# How to preserve state of ray serve on ray cluster restart?

**URL:** <https://discuss.ray.io/t/how-to-preserve-state-of-ray-serve-on-ray-cluster-restart/2033>\
**Category:** Ray Serve\
**Created:** [May 4, 2021, 1:31pm UTC](https://discuss.ray.io/t/how-to-preserve-state-of-ray-serve-on-ray-cluster-restart/2033 "2021-05-04T13:31:39Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Aleh\_Sauko](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/aleh_sauko/32/697_2.png) [@Aleh\_Sauko](https://discuss.ray.io/u/Aleh_Sauko)\
**Post date:** [May 4, 2021, 1:31pm UTC](https://discuss.ray.io/t/how-to-preserve-state-of-ray-serve-on-ray-cluster-restart/2033/1 "2021-05-04T13:31:40Z")

</div>

I tried to use ray serve for model deployment but when something go wrong with cluster,and I need to restart it how to preserve model already deployed to them?

Currently when cluster is restarted no any models deployed to ray[serve].