# How to do backup locally without data duplication?

**URL:** <https://discuss.elastic.co/t/how-to-do-backup-locally-without-data-duplication/49273>\
**Category:** Elasticsearch\
**Created:** [May 5, 2016, 9:20am UTC](https://discuss.elastic.co/t/how-to-do-backup-locally-without-data-duplication/49273 "2016-05-05T09:20:01Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![LaszloHont](https://avatars.discourse-cdn.com/v4/letter/l/b2d939/32.png) [@LaszloHont](https://discuss.elastic.co/u/LaszloHont)\
**Post date:** [May 5, 2016, 9:20am UTC](https://discuss.elastic.co/t/how-to-do-backup-locally-without-data-duplication/49273/1 "2016-05-05T09:20:01Z")

</div>

I've read a [previous question](https://discuss.elastic.co/t/hardware-snapshot-best-practice-advice/2301) which is now already a year old.

I have a backup agent running on a server, and I'd like to backup my elasticsearch data. I'd like to make sure that any backup I run will be consistent in case I need to restore it.

I've read the documentation about snapshotting, and it seems focused on backing up to remote machines, or to a different place on the same machine. I cannot backup to a remote machine and I'd like to avoid creating another copy of the data locally: ideally it would be great if I could quiesce the data locally and then let the backup agent run (unfreezing the data after the agent is done).

Is there a good way of achieving this?

I can see three options so far: i) stop elasticsearch before the backup, start it after the backup; ii) close all the indexes before the backup, reopen them after the backup; iii) just run the backup with no quiescing at all

Any recommendations?

Thanks a lot in advance for your help.

---

<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:** [May 5, 2016, 9:26am UTC](https://discuss.elastic.co/t/how-to-do-backup-locally-without-data-duplication/49273/2 "2016-05-05T09:26:40Z")

</div>

The first two would work, if you can deal with your cluster being unavailable. The third wouldn't as your data would change.

---

<div class="post-metadata">

**Author:** ![LaszloHont](https://avatars.discourse-cdn.com/v4/letter/l/b2d939/32.png) [@LaszloHont](https://discuss.elastic.co/u/LaszloHont)\
**Post date:** [May 5, 2016, 10:16am UTC](https://discuss.elastic.co/t/how-to-do-backup-locally-without-data-duplication/49273/3 "2016-05-05T10:16:23Z")

</div>

Thanks @warkolm

Can you think of a way of doing this without the downtime? Do you think elastic will consider adding a data quiesce feature in their commercial offering?

---

<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:** [May 5, 2016, 11:40am UTC](https://discuss.elastic.co/t/how-to-do-backup-locally-without-data-duplication/49273/4 "2016-05-05T11:40:54Z")

</div>

Probably not, no, we have the snapshot API.

---

<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:** [July 5, 2017, 10:53pm UTC](https://discuss.elastic.co/t/how-to-do-backup-locally-without-data-duplication/49273/5 "2017-07-05T22:53:44Z")

</div>