# Discovery.zen.ping.unicast.hosts configuration

**URL:** <https://discuss.elastic.co/t/discovery-zen-ping-unicast-hosts-configuration/69234>\
**Category:** Elasticsearch\
**Created:** [December 16, 2016, 4:39am UTC](https://discuss.elastic.co/t/discovery-zen-ping-unicast-hosts-configuration/69234 "2016-12-16T04:39:17Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![ginger](https://avatars.discourse-cdn.com/v4/letter/g/34f0e0/32.png) [@ginger](https://discuss.elastic.co/u/ginger)\
**Post date:** [December 16, 2016, 4:39am UTC](https://discuss.elastic.co/t/discovery-zen-ping-unicast-hosts-configuration/69234/1 "2016-12-16T04:39:17Z")

</div>

Hi.

When I setup our es cluster, discovery.zen.ping.unicast.hosts of all nodes is set to ip of three master nodes. After a period of time, all the master nodes are replaced one by one. Now, I restart some of the data nodes, these nodes can never rejoin the cluster again except I update discovery.zen.ping.unicast.hosts configuration item.

It is not elegant that user have to update every node's configuration when all master nodes are replaced. I wonder that if node store cluster info locally or not? can node rejoin the cluster through these info automatically?

Any suggestions would be appreciated : )

---

<div class="post-metadata">

**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\
**Post date:** [December 16, 2016, 4:53am UTC](https://discuss.elastic.co/t/discovery-zen-ping-unicast-hosts-configuration/69234/2 "2016-12-16T04:53:29Z")

</div>

Use a CNAME that points to multiple hosts, then just manage that cname as needed.

---

<div class="post-metadata">

**Author:** ![Christian\_Dahlqvist](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/christian_dahlqvist/32/4617_2.png) [@Christian\_Dahlqvist](https://discuss.elastic.co/u/Christian_Dahlqvist)\
**Post date:** [December 16, 2016, 6:43am UTC](https://discuss.elastic.co/t/discovery-zen-ping-unicast-hosts-configuration/69234/3 "2016-12-16T06:43:04Z")

</div>

Why are you replacing all of your master nodes in the first place?

---

<div class="post-metadata">

**Author:** ![ginger](https://avatars.discourse-cdn.com/v4/letter/g/34f0e0/32.png) [@ginger](https://discuss.elastic.co/u/ginger)\
**Post date:** [December 19, 2016, 1:51am UTC](https://discuss.elastic.co/t/discovery-zen-ping-unicast-hosts-configuration/69234/4 "2016-12-19T01:51:38Z")

</div>

It's a good solution, but need some effort to manage CNAME. It's more elegant to allow node rejoin the cluster automatically.

---

<div class="post-metadata">

**Author:** ![ginger](https://avatars.discourse-cdn.com/v4/letter/g/34f0e0/32.png) [@ginger](https://discuss.elastic.co/u/ginger)\
**Post date:** [December 19, 2016, 2:36am UTC](https://discuss.elastic.co/t/discovery-zen-ping-unicast-hosts-configuration/69234/5 "2016-12-19T02:36:01Z")

</div>

> [@Christian\_Dahlqvist](#):
>
> Why are you replacing all of your master nodes in the first place?

It's has the same problem if I use some other node as the value of discovery.zen.ping.unicast.hosts.

---

<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:** [January 16, 2017, 2:36am UTC](https://discuss.elastic.co/t/discovery-zen-ping-unicast-hosts-configuration/69234/6 "2017-01-16T02:36:05Z")

</div>

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