Disable uuid checks on XFS #788
Conversation
k8s-ci-robot
added
release-note
k8s-ci-robot
added
the
size/S
|
/assign |
k8s-ci-robot
added
the
lgtm
By default, XFS does not allow mounting two devices that have the same UUID of the XFS filesystem. Therefore it's not possible to mount a volume + its restored snapshot. Therefore disable UUID check in XFS using a mount option.
k8s-ci-robot
removed
the
lgtm
|
Fixed |
|
/retests |
|
/lgtm I agree those look like flakes, created #789. |
k8s-ci-robot
added
the
lgtm
|
@mattcary, can you please also |
|
/approve It doesn't look like you need the approval but I never figure out what github wants. Unfortunately it looks like the test flake is the problem. We're looking into that. |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: jsafrane, mattcary 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
the
approved
|
/retest It seems like the flake is pretty severe but ATM not obvious what is going on. |
|
/retest |
1 similar comment
|
/retest |
|
fyi I think we're going to push a change reducing parallelism in the
presubmits that hopefully will reduce these flakes.
…On Mon, Jun 14, 2021 at 11:20 AM Elana Hashman ***@***.***> wrote:
/retest
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#788 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AIJCBADPAERNEQNT4DTR6N3TSZB67ANCNFSM46MMJ3XQ>
.
|
|
By reducing the parallelism of the tests in #790 , all the tests passed except "Kubernetes e2e suite: External Storage [Driver: csi-gcepd-sc-standard] [Testpattern: Dynamic PV (xfs)][Slow] multiVolume [Slow] should concurrently access the volume and restored snapshot from pods on the same node [LinuxOnly][Feature:VolumeSnapshotDataSource][Feature:VolumeSourceXFS]", which needs the fix in #788 . So @mattcary , @jsafrane can you please make a change similar to #790 in this PR to unblock this. |
|
I think I can force-merge #790, that will be easier.
…On Mon, Jun 14, 2021 at 4:56 PM Saikat Roychowdhury < ***@***.***> wrote:
By reducing the parallelism of the tests in #790
<#790>
, all the tests passed except "Kubernetes e2e suite: External Storage
[Driver: csi-gcepd-sc-standard] [Testpattern: Dynamic PV (xfs)][Slow]
multiVolume [Slow] should concurrently access the volume and restored
snapshot from pods on the same node
[LinuxOnly][Feature:VolumeSnapshotDataSource][Feature:VolumeSourceXFS]",
which needs the fix in #788
<#788>
. So @mattcary <https://github.com/mattcary> , @jsafrane
<https://github.com/jsafrane> can you please make a change similar to #790
<#790>
in this PR to unblock this.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#788 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AIJCBAFNCNMSROF7BZ6HT3DTS2JLBANCNFSM46MMJ3XQ>
.
|
|
That worked, feel the admin power coursing through my fingers :-) /test pull-gcp-compute-persistent-disk-csi-driver-kubernetes-integration |
Thanks that works too! |
|
/retest |
|
Sigh |
|
/retest |
|
@jsafrane: The following test failed, say
Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR. 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. I understand the commands that are listed here. |
|
My suggestion would be to merge this PR, and continue investigation as part of #789. The internal test grids https://testgrid.corp.google.com/gcp-compute-persistent-disk-csi-driver#pd-master-sidecars-latest-k8s-master-on-gce have significantly reduced flakes except for one test failure which will be fixed by this PR. |
|
/retest |
Cherry pick #788 to release-1.2
What type of PR is this?
/kind bug
What this PR does / why we need it:
By default, XFS does not allow mounting two devices that have the same UUID of the XFS filesystem. Therefore it's not possible to mount a volume + its restored snapshot.
Therefore disable UUID check in XFS using a mount option.
Since kubernetes/kubernetes#102538 was merged, all pull-kubernetes-e2e-gce-storage-slow tests fail in Kubernetes, e.g. https://prow.k8s.io/view/gs/kubernetes-jenkins/pr-logs/pull/102690/pull-kubernetes-e2e-gce-storage-slow/1402241210831081472/
Does this PR introduce a user-facing change?: