Parameter `fsGroupChangePolicy: Always` causes very slow pod start

[juan-ferrer-toribio]

Having fsGroupChangePolicy set to Always causes all pods using the volume defined in zammadConfig.storageVolume to take a very long time to start, when that volume begins to grow and uses a shared FS like CephFS or NFS it becomes unfeasible. In our case Helm deploy aborts because the timeout is exceeded.

I believe this is not viable for large environments and it would be better to set it to OnRootMismatch by default.

References:

• https://kubernetes.io/docs/tasks/configure-pod-container/security-context/#configure-volume-permission-and-ownership-change-policy-for-pods
• https://kubernetes.io/blog/2020/12/14/kubernetes-release-1.20-fsgroupchangepolicy-fsgrouppolicy/
• Allow volume ownership to be only set after fs formatting. kubernetes/kubernetes#69699

It was set by @alexkorotysh on #210

[monotek]

Feel free to create a PR which changes that.
I'm not entirely sure if changing the current default might be an issue for some users so maybe making it configurable would be the way to go....

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[juan-ferrer-toribio]

Here is the PR:

• Update fsGroupChangePolicy default on values.yaml #427