# Replication to DR site, without CCR

**URL:** <https://discuss.elastic.co/t/replication-to-dr-site-without-ccr/253316>\
**Category:** Elasticsearch\
**Tags:** ccr-cross-cluster-replication, slm-snapshot-lifecycle-management\
**Created:** [October 26, 2020, 1:40pm UTC](https://discuss.elastic.co/t/replication-to-dr-site-without-ccr/253316 "2020-10-26T13:40:38Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![AndySw](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/andysw/32/77879_2.png) [@AndySw](https://discuss.elastic.co/u/AndySw)\
**Post date:** [October 26, 2020, 1:40pm UTC](https://discuss.elastic.co/t/replication-to-dr-site-without-ccr/253316/1 "2020-10-26T13:40:38Z")

</div>

Hi all. I was trying to find out which options exist if you want to synchronise a primary site with a DR site, both running seperate Elasticsearch clusters.

I'm aware of CCR, but it is cost prohibitive. What is the art of the possible without using CCR and what are the issues & edge cases we would need to bear in mind?

Many thanks in advance,

Andy.

---

<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:** [October 26, 2020, 1:45pm UTC](https://discuss.elastic.co/t/replication-to-dr-site-without-ccr/253316/2 "2020-10-26T13:45:48Z")

</div>

Easiest is probably to index/update both clusters in parallel, possibly using a message queue to enable buffering.

---

<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:** [October 26, 2020, 2:12pm UTC](https://discuss.elastic.co/t/replication-to-dr-site-without-ccr/253316/4 "2020-10-26T14:12:33Z")

</div>

It is a complex problem to solve well, especially when considering various failure scenarios. That is probably why it is an enterprise feature. It is not easy, but I still think it is the easiest option available.

---

<div class="post-metadata">

**Author:** ![AndySw](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/andysw/32/77879_2.png) [@AndySw](https://discuss.elastic.co/u/AndySw)\
**Post date:** [November 4, 2020, 11:07am UTC](https://discuss.elastic.co/t/replication-to-dr-site-without-ccr/253316/5 "2020-11-04T11:07:10Z")

</div>

Thank you @Christian_Dahlqvist & @riferrei for your advice.

We've conducted an analysis of the data we are storing in Elasticsearch and have identified that the business critical data is actually being saved using Filebeat & Logstash.

I think there might be a simpler approach where we synchronise the logs to the secondary application server and let them be ingested into the secondary Elasticsearch node, and vice-versa.

Any thoughts you guys have would be gladly recieved.

Andy.

---

<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:** [December 2, 2020, 11:07am UTC](https://discuss.elastic.co/t/replication-to-dr-site-without-ccr/253316/6 "2020-12-02T11:07:19Z")

</div>

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