Matt Cary
Matt Cary
CVE-2022-2068 highlights that libssl and openssl are an active CVE service. It's not clear how many applications need those libraries. Have we considered a non-ssl base image?
See https://testgrid.k8s.io/sig-storage-kubernetes#volume-snapshot. The failures started after https://github.com/kubernetes/kubernetes/pull/110204 was merged, which updated the controller from v4 to v6. Failures for the mock csi driver tests are due to the VolumeSnapshot not...
See https://github.com/kubernetes/kubernetes/pull/109541. Since cloud-provider-gcp installs the pd csi driver, it's safe to enable this now. Tested manually, the best I could and it seemed to pass. /assign @leiyiz
Expected behavior: kubetest2 builds and uses a local e2e.test Actual behavior: kubetest2 builds a local e2e.test, but uses a different one. I'm trying to test out a new e2e.test config...
kubetest2 does not plumb env vars to its dependent processes as described here: https://github.com/kubernetes-sigs/kubetest2/blob/master/kubetest2-gce/deployer/common.go#L89. From the comment that seems mostly an aesthetic decision, and a desire to understand the API...
It's already causing problems with #972. Refactoring is needed any to deal with the fact that GCP state is carried in a single driver instance across test cases, see #990.
/kind bug **What this PR does / why we need it**: Using a space causes an annoying spurious message to be printed during making. ```release-note None ``` /assign @amacaskill
Regional disks aren't listed along with zonal disks. So they never appear in the list volume calls, meaning they are continually getting re-attached, causing api quota problems and load across...
I don't think any code changes are needed, it's just adding the fsGroupPolicy field to the CSIDriver object for the appropriate overlays (the actual work here being to figure out...
Currently we just add labels to the standard storageclass in test/k8s-integration/configs/. This causes problems for test environments that are testing against master. /assign