Add e2e test for mounting different paths on same volume on same node #173
Conversation
k8s-ci-robot
added
do-not-merge/work-in-progress
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: wongma7 The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
k8s-ci-robot
added
approved
wongma7
force-pushed
the
subpath-test
branch
2 times, most recently
from
841b93d
to
416d786
Compare
k8s-ci-robot
added
size/XL
wongma7
force-pushed
the
subpath-test
branch
2 times, most recently
from
718905b
to
21de65d
Compare
k8s-ci-robot
added
needs-rebase
k8s-ci-robot
added
size/XL
k8s-ci-robot
added
size/L
wongma7
changed the title
k8s-ci-robot
removed
the
do-not-merge/work-in-progress
|
/test pull-aws-efs-csi-driver-e2e https://prow.k8s.io/view/gcs/kubernetes-jenkins/pr-logs/pull/kubernetes-sigs_aws-efs-csi-driver/173/pull-aws-efs-csi-driver-e2e/1268220967159599105 looks like an instance of #141. Saw similar 'mount.nfs4...blocked for more than 120 seconds.' messages in dmesg with no mounts of fs-286c882d on node ip-172-20-119-122.us-west-2.compute.internal ever succeeding |
|
/assign @leakingtapan |
| defer func() { _ = f.ClientSet.CoreV1().PersistentVolumes().Delete(pvRoot.Name, &metav1.DeleteOptions{}) }() | ||
|
|
||
| ginkgo.By(fmt.Sprintf("Creating pod to make subpaths /a and /b")) | ||
| pod := e2epod.MakePod(f.Namespace.Name, nil, []*v1.PersistentVolumeClaim{pvcRoot}, false, "mkdir /mnt/volume1/a && mkdir /mnt/volume1/b") |
There was a problem hiding this comment.
Where "/mnt/volume1" is defined:
volumeMounts[index] = v1.VolumeMount{Name: volumename, MountPath: "/mnt/" + volumename}
| framework.ExpectNoError(e2epod.WaitForPodSuccessInNamespace(f.ClientSet, pod.Name, f.Namespace.Name), "waiting for pod success") | ||
|
|
||
| ginkgo.By(fmt.Sprintf("Creating efs pvc & pv with subpath /a")) | ||
| pvcA, pvA, err := createEFSPVCPV(f.ClientSet, f.Namespace.Name, f.Namespace.Name+"-a", "/a") |
There was a problem hiding this comment.
Wish Golang support keywords argument like Python does. It's not easy to tell what those parameters really represent.
There was a problem hiding this comment.
agree lol. a pattern I have seen is to add /* */ after every arg but thats gets tiring and verbose fast. IDE can help here.
| defer func() { _ = f.ClientSet.CoreV1().PersistentVolumes().Delete(pvRoot.Name, &metav1.DeleteOptions{}) }() | ||
|
|
||
| ginkgo.By(fmt.Sprintf("Creating pod to make subpaths /a and /b")) | ||
| pod := e2epod.MakePod(f.Namespace.Name, nil, []*v1.PersistentVolumeClaim{pvcRoot}, false, "mkdir /mnt/volume1/a && mkdir /mnt/volume1/b") |
There was a problem hiding this comment.
The interface of MakePod
func MakePod(ns string, nodeSelector map[string]string, pvclaims []*v1.PersistentVolumeClaim, isPrivileged bool, command string) *v1.Pod {
| pod = e2epod.MakePod(f.Namespace.Name, nil, []*v1.PersistentVolumeClaim{pvcA, pvcB}, false, "") | ||
| pod, err = f.ClientSet.CoreV1().Pods(f.Namespace.Name).Create(pod) | ||
| framework.ExpectNoError(err, "creating pod") | ||
| framework.ExpectNoError(e2epod.WaitForPodNameRunningInNamespace(f.ClientSet, pod.Name, f.Namespace.Name), "waiting for pod running") |
There was a problem hiding this comment.
Is an additional check on whether the mount is accessible necessary, i.e. kubectl exec ... ls /mnt/volume1?
There was a problem hiding this comment.
no, the pod from the initial step "ginkgo.By(fmt.Sprintf("Creating pod to make subpaths /a and /b"))" would have failed if somehow mnt is not accessible
| ginkgo.By(fmt.Sprintf("Creating efs pvc & pv with no subpath")) | ||
| pvcRoot, pvRoot, err := createEFSPVCPV(f.ClientSet, f.Namespace.Name, f.Namespace.Name+"-root", "/") | ||
| framework.ExpectNoError(err, "creating efs pvc & pv with no subpath") | ||
| defer func() { _ = f.ClientSet.CoreV1().PersistentVolumes().Delete(pvRoot.Name, &metav1.DeleteOptions{}) }() |
There was a problem hiding this comment.
Are PVC and Pods getting cleaned up automatically?
There was a problem hiding this comment.
yes, since namespaces gets cleaned up automatically
|
@jqmichael: changing LGTM is restricted to collaborators In response to this:
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. |
|
/lgtm |
k8s-ci-robot
added
the
lgtm
|
/test pull-aws-efs-csi-driver-verify |
Is this a bug fix or adding new feature? fix #105
What is this PR about? / Why do we need it? #100 was fixed but we need an e2e test for it.
What testing is done?
Testing on my EKS cluster:
Testing on CI's kops cluster (https://prow.k8s.io/view/gcs/kubernetes-jenkins/pr-logs/pull/kubernetes-sigs_aws-efs-csi-driver/173/pull-aws-efs-csi-driver-e2e/1268250579340103680):