# Elasticsearch and Snapshots?

**URL:** <https://discuss.elastic.co/t/elasticsearch-and-snapshots/243717>\
**Category:** Elasticsearch\
**Created:** [August 4, 2020, 1:46pm UTC](https://discuss.elastic.co/t/elasticsearch-and-snapshots/243717 "2020-08-04T13:46:29Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![Rysiu](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/rysiu/32/61919_2.png) [@Rysiu](https://discuss.elastic.co/u/Rysiu)\
**Post date:** [August 4, 2020, 1:46pm UTC](https://discuss.elastic.co/t/elasticsearch-and-snapshots/243717/1 "2020-08-04T13:46:29Z")

</div>

Hi,

I have very quick information about making snapshots at Elasticsearch.

The documentation is here:  
[https://www.elastic.co/guide/en/elasticsearch/reference/current/snapshots-register-repository.html](https://www.elastic.co/guide/en/elasticsearch/reference/current/snapshots-register-repository.html)

However, I haven't found any information on which elasticseach nodes must be mapped with additional remote disk resource for snapshots?

Does this resource have to be available for absolutely all elasticsearch nodes?  
Or only the data nodes can by used?  
What is a typical configuration?

---

<div class="post-metadata">

**Author:** ![elasticforme](https://avatars.discourse-cdn.com/v4/letter/e/f05b48/32.png) [@elasticforme](https://discuss.elastic.co/u/elasticforme)\
**Post date:** [August 4, 2020, 2:21pm UTC](https://discuss.elastic.co/t/elasticsearch-and-snapshots/243717/2 "2020-08-04T14:21:05Z")

</div>

all masters and data nodes

---

<div class="post-metadata">

**Author:** ![elasticforme](https://avatars.discourse-cdn.com/v4/letter/e/f05b48/32.png) [@elasticforme](https://discuss.elastic.co/u/elasticforme)\
**Post date:** [August 4, 2020, 2:26pm UTC](https://discuss.elastic.co/t/elasticsearch-and-snapshots/243717/3 "2020-08-04T14:26:50Z")

</div>

what I have done is that mount xfs file system to all masters/data node

and then run a bash script to do snapshot every other day and remove week old snapshot

It is three step process  
mount fs  
register snapshot (elk knows where to save data, name etc..)  
run actual backup (data gets backed up here)

[https://discuss.elastic.co/t/automatic-elasticsearch-snapshot-backup/173223](https://discuss.elastic.co/t/automatic-elasticsearch-snapshot-backup/173223)

---

<div class="post-metadata">

**Author:** ![Rysiu](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/rysiu/32/61919_2.png) [@Rysiu](https://discuss.elastic.co/u/Rysiu)\
**Post date:** [August 4, 2020, 6:00pm UTC](https://discuss.elastic.co/t/elasticsearch-and-snapshots/243717/4 "2020-08-04T18:00:22Z")

</div>

Coordinate node doesn't need access to this storage?

---

<div class="post-metadata">

**Author:** ![Armin\_Braun](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/armin_braun/32/20092_2.png) [@Armin\_Braun](https://discuss.elastic.co/u/Armin_Braun)\
**Post date:** [August 4, 2020, 7:48pm UTC](https://discuss.elastic.co/t/elasticsearch-and-snapshots/243717/5 "2020-08-04T19:48:00Z")

</div>

> [@Rysiu](#):
>
> Coordinate node doesn't need access to this storage?

No, only master- and data-nodes need access to the repository.

---

<div class="post-metadata">

**Author:** ![Steve\_Mushero](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/steve_mushero/32/22441_2.png) [@Steve\_Mushero](https://discuss.elastic.co/u/Steve_Mushero)\
**Post date:** [August 5, 2020, 4:44am UTC](https://discuss.elastic.co/t/elasticsearch-and-snapshots/243717/6 "2020-08-05T04:44:11Z")

</div>

I just wrote a blog on how Snapshots work that you may find useful - note why the repeated mounts & registrations? I'd think you'd leave mounted and registered all the time. Be sure to prune and note new wildcard deletes in V7.8. Also why every other day, not every day or more often? Pretty nice process that lends itself to frequent backups:

> **[How Elasticsearch Snapshots Work](https://www.elkman.io/blog/how-elasticsearch-snapshots-work)**
>
> Elasticsearch is a powerful and dynamic distributed data system, and such things can be hard to backup, especially as they scale into the terabytes and beyond. Fortunately, the fol

---

<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:** [September 2, 2020, 4:44am UTC](https://discuss.elastic.co/t/elasticsearch-and-snapshots/243717/7 "2020-09-02T04:44:12Z")

</div>

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