Allow mounting two different subdirectories on the same pod

[colearendt]

/kind bug

What happened?

Mounting two different subdirectories (/one and /two) of the same lustre filesystem onto the same pod fails, never resolves.

What you expected to happen?

Mounting two different subdirectories of the same lustre filesystem should succeed (without making other directories, e.g. /three and /four, available)

How to reproduce it (as minimally and precisely as possible)?

• Create two PVs referencing the same Lustre filesystem at different subdirectories
• Create two PVCs referencing these PVs
• Create two volumes / volumeMounts on the same pod that reference these PVCs and mount at different locations
• Notice that ContainerCreating never succeeds

  Type     Reason       Age    From     Message
  ----     ------       ----   ----     -------
  Warning  FailedMount  2m36s  kubelet  Unable to attach or mount volumes: unmounted volumes=[storage2], unattached volumes=[storage3 kube-api-access-8tsdh storage2]: timed out waiting for the condition
  Warning  FailedMount  20s    kubelet  Unable to attach or mount volumes: unmounted volumes=[storage2], unattached volumes=[storage2 storage3 kube-api-access-8tsdh]: timed out waiting for the condition

Anything else we need to know?:

Related to #247

Environment

• Kubernetes version (use kubectl version):
• Driver version:

[DevSusu]

+1 for the issue

Since Lustre storage can only be provisioned in unit of 1200GiBs, it is important that the volume can be distributed across namespaces, which needs multiple static PVs in 1 FSx volume.

[jacobwolfaws]

Hi @colearendt, I tried reproducing with your steps and didn't experience any issues. Can you share your pv.yaml's, pvc.yaml's, and pod.yaml (excluding any sensitive information)?

Edit: Can you also speak a little more about your use case here, where you want to mount multiple different subdirectories to a single pod? If you feel this is something multiple other users might hit we can add an example for future ref

[jacobwolfaws]

@colearendt Following up on this, are you still having this issue?

[colearendt]

Woops. Thanks for the follow up! I’ll give reproducing with more details a shot next week! Message ID: ***@***.***>

[jacobwolfaws]

Going to close out this issue since I wasn't able to repro and it doesn't seem like it's continuing to have impact, feel free to reopen with more details and/or this continues causing issues. Thanks!
/close

[k8s-ci-robot]

@jacobwolfaws: Closing this issue.

In response to this:

Going to close out this issue since I wasn't able to repro and it doesn't seem like it's continuing to have impact, feel free to reopen with more details and/or this continues causing issues. Thanks!
/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.