# RPO and backup

**URL:** <https://discuss.elastic.co/t/rpo-and-backup/208558>\
**Category:** Elasticsearch\
**Created:** [November 19, 2019, 6:24pm UTC](https://discuss.elastic.co/t/rpo-and-backup/208558 "2019-11-19T18:24:00Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![granier](https://avatars.discourse-cdn.com/v4/letter/g/b3f665/32.png) [@granier](https://discuss.elastic.co/u/granier)\
**Post date:** [November 19, 2019, 6:24pm UTC](https://discuss.elastic.co/t/rpo-and-backup/208558/1 "2019-11-19T18:24:01Z")

</div>

Hi,

After reading some topics here, i get a question about backup and RPO.

If I take a snapshot every day, my RPO could be 24h : that's a lots.

But taking a snapshot every hour or every 10' or more is a lots too ...

With Elasticsearch is there something like Oracle transaction logs ?

I remembered that there are some luccene buffers but they are in memory and as i understood they can not be used to restore a cluster.

Thanks for any help.

B. Granier  
Elasticsearch certified

---

<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:** [November 19, 2019, 6:32pm UTC](https://discuss.elastic.co/t/rpo-and-backup/208558/2 "2019-11-19T18:32:49Z")

</div>

You could use cross cluster replication to replicate to a separate cluster. That is probably the best way to minimize RPO.

---

<div class="post-metadata">

**Author:** ![granier](https://avatars.discourse-cdn.com/v4/letter/g/b3f665/32.png) [@granier](https://discuss.elastic.co/u/granier)\
**Post date:** [November 20, 2019, 7:56am UTC](https://discuss.elastic.co/t/rpo-and-backup/208558/3 "2019-11-20T07:56:10Z")

</div>

Yes but for that a "new" cluster is needed and there is Always the question of the distance between clusters nodes.

---

<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:** [November 20, 2019, 8:00am UTC](https://discuss.elastic.co/t/rpo-and-backup/208558/4 "2019-11-20T08:00:12Z")

</div>

A new cluster is needed but distance should not be a problem with cross cluster replication as it is designed to replicate data across regions.

---

<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 18, 2019, 8:09am UTC](https://discuss.elastic.co/t/rpo-and-backup/208558/5 "2019-12-18T08:09:05Z")

</div>

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