# How to gracefully restart the cluster without downtime?

**URL:** <https://forums.percona.com/t/how-to-gracefully-restart-the-cluster-without-downtime/23142>\
**Category:** Percona Operator for MySQL\
**Created:** [June 27, 2023, 9:36am UTC](https://forums.percona.com/t/how-to-gracefully-restart-the-cluster-without-downtime/23142 "2023-06-27T09:36:19Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![b602fa9c](https://avatars.discourse-cdn.com/v4/letter/b/6bbea6/32.png) [@b602fa9c](https://forums.percona.com/u/b602fa9c)\
**Post date:** [June 27, 2023, 9:36am UTC](https://forums.percona.com/t/how-to-gracefully-restart-the-cluster-without-downtime/23142/1 "2023-06-27T09:36:19Z")

</div>

Hi,

Is there a recommended way to gracefully restart the pods one by one without taking the service down?  
I tried a `kubectl rollout restart statefulset perconamysql-pxc-db-pxc` but it didn’t trigger any restarts.

Thanks

---

<div class="post-metadata">

**Author:** ![matthewb](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/matthewb/32/34_2.png) [@matthewb](https://forums.percona.com/u/matthewb)\
**Post date:** [June 27, 2023, 12:55pm UTC](https://forums.percona.com/t/how-to-gracefully-restart-the-cluster-without-downtime/23142/2 "2023-06-27T12:55:52Z")

</div>

Hello @b602fa9c,  
You might want to take a look at [this documentation](https://docs.percona.com/percona-operator-for-mysql/pxc/update_manually.html). Why do you need to restart the cluster?

---

<div class="post-metadata">

**Author:** ![b602fa9c](https://avatars.discourse-cdn.com/v4/letter/b/6bbea6/32.png) [@b602fa9c](https://forums.percona.com/u/b602fa9c)\
**Post date:** [June 29, 2023, 11:54am UTC](https://forums.percona.com/t/how-to-gracefully-restart-the-cluster-without-downtime/23142/3 "2023-06-29T11:54:08Z")

</div>

Ok so we need to modify the pxc config in some way? That’s what I did previously (changed the CPU limit slightly) but it would be nice to do it without having to change any settings.

The use case is if we need to pull in a new version of a non-percona sidecar container, e.g istio-proxy.

---

<div class="post-metadata">

**Author:** ![Slava\_Sarzhan](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/slava_sarzhan/32/3196_2.png) [@Slava\_Sarzhan](https://forums.percona.com/u/Slava_Sarzhan)\
**Post date:** [June 29, 2023, 12:01pm UTC](https://forums.percona.com/t/how-to-gracefully-restart-the-cluster-without-downtime/23142/4 "2023-06-29T12:01:43Z")

</div>

Hi @b602fa9c, I suggest using custom annotation for it. You can add `spec.pxc.annotations: restarted-at: "28-06-23` to your CR, and all pxc pods will be restarted.
