-
Notifications
You must be signed in to change notification settings - Fork 211
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
update images, add Kubernetes 1.20, prepare for v1.5.0 release #238
Conversation
/retest Test flake (?) on 1.20: "External Storage [Driver: hostpath.csi.k8s.io] [Testpattern: Dynamic PV (default fs)] subPath should support restarting containers using directory as subpath [Slow]" |
@@ -91,7 +91,7 @@ spec: | |||
name: csi-data-dir | |||
|
|||
- name: hostpath | |||
image: k8s.gcr.io/sig-storage/hostpathplugin:v1.4.0 | |||
image: k8s.gcr.io/sig-storage/hostpathplugin:v1.5.0 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Will CI start to fail for other repos if we merge this before releasing hostpathplugin v1.5.0?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
No. Those other pre-merge jobs intentionally use a fixed version of the hostpath repo to prevent such breakage.
What normally would have happened is that the periodic jobs (briefly) fail until the new image is available. But those jobs were already broken when adding the health check sidecar, so this is not going to make them worse 😅
CHANGELOG/CHANGELOG-1.5.md
Outdated
## Changes by Kind | ||
|
||
### Feature | ||
- Adds snapshot beta CRD deployment for 1.17 ([#98](https://github.com/kubernetes-csi/csi-driver-host-path/pull/98), [@xing-yang](https://github.com/xing-yang)) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We deploy snapshot CRDs in prow, not the hostpath driver right?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It was another PR that doesn't belong here.
CHANGELOG/CHANGELOG-1.5.md
Outdated
|
||
### Feature | ||
- Adds snapshot beta CRD deployment for 1.17 ([#98](https://github.com/kubernetes-csi/csi-driver-host-path/pull/98), [@xing-yang](https://github.com/xing-yang)) | ||
- Ephemeral inline volumes in addition to normal volumes on Kubernetes 1.16 ([#97](https://github.com/kubernetes-csi/csi-driver-host-path/pull/97), [@pohly](https://github.com/pohly)) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This has been available since 1.2.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think the release-notes tool is confused by the release-tools subtree commit messages and picking up csi-release-tools PR numbers. Does the new method of squashing fix this?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Darn. I did check one PR which looked odd and that turned out to be correct, so I trusted the tool for the rest 😢
I suppose squashing helps, but I haven't tried that myself - I don't do releases that often.
CHANGELOG/CHANGELOG-1.5.md
Outdated
|
||
### Bug or Regression | ||
- Add topology to the CreateVolumeResponse for existing volumes ([#231](https://github.com/kubernetes-csi/csi-driver-host-path/pull/231), [@Jiawei0227](https://github.com/Jiawei0227)) | ||
- Fixed raw block volumes on hosts that don't have /dev/loop- devices pre-defined ([#109](https://github.com/kubernetes-csi/csi-driver-host-path/pull/109), [@pohly](https://github.com/pohly)) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this is old?
CHANGELOG/CHANGELOG-1.5.md
Outdated
|
||
### Uncategorized | ||
- Build with Go 1.15 ([#198](https://github.com/kubernetes-csi/csi-driver-host-path/pull/198), [@pohly](https://github.com/pohly)) | ||
- Migrated to Go modules, so the source builds also outside of GOPATH. ([#103](https://github.com/kubernetes-csi/csi-driver-host-path/pull/103), [@pohly](https://github.com/pohly)) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
old?
@@ -1,4 +1,4 @@ | |||
apiVersion: storage.k8s.io/v1beta1 | |||
apiVersion: storage.k8s.io/v1 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
v1 is only available in 1.18
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So shall we rename kubernetes-1.17
to kubernetes-1.18
after all?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sounds good
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done and force pushed (because it is a fix for the incorrect use of v1 in kubernetes-1.17).
I've pushed a version where I manually verified each PR that gets listed in the changelog. |
Kubernetes 1.17 is no longer supported and by updating to 1.18 as minimal version we can use the GA CSIDriver v1 API. Also, all sidecar and driver images get updated. The CSI hostpath image update is important because the recently added health check sidecar depends on the new version. Kubernetes 1.20 is where officially the v1 snapshotter API is introduced. Therefore the new deployment for it also uses the external-snapshotter 4.x and the v1 APIs.
# are needed only because of condition explained in | ||
# https://github.com/kubernetes/kubernetes/issues/69608 | ||
|
||
kind: Service |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
did we miss removing these as part of #196?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Probably.
volumeMounts: | ||
- name: socket-dir | ||
mountPath: /csi | ||
- name: csi-external-health-monitor-controller |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Putting the controller here seems inconsistent with how we currently have other controller sidecars in separate manifests. I guess we didn't really conclude how we want to go in #192
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That came from #210. I just copied it.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
FWIW, I agree that it seems inconsistent. I don't know whether it had to be done this way.
/lgtm Let's followup on these cleanup items separately. |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: msau42, pohly 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 |
What type of PR is this?
/kind cleanup
What this PR does / why we need it:
Kubernetes 1.17 is no longer supported, but because all sidecars still
work on it there is no need to remove the "deploy/kubernetes-1.17"
folder either. Instead, all sidecar and driver images merely get
updated.
The CSI hostpath image update is important because the recently added
health check sidecar depends on the new version.
Kubernetes 1.20 is where officially the v1 snapshotter API is
introduced. Therefore the new deployment for it also uses the
external-snapshotter 4.x and the v1 APIs.
** Note for reviewers ***
I am including a
CHANGELOG/CHANGELOG-1.5.md
in this PR for the sake of expedience. It already includes this PR, so once it is merged, we can tag v1.5.0.Does this PR introduce a user-facing change?: