Roman Bednář
Roman Bednář
@wongma7 Hi Matthew, do you thing can we expect a fix for this issue?
/lgtm @wongma7 Can you please review an approve?
> @RomanBednar can you please take a look again to the latest commit? Thank you for the patch, I added lgtm label but we still need someone who can approve/merge...
Performance was tested with [kind](kind.sigs.k8s.io) on single node with 10 pods and 10 volumes per pod (=100 PVCs). Dynamic provisioning was enabled during the test. Cluster has been modified to...
Rollout-rollback-rollout testing was performed, feature behaves as expected and no issues were observed. Posting my tests results: ## Perform pre-upgrade tests Set default storage class: ```bash $ kc patch sc/csi-hostpath-sc...
Tested version skew and sharing results below. First case is with feature enabled in API server and disabled in KCM and the second case is the other way around. ##...
@mattcary Hi, I've noticed there are tests for NodeExpandVolume that are commented out with a note about fakeexec being too brittle: https://github.com/kubernetes-sigs/gcp-compute-persistent-disk-csi-driver/blob/00986499d2d4e8453cc85b2bee2fede5600dcc97/pkg/gce-pd-csi-driver/node_test.go#L454 But, by adding mount-utils resizer to NodeStageVolume I...
> I dunno, creating a mock that would actually catch errors or regressions is very hard. Maybe just stick with fakeexec for now and we'll see how it does. >...
I've noticed that mount-utils does not support [Resize()](https://github.com/kubernetes/mount-utils/blob/0141062e1d011f295eec1cdcea5c80280146bae3/resizefs_unsupported.go#L39) on platforms other than Linux. And later NeedResize() was [added](https://github.com/kubernetes/kubernetes/pull/109721) as well which is reasonable because having it without Resize() would not...