# Auto-backup + auto-restore from S3? (failure mitigation best practices)

**URL:** <https://discuss.elastic.co/t/auto-backup-auto-restore-from-s3-failure-mitigation-best-practices/178505>\
**Category:** Elasticsearch\
**Created:** [April 25, 2019, 4:20pm UTC](https://discuss.elastic.co/t/auto-backup-auto-restore-from-s3-failure-mitigation-best-practices/178505 "2019-04-25T16:20:10Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![alexandergunnarson](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/alexandergunnarson/32/45038_2.png) [@alexandergunnarson](https://discuss.elastic.co/u/alexandergunnarson)\
**Post date:** [April 25, 2019, 4:20pm UTC](https://discuss.elastic.co/t/auto-backup-auto-restore-from-s3-failure-mitigation-best-practices/178505/1 "2019-04-25T16:20:11Z")

</div>

Hey — so I'm trying to set up an Elasticsearch cluster while making sure that I'm accounting for failure situations. I understand that Curator can easily snapshot to S3 (I'm running on AWS), which is great — I was considering running it as a cron job on every node (passing the `--master-only` flag so it only runs on the elected master node) every N minutes. However, I'm not aware of an easy way to auto-restore in the case of a node that fails, assuming the node is not replicated. What is best practice here?

And while we're at it, what are some good ways to mitigate snapshot corruption?

Thanks!

---

<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:** [May 23, 2019, 4:20pm UTC](https://discuss.elastic.co/t/auto-backup-auto-restore-from-s3-failure-mitigation-best-practices/178505/2 "2019-05-23T16:20:13Z")

</div>

Closed on May 23, 2019, 4:20pm UTC

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