# Restore strategy

**URL:** <https://forums.percona.com/t/restore-strategy/31218>\
**Category:** Percona XtraBackup\
**Created:** [June 25, 2024, 9:26am UTC](https://forums.percona.com/t/restore-strategy/31218 "2024-06-25T09:26:06Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![midson](https://avatars.discourse-cdn.com/v4/letter/m/f14d63/32.png) [@midson](https://forums.percona.com/u/midson)\
**Post date:** [June 25, 2024, 9:26am UTC](https://forums.percona.com/t/restore-strategy/31218/1 "2024-06-25T09:26:06Z")

</div>

In K8s, MySQL replica pod might restart due to various reasons. When it is coming up, how do we decide, whether to restore or not? Any guidelines?

---

<div class="post-metadata">

**Author:** ![erpramodprajapat](https://avatars.discourse-cdn.com/v4/letter/e/c37758/32.png) [@erpramodprajapat](https://forums.percona.com/u/erpramodprajapat)\
**Post date:** [June 25, 2024, 1:35pm UTC](https://forums.percona.com/t/restore-strategy/31218/2 "2024-06-25T13:35:12Z")

</div>

What kind of setup you are using, is it cluster or stand alone ?  
if its stand alone than there is no need to restore until its DB is corrupted and if its cluster than it should join back to cluster when its come back

---

<div class="post-metadata">

**Author:** ![midson](https://avatars.discourse-cdn.com/v4/letter/m/f14d63/32.png) [@midson](https://forums.percona.com/u/midson)\
**Post date:** [June 26, 2024, 8:52am UTC](https://forums.percona.com/t/restore-strategy/31218/3 "2024-06-26T08:52:27Z")

</div>

@erpramodprajapat Thanks.

Cluster Mode. If we have GBs of Data, how do we know that the DB is in right state?. Bcoz if there is any issue, then manual/automatic intervention needed to bring back to right state. Due to the size of data, Sync up of full data from Master might not be the solution. How to find (automatic/ manual) and handle the restore?