability/39460 "2022-11-15T15:08:10Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Herman\_Milton\_Maleia](https://sea2.discourse-cdn.com/flex022/user_avatar/forums.suse.com/herman_milton_maleia/32/9633_2.png) [@Herman\_Milton\_Maleia](https://forums.suse.com/u/Herman_Milton_Maleia)\
**Post date:** [November 15, 2022, 3:08pm UTC](https://forums.suse.com/t/cluster-provision-in-high-avaliability/39460/1 "2022-11-15T15:08:10Z")

</div>

Hi team,

Hope you are doing.  
I want to implement a high avaliabilty cluster for kubernetes using rancher.  
I’m following this link.

> **[Building a Highly Available Kubernetes Cluster | SUSE Communities](https://www.suse.com/c/rancher_blog/building-a-highly-available-kubernetes-cluster/)**
>
>   Introduction As Kubernetes becomes more and more prevalent as a container orchestration system, it is important to properly architect your Kubernetes clusters to be highly available. This guide covers some of the options that are available for...

 ![Diagrama sem nome.drawio](https://us1.discourse-cdn.com/flex022/uploads/suse/original/2X/1/1f5dc55b3a4e5fe06a316ee4ad6887025b4aaaee.png)

Do i need to create a vip in front of the etcd cluster?  
If yes, how does api-server will know about this vip?  
Where can i setup this configuration?  
Do i need to configure keepalived and other options or these options are granted by default?

---

<div class="post-metadata">

**Author:** ![Kamil\_Maslowski](https://sea2.discourse-cdn.com/flex022/user_avatar/forums.suse.com/kamil_maslowski/32/9588_2.png) [@Kamil\_Maslowski](https://forums.suse.com/u/Kamil_Maslowski)\
**Post date:** [November 19, 2022, 7:43pm UTC](https://forums.suse.com/t/cluster-provision-in-high-avaliability/39460/2 "2022-11-19T19:43:46Z")

</div>

Hi. I can recommend this guide [https://computingforgeeks.com/deploy-kubernetes-on-rocky-using-rke2/](https://computingforgeeks.com/deploy-kubernetes-on-rocky-using-rke2/)

---

<div class="post-metadata">

**Author:** ![Fraser\_Goffin](https://sea2.discourse-cdn.com/flex022/user_avatar/forums.suse.com/fraser_goffin/32/6608_2.png) [@Fraser\_Goffin](https://forums.suse.com/u/Fraser_Goffin)\
**Post date:** [November 29, 2022, 9:05am UTC](https://forums.suse.com/t/cluster-provision-in-high-avaliability/39460/3 "2022-11-29T09:05:51Z")

</div>

I don’t think that answers the OP’s question. In the topology he is proposing he has HA for Etcd separately from the other Control Plane nodes, whereas in that blog Etcd is running on the same nodes.