OCPNODE-1515 : Support Evented PLEG feature in Openshift by sairameshv · Pull Request #1458 · openshift/api

sairameshv · 2023-05-10T13:34:30Z

CRI-O sends the container events to the Kubelet so that the pod cache can be updated based on the received events.
More about the Evented PLEG is here - KEP Reference
This feature can be enabled in OCP by adding a new field in the node config custom resource
that can be monitored by the MCO and update both the Kubelet and CRI-O configurations
Enhancement PR: OCPNODE-1525: Support Evented PLEG in Openshift enhancements#1368

openshift-ci · 2023-05-10T13:34:38Z

Skipping CI for Draft Pull Request.
If you want CI signal for your change, please convert it to an actual PR.
You can still manually trigger a test run with /test all

openshift-ci · 2023-05-10T13:34:40Z

Hello @sairameshv! Some important instructions when contributing to openshift/api:
API design plays an important part in the user experience of OpenShift and as such API PRs are subject to a high level of scrutiny to ensure they follow our best practices. If you haven't already done so, please review the OpenShift API Conventions and ensure that your proposed changes are compliant. Following these conventions will help expedite the api review process for your PR.

sairameshv · 2023-05-24T17:02:45Z

/test verify

JoelSpeed · 2023-05-30T10:40:46Z

 	// +optional
 	WorkerLatencyProfile WorkerLatencyProfileType `json:"workerLatencyProfile,omitempty"`
+
+	// EventedPleg enables event based PLEG between the kubelet and the CRI-O


I think we want to expand this comment to provide more detail.

Questions I would be asking would be:

What are the valid values?

What happens if i didn't specify any value? Is this behaviour likely to change over time?

Nit, the comment should start with the json tag version of the name not the Go field version
So I'd be expecting something more along the lines of:

Suggested change

// EventedPleg enables event based PLEG between the kubelet and the CRI-O

// eventedPLEG enables event based PLEG between the kubelet and the CRI-O.

// Valid values are `Enabled`, `Disabled` and omitted.

// When omitted, this means no opinion and the platform is left to choose a reasonable default

// which is subject to change over time.

// The current default is `Disabled`.

Also, is it worthing rather than saying CRI-O, spelling that out in human readable prose?

I assume a user will know what PLEG is if they are turning this on?

Updated as suggested along with a reference to the KEP !!

JoelSpeed · 2023-05-30T10:41:37Z

 	DefaultUpdateDefaultReaction WorkerLatencyProfileType = "Default"
 )

+// +kubebuilder:validation:Enum=Enabled;Disabled;""


If you are using omitempty which I think you are doing because this is a workload API rather than a configuration API, you don't need to have "" in place, when the field is omitted the validation for the enum will not execute

Yeah, Removed the "" option and I think nodes.config API is a configuration API as this resource is unique cluster-wide

Is it a singleton resource or is there one per Node in the cluster? If it's a singleton then yep, it's a configuration API, in which case I would suggest including "" in the enum and dropping omitempty. This improves the discoverability of the API for the user to configure it

It is a singleton resource named "cluster" that has some of the node related configs like cgroupMode, workerlatencyprofiles etc.
I agree that removing omitempty improves the discoverability of the API for the user. At the same time, I want to maintain consistency with the other API fields already present which are again optional
WDYT?

At the same time, I want to maintain consistency with the other API fields already present which are again optional

We tend to say we don't repeat the sins of the past here. We have conventions that evolve over time so new fields should follow current conventions even when they make the field look inconsistent with older fields.

IMO this should be no omitempty, but allow empty string on the enum please.

Agreed with your point, Updated !!

JoelSpeed · 2023-06-02T16:02:15Z

How does this interact with #1471? What's the goal in 4.14?

Looks like 1471 enabled the feature by default in tech preview clusters, so is the intention here that a user will have the choice? It will be on by default but can be disabled by using the fields added here?

sairameshv · 2023-06-02T16:57:57Z

How does this interact with #1471? What's the goal in 4.14?

Looks like 1471 enabled the feature by default in tech preview clusters, so is the intention here that a user will have the choice? It will be on by default but can be disabled by using the fields added here?

#1471 Just provides a way to enable this feature via featuregate. I don't think it enables the EventedPLEG without getting this PR & the related MCO PR merged.
Default behavior of the EventedPLEG is disabled in the cluster.

JoelSpeed · 2023-06-02T17:00:29Z

I don't think it enables the EventedPLEG without getting this PR & the related MCO PR merged.

It does enable it for TechPreviewNoUpgrade clusters. MCO gets the feature gate passed through to kubelet and so it gets enabled by default on any TechPreviewNoUpgrade cluster, guessing that wasn't the intention?

Note, this has been noticed because of debugging issues in openshift/machine-config-operator#3688

JoelSpeed · 2023-06-06T10:21:41Z

+	// eventedPleg enables the event based PLEG between the kubelet and CRI-O
+	// Reference: https://github.com/kubernetes/enhancements/blob/master/keps/sig-node/3386-kubelet-evented-pleg/README.md
+	// Valid values are `Enabled`, `Disabled` and ""
+	// By default, the evented pleg feature is not enabled in the cluster


We typically have a particular prose we use for this, can you update please

Suggested change

// By default, the evented pleg feature is not enabled in the cluster

// When omitted, this means no opinion and the platform is left to choose a reasonable default, which is subject to change over time.

// The current default value is Disabled.

JoelSpeed · 2023-06-06T10:22:55Z

 	WorkerLatencyProfile WorkerLatencyProfileType `json:"workerLatencyProfile,omitempty"`
+
+	// eventedPleg enables the event based PLEG between the kubelet and CRI-O
+	// Reference: https://github.com/kubernetes/enhancements/blob/master/keps/sig-node/3386-kubelet-evented-pleg/README.md


Is there a user friendly explanation we can include? A link to a KEP is quite a lot of context.

Is the intention when a user enables this feature that CRIO and Kubelet are both configured? Is there explicit configuration required for both?

EventedPLEG is an upstream feature gate, what happens when that is enabled by default?

openshift-ci · 2024-01-24T11:13:17Z

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: sairameshv
Once this PR has been reviewed and has the lgtm label, please assign mfojtik for approval. For more information see the Kubernetes Code Review Process.

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

1. Incase of Evented PLEG, CRI-O sends the container events to the Kubelet so that the pod cache can be updated based on the received events. KEP Reference: https://github.com/kubernetes/enhancements/blob/master/keps/sig-node/3386-kubelet-evented-pleg/README.md 2. This feature can be enabled in OCP by adding a new field in the node config custom resource that can be monitored by the MCO and update both the required Kubelet and CRI-O configurations Enhancement PR: openshift/enhancements#1368 Signed-off-by: Sai Ramesh Vanka <svanka@redhat.com>

openshift-merge-robot · 2024-03-19T13:10:13Z

PR needs rebase.

Details

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.

openshift-bot · 2024-06-18T09:00:26Z

Issues go stale after 90d of inactivity.

Mark the issue as fresh by commenting /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.
Exclude this issue from closing by commenting /lifecycle frozen.

If this issue is safe to close now please do so with /close.

/lifecycle stale

openshift-bot · 2024-08-11T08:30:52Z

Stale issues rot after 30d of inactivity.

Mark the issue as fresh by commenting /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.
Exclude this issue from closing by commenting /lifecycle frozen.

If this issue is safe to close now please do so with /close.

/lifecycle rotten
/remove-lifecycle stale

openshift-ci · 2024-08-15T15:25:33Z

@sairameshv: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
ci/prow/e2e-aws-ovn	`8b0fb6b`	link	true	`/test e2e-aws-ovn`
ci/prow/e2e-aws-ovn-techpreview	`8b0fb6b`	link	true	`/test e2e-aws-ovn-techpreview`
ci/prow/e2e-aws-serial-techpreview	`8b0fb6b`	link	true	`/test e2e-aws-serial-techpreview`
ci/prow/e2e-aws-serial	`8b0fb6b`	link	true	`/test e2e-aws-serial`
ci/prow/e2e-aws-ovn-hypershift	`8b0fb6b`	link	true	`/test e2e-aws-ovn-hypershift`
ci/prow/e2e-upgrade-minor	`8b0fb6b`	link	true	`/test e2e-upgrade-minor`
ci/prow/e2e-upgrade	`8b0fb6b`	link	true	`/test e2e-upgrade`
ci/prow/minor-e2e-upgrade-minor	`8b0fb6b`	link	true	`/test minor-e2e-upgrade-minor`
ci/prow/minor-images	`8b0fb6b`	link	true	`/test minor-images`

Full PR test history. Your PR dashboard.

Details

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-sigs/prow repository. I understand the commands that are listed here.

openshift-bot · 2024-09-15T08:01:10Z

Rotten issues close after 30d of inactivity.

Reopen the issue by commenting /reopen.
Mark the issue as fresh by commenting /remove-lifecycle rotten.
Exclude this issue from closing again by commenting /lifecycle frozen.

/close

openshift-ci · 2024-09-15T08:01:50Z

@openshift-bot: Closed this PR.

Details

In response to this:

Rotten issues close after 30d of inactivity.

Reopen the issue by commenting /reopen.
Mark the issue as fresh by commenting /remove-lifecycle rotten.
Exclude this issue from closing again by commenting /lifecycle frozen.

/close

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-sigs/prow repository.

openshift-ci Bot added the do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. label May 10, 2023

openshift-ci Bot added the size/S Denotes a PR that changes 10-29 lines, ignoring generated files. label May 10, 2023

sairameshv changed the title ~~Support Evented PLEG feature in Openshift~~ OCPNODE-1515 : Support Evented PLEG feature in Openshift May 10, 2023

openshift-merge-robot added the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label May 22, 2023

mrunalp reviewed May 24, 2023

View reviewed changes

Comment thread config/v1/0000_10_config-operator_01_node.crd.yaml Outdated

sairameshv force-pushed the evented_pleg branch from 0f2d0a5 to adcc1b2 Compare May 24, 2023 16:09

openshift-merge-robot removed the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label May 24, 2023

sairameshv marked this pull request as ready for review May 24, 2023 16:11

openshift-ci Bot removed the do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. label May 24, 2023

openshift-ci Bot requested review from bparees and sjenning May 24, 2023 16:12

sairameshv force-pushed the evented_pleg branch from adcc1b2 to 92f0e84 Compare May 25, 2023 11:46

deads2k reviewed May 25, 2023

View reviewed changes

Comment thread config/v1/types_node.go Outdated

sairameshv marked this pull request as draft May 25, 2023 15:55

openshift-ci Bot added the do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. label May 25, 2023

sairameshv force-pushed the evented_pleg branch from 92f0e84 to b40ba4c Compare May 26, 2023 10:59

openshift-ci Bot added size/M Denotes a PR that changes 30-99 lines, ignoring generated files. and removed size/S Denotes a PR that changes 10-29 lines, ignoring generated files. labels May 26, 2023

sairameshv marked this pull request as ready for review May 26, 2023 11:08

openshift-ci Bot removed the do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. label May 26, 2023

openshift-ci Bot requested a review from JoelSpeed May 26, 2023 11:09

sairameshv force-pushed the evented_pleg branch 2 times, most recently from 7661898 to 6c1e4f0 Compare May 29, 2023 12:07

openshift-ci Bot added size/L Denotes a PR that changes 100-499 lines, ignoring generated files. and removed size/M Denotes a PR that changes 30-99 lines, ignoring generated files. labels May 29, 2023

JoelSpeed reviewed May 30, 2023

View reviewed changes

sairameshv force-pushed the evented_pleg branch from 6c1e4f0 to bcafa17 Compare May 31, 2023 08:22

sairameshv requested a review from JoelSpeed May 31, 2023 10:38

sairameshv force-pushed the evented_pleg branch from bcafa17 to 761cb1f Compare May 31, 2023 14:27

sairameshv mentioned this pull request Jun 5, 2023

Update openshift/api to disable EventedPLEG featuregate in techpreview openshift/cluster-config-operator#317

Merged

JoelSpeed reviewed Jun 6, 2023

View reviewed changes

sairameshv marked this pull request as draft January 24, 2024 11:12

openshift-ci Bot added the do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. label Jan 24, 2024

sairameshv force-pushed the evented_pleg branch from 761cb1f to 23a1dcc Compare January 24, 2024 11:12

sairameshv added 2 commits January 24, 2024 17:09

Enable evented pleg flag

8b0fb6b

sairameshv force-pushed the evented_pleg branch from 23a1dcc to 8b0fb6b Compare January 24, 2024 11:49

openshift-merge-robot added the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Mar 19, 2024

openshift-ci Bot added the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Jun 18, 2024

openshift-ci Bot added lifecycle/rotten Denotes an issue or PR that has aged beyond stale and will be auto-closed. and removed lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. labels Aug 11, 2024

openshift-ci Bot closed this Sep 15, 2024

-	// EventedPleg enables event based PLEG between the kubelet and the CRI-O
+// eventedPLEG enables event based PLEG between the kubelet and the CRI-O.
+// Valid values are `Enabled`, `Disabled` and omitted.
+// When omitted, this means no opinion and the platform is left to choose a reasonable default
+// which is subject to change over time.
+// The current default is `Disabled`.

	// By default, the evented pleg feature is not enabled in the cluster
	// When omitted, this means no opinion and the platform is left to choose a reasonable default, which is subject to change over time.
	// The current default value is Disabled.

Conversation

sairameshv commented May 10, 2023 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

openshift-ci Bot commented May 10, 2023

Uh oh!

openshift-ci Bot commented May 10, 2023

Uh oh!

Uh oh!

sairameshv commented May 24, 2023

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

JoelSpeed commented Jun 2, 2023

Uh oh!

sairameshv commented Jun 2, 2023

Uh oh!

JoelSpeed commented Jun 2, 2023

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

openshift-ci Bot commented Jan 24, 2024

Uh oh!

openshift-merge-robot commented Mar 19, 2024

Uh oh!

openshift-bot commented Jun 18, 2024

Uh oh!

openshift-bot commented Aug 11, 2024

Uh oh!

openshift-ci Bot commented Aug 15, 2024

Uh oh!

openshift-bot commented Sep 15, 2024

Uh oh!

openshift-ci Bot commented Sep 15, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

sairameshv commented May 10, 2023 •

edited

Loading