Add a subdirectory provisioner mechanism

[colearendt]

Is your feature request related to a problem?/Why is this needed

FSx Lustre filesystems are large (minimum 1.2T). As a result, it would be nice to be able to couple related applications within a particular FSx Lustre filesystem.

/feature

Describe the solution you'd like in detail

For instance:

• Create a StorageClass that targets an existing / static FSx Lustre filesystem
• When a PVC is requested against that storage class, a directory is created within the root of the system
• The PVC is filled / bound to this directory and made available to the application
• As a result, several applications have mutually exclusive access to a subdirectory of the FSx Lustre filesystem

Describe alternatives you've considered

• Independent FSx Lustre filesystem per application (cost prohibitive)
• Use several PVs (one per application) to target the same filesystem at different mountnames (mount sub-paths)
  • Select the corresponding PV with the application's PVC
  • Less than ideal because requires a PV per application
  • Also requires the subdirectories to be created manually within the FSx Lustre filesystem (or the mount / PVC binding will not succeed)

Additional context

This feature is prompted by the nfs-subdir-external-provisioner's behavior (although nfs-server-provisioner works much the same way too).

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[colearendt]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[colearendt]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[pat-s]

/remove-lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[pat-s]

/remove-lifecycle stale

[jacobwolfaws]

/kind feature

[jacobwolfaws]

/lifecycle frozen

[pat-s]

@jacobwolfaws Any news why it has been moved to 'frozen'?

[jacobwolfaws]

@pat-s Moving it to frozen so people don't have to keep removing the lifecycle stale tag. No updates on this particular feature

[DevSusu]

@colearendt I'm facing similar problem, where I want to use 1 FSx volume (1200GiB) in multiple namespaces, with needs multiple PV and PVCs. How would using two PVs with same mountpath work in this case? (seperate tenants by Group/User)