# GCS Snapshots suddenly failing in ECK

**URL:** <https://discuss.elastic.co/t/gcs-snapshots-suddenly-failing-in-eck/210949>\
**Category:** Elastic Cloud on Kubernetes (ECK)\
**Created:** [December 7, 2019, 3:53am UTC](https://discuss.elastic.co/t/gcs-snapshots-suddenly-failing-in-eck/210949 "2019-12-07T03:53:06Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![tadgh](https://avatars.discourse-cdn.com/v4/letter/t/9f8e36/32.png) [@tadgh](https://discuss.elastic.co/u/tadgh)\
**Post date:** [December 7, 2019, 3:53am UTC](https://discuss.elastic.co/t/gcs-snapshots-suddenly-failing-in-eck/210949/1 "2019-12-07T03:53:07Z")

</div>

Hello Elastic Wizards!

I'm on ECK 0.8 at the moment, and I have been using the snapshot functionality on GCP quite successfully, with no issues for a long time. [I followed these instructions](https://www.elastic.co/guide/en/cloud-on-k8s/0.9/k8s-snapshot.html) and it has been working quite well. I literally have not touched the cluster in forever. I went to check on my snapshots, and I now notice that since a few weeks ago, I get partial shard failures .

 ![](https://us1.discourse-cdn.com/elastic/original/3X/4/2/4245d0048b46565af29e22a364e1b9ad2a0967c6.png)

Nothing changed in the cluster state during this time. I've gone over elastic logs for those days, nothing was upgraded or added.

The errors look like this:

 ![](https://us1.discourse-cdn.com/elastic/original/3X/6/5/65f1bd6a86a5a61fcb146e4318f049f7c5132b5c.png)

If there were an error accessing the bucket, you would think that _all_ of the indices would fail.

Am I missing something obvious here?

Thanks for reading,

--Tadgh

---

<div class="post-metadata">

**Author:** ![tadgh](https://avatars.discourse-cdn.com/v4/letter/t/9f8e36/32.png) [@tadgh](https://discuss.elastic.co/u/tadgh)\
**Post date:** [December 13, 2019, 5:36am UTC](https://discuss.elastic.co/t/gcs-snapshots-suddenly-failing-in-eck/210949/2 "2019-12-13T05:36:43Z")

</div>

For anybody stumbling around trying to solve this issue for themselves, I've gotten it to disappear by literally just creating a new service account, and using it instead.

---

<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:** [November 4, 2022, 7:36am UTC](https://discuss.elastic.co/t/gcs-snapshots-suddenly-failing-in-eck/210949/3 "2022-11-04T07:36:28Z")

</div>


