# Does anybody else have problems using EBS-backed PersistentVolume on Kubernetes?

**URL:** <https://forums.suse.com/t/does-anybody-else-have-problems-using-ebs-backed-persistentvolume-on-kubernetes/5215>\
**Category:** Rancher 1.x\
**Created:** [January 16, 2017, 11:22am UTC](https://forums.suse.com/t/does-anybody-else-have-problems-using-ebs-backed-persistentvolume-on-kubernetes/5215 "2017-01-16T11:22:00Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![emk](https://sea2.discourse-cdn.com/flex022/user_avatar/forums.suse.com/emk/32/2050_2.png) [@emk](https://forums.suse.com/u/emk)\
**Post date:** [January 16, 2017, 11:22am UTC](https://forums.suse.com/t/does-anybody-else-have-problems-using-ebs-backed-persistentvolume-on-kubernetes/5215/1 "2017-01-16T11:22:00Z")

</div>

Rancher is a really impressive piece of software! I love the GUI and the REST API.

I’m currently evaluating Rancher for hosting a Kubernetes cluster for Pachyderm. Pachyderm requires a `PersistentVolume` backed by an EBS volume, which fails with the following error:

> MountVolume.SetUp failed […] Output: mount: special device /var/lib/kubelet/plugins/kubernetes.io/aws-ebs/mounts/vol-\* does not exist

All the technical details are [described in issue #7370](https://github.com/rancher/rancher/issues/7370). There appears to have been an earlier bug with the same symptoms reported at #4444, but it was supposedly fixed a while ago. But if you read the issue on GitHub, you’ll see [another user has encountered this issue in a recent version](https://github.com/rancher/rancher/issues/4444#issuecomment-266211104).

Notably, this works fine on standard Kubernetes for me.

So, my questions:

- Has anybody else gotten EBS-backed `PersistentVolume`s to work with Rancher 1.3.0?
- Is it generally worthwhile to file bugs on the GitHub issues page if you don’t have a support contract? I see a lot of open issues with nobody assigned, so I’m guessing there’s just too much incoming issue volume.
- If this is a bug, where should I look in the code to figure out what’s breaking? I’m willing to attempt to fix this myself.

I’d be happy to try to dig into this, but it would probably take me a day just to find out where to begin, given the complexity of Kubernetes and the Rancher integration.

Anyway, thank you for any data that you can provide!

---

<div class="post-metadata">

**Author:** ![emk](https://sea2.discourse-cdn.com/flex022/user_avatar/forums.suse.com/emk/32/2050_2.png) [@emk](https://forums.suse.com/u/emk)\
**Post date:** [January 17, 2017, 11:07pm UTC](https://forums.suse.com/t/does-anybody-else-have-problems-using-ebs-backed-persistentvolume-on-kubernetes/5215/2 "2017-01-17T23:07:22Z")

</div>

This can be fixed by changing the Kubernetes cluster type from Rancher to AWS. See [issue #7370](https://github.com/rancher/rancher/issues/7370#issuecomment-273325303) for detailed steps. Thank you to everybody who helped sort this out!
