Add docs on scheduler MultiPoint config #30442

damemi · 2021-11-11T14:51:25Z

This adds documentation for the scheduler's new MultiPoint config field, along with examples and comparisons to the old way of configuring the scheduler.

Enhancement issue: kubernetes/enhancements#2891
Work PR: kubernetes/kubernetes#105611

netlify · 2021-11-11T14:51:30Z

👷 Deploy Preview for kubernetes-io-vnext-staging processing.

🔨 Explore the source changes: f7f336c

🔍 Inspect the deploy log: https://app.netlify.com/sites/kubernetes-io-vnext-staging/deploys/61a65ee8ae674f0007fa2756

damemi

/cc @alculquicondor @ahg-g @Huang-Wei
ptal, and let me know any other edge cases I should cover

jlbutler · 2021-11-11T15:34:01Z

/assign @nate-double-u

nate-double-u · 2021-11-15T21:03:18Z

Deploy preview: https://deploy-preview-30442--kubernetes-io-vnext-staging.netlify.app/docs/reference/scheduling/config/#multipoint-config

RinkiyaKeDad · 2021-11-16T10:01:32Z

content/en/docs/reference/scheduling/config.md

+```
+
+Specific extension points can be excluded from `MultiPoint` expansion using 
+the `disabled` fields. This works with disabling default plugins, out-of-tree plugins, 


nit: should it be field instead?

Suggested change

the `disabled` fields. This works with disabling default plugins, out-of-tree plugins,

the `disabled` field. This works with disabling default plugins, out-of-tree plugins,

I think we should avoid talking about out-of-tree plugins in official k8s documentation.

I don't disagree, but since this is different than how previous extension points treated out-of-tree plugins and this does apply to them, should those aspects be noted somewhere? Maybe part of the scheduling framework docs?

content/en/docs/reference/scheduling/config.md

sftim · 2021-11-16T17:46:42Z

content/en/docs/reference/scheduling/config.md

@@ -255,7 +255,183 @@ the same configuration parameters (if applicable). This is because the scheduler
 only has one pending pods queue.
 {{< /note >}}

-## {{% heading "whatsnext" %}}
+### MultiPoint config


I'd write:

Suggested change

### MultiPoint config

## Plugins that apply to all extension points {#multipoint}

The fragment identifier will then be multipoint.

Also, under “Extension points” (line 47), I'd mention and hyperlink to this section.

made this multiple extension points rather than all, think that's clearer

sftim · 2021-11-16T17:47:07Z

/sig scheduling

sftim · 2021-11-16T17:47:50Z

If this is a new field that doesn't work in older Kubernetes, we should also document that detail.

Huang-Wei

Thanks for the write-up! LGTM except for some nits.

content/en/docs/reference/scheduling/config.md

nate-double-u

A couple of comments inline:

content/en/docs/reference/scheduling/config.md

alculquicondor · 2021-11-22T18:43:31Z

content/en/docs/reference/scheduling/config.md

+```
+
+Specific extension points can be excluded from `MultiPoint` expansion using 
+the `disabled` fields. This works with disabling default plugins, out-of-tree plugins, 


I think we should avoid talking about out-of-tree plugins in official k8s documentation.

content/en/docs/reference/scheduling/config.md

alculquicondor · 2021-11-22T18:46:40Z

content/en/docs/reference/scheduling/config.md

+profiles:
+  - schedulerName: multipoint-scheduler
+    plugins:
+      score:


multiPoint? Order only really matters for Filter

Yeah I was trying to show a couple things here, that we can reorder and re-weight the plugins. Order technically only matters for Filter, true, but I think this still gets both of those points across

sounds good

alculquicondor · 2021-11-22T18:49:57Z

content/en/docs/reference/scheduling/config.md

+* Disables `DefaultPlugin1`, but only for `Filter`
+* Reorders `DefaultPlugin2` to run first in `Score` (even before the custom plugins)
+
+Without `MultiPoint` configuration, the above snippet would equate to this:


Maybe explicitly disable * in multiPoint to make it clearer?

In this section I'm actually showing what the equivalent in a pre-v1beta3 config would be. I tried updating the wording here to make that clearer. I could still add the disable: * to multiPoint if you think that would help too

alculquicondor · 2021-11-23T20:16:59Z

content/en/docs/reference/scheduling/config.md

+
+In older versions of the config, without `multiPoint`, the above snippet would equate to this:
+```yaml
+apiVersion: kubescheduler.config.k8s.io/v1beta3


I think this either should be v1beta2 or just add the disable *.

went with v1beta2 and pointed that out. multiPoint.disabled: * would mean all of the other default plugins need to be explicitly enabled which isn't exactly the comparison I was trying to draw here

alculquicondor · 2021-11-23T20:17:49Z

content/en/docs/reference/scheduling/config.md

+profiles:
+  - schedulerName: multipoint-scheduler
+    plugins:
+      score:


sounds good

content/en/docs/reference/scheduling/config.md

alculquicondor · 2021-11-24T14:33:58Z

content/en/docs/reference/scheduling/config.md

@@ -341,7 +341,7 @@ extension point in the future, the `multiPoint` config will automatically enable
 for the new extension.

 Specific extension points can be excluded from `MultiPoint` expansion using 
-the `disabled` field. This works with disabling default plugins, out-of-tree plugins, 
+the `disabled` field. This works with disabling default plugins, non-default plugins, 


Maybe let's say:

using the disabled field for the extension point.

alculquicondor

Do we have to add any extra notes in #30404?

IIUC, people using v1beta2 could just swap the version without needing extra changes, right?

alculquicondor · 2021-11-24T14:42:36Z

content/en/docs/reference/scheduling/config.md

@@ -78,6 +78,8 @@ extension points:
   least one bind plugin is required.
 1. `postBind`: This is an informational extension point that is called after
   a Pod has been bound.
+1. `multiPoint`: This is a config-only field that allows plugins to be enabled for 


enabled or disabled

damemi · 2021-11-24T14:49:44Z

Do we have to add any extra notes in #30404?

IIUC, people using v1beta2 could just swap the version without needing extra changes, right?

Right, an existing v1beta2 config will work with v1beta3 without any changes

alculquicondor · 2021-11-24T15:17:19Z

/lgtm

/sig scheduling

k8s-ci-robot · 2021-11-24T15:17:40Z

LGTM label has been added.

Git tree hash: 99fd7c5079bc28eedd04ff2c0b2e6d721f4969fb

nate-double-u

This looks good from a docs standpoint. I have a couple suggestions inline to simplify or clarify a couple sentences. I don't think these are strictly necessary for merge though.

/lgtm
/cc @jlbutler

content/en/docs/reference/scheduling/config.md

nate-double-u

I'm sorry, I didn't scroll down far enough on the deploy preview.

The "What's next" heading needs to be fixed:

content/en/docs/reference/scheduling/config.md

damemi · 2021-11-30T15:28:22Z

Updated and squashed, will need another lgtm when anyone has the chance

content/en/docs/reference/scheduling/config.md

nate-double-u · 2021-11-30T17:55:46Z

/lgtm

k8s-ci-robot · 2021-11-30T17:56:07Z

LGTM label has been added.

Git tree hash: b2573535f5bab19c8998ebb0b5eb577eacf1b067

jlbutler · 2021-11-30T18:53:22Z

looks like final editorial and tech comments have been addressed, thanks!
/approve

k8s-ci-robot · 2021-11-30T18:54:26Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: jlbutler

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:

~~content/en/OWNERS~~ [jlbutler]

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

k8s-ci-robot added this to the 1.23 milestone Nov 11, 2021

k8s-ci-robot added the cncf-cla: yes Indicates the PR's author has signed the CNCF CLA. label Nov 11, 2021

damemi commented Nov 11, 2021

View reviewed changes

k8s-ci-robot requested review from ahg-g, alculquicondor and Huang-Wei November 11, 2021 14:51

damemi mentioned this pull request Nov 11, 2021

Simplified Multi-point plugin configuration for scheduler kubernetes/enhancements#2891

Closed

4 tasks

k8s-ci-robot added the size/L Denotes a PR that changes 100-499 lines, ignoring generated files. label Nov 11, 2021

k8s-ci-robot requested review from onlydole and rajeshdeshpande02 November 11, 2021 14:53

k8s-ci-robot added language/en Issues or PRs related to English language sig/docs Categorizes an issue or PR as relevant to SIG Docs. labels Nov 11, 2021

k8s-ci-robot assigned nate-double-u Nov 11, 2021

damemi mentioned this pull request Nov 11, 2021

Scheduler simplified MultiPoint plugin config kubernetes/kubernetes#105611

Merged

RinkiyaKeDad reviewed Nov 16, 2021

View reviewed changes

sftim requested changes Nov 16, 2021

View reviewed changes

Huang-Wei reviewed Nov 16, 2021

View reviewed changes

content/en/docs/reference/scheduling/config.md Outdated Show resolved Hide resolved

nate-double-u reviewed Nov 16, 2021

View reviewed changes

content/en/docs/reference/scheduling/config.md Outdated Show resolved Hide resolved

content/en/docs/reference/scheduling/config.md Outdated Show resolved Hide resolved

k8s-ci-robot added the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Nov 17, 2021

alculquicondor reviewed Nov 22, 2021

View reviewed changes

damemi force-pushed the scheduler-multipoint-config branch from d99496d to 4c1c01e Compare November 23, 2021 19:38

k8s-ci-robot removed the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Nov 23, 2021

alculquicondor reviewed Nov 23, 2021

View reviewed changes

alculquicondor reviewed Nov 24, 2021

View reviewed changes

damemi force-pushed the scheduler-multipoint-config branch from 618450e to d1b0a46 Compare November 24, 2021 14:49

k8s-ci-robot added the sig/scheduling Categorizes an issue or PR as relevant to SIG Scheduling. label Nov 24, 2021

k8s-ci-robot assigned alculquicondor Nov 24, 2021

k8s-ci-robot added the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Nov 24, 2021

nate-double-u reviewed Nov 26, 2021

View reviewed changes

content/en/docs/reference/scheduling/config.md Outdated Show resolved Hide resolved

content/en/docs/reference/scheduling/config.md Outdated Show resolved Hide resolved

k8s-ci-robot requested a review from jlbutler November 26, 2021 22:41

nate-double-u requested changes Nov 26, 2021

View reviewed changes

content/en/docs/reference/scheduling/config.md Outdated Show resolved Hide resolved

damemi force-pushed the scheduler-multipoint-config branch from d1b0a46 to e880cd2 Compare November 30, 2021 15:28

k8s-ci-robot removed the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Nov 30, 2021

alculquicondor reviewed Nov 30, 2021

View reviewed changes

content/en/docs/reference/scheduling/config.md Show resolved Hide resolved

content/en/docs/reference/scheduling/config.md Outdated Show resolved Hide resolved

Add docs on scheduler MultiPoint config

f7f336c

damemi force-pushed the scheduler-multipoint-config branch from e880cd2 to f7f336c Compare November 30, 2021 17:27

k8s-ci-robot added the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Nov 30, 2021

k8s-ci-robot added the approved Indicates a PR has been approved by an approver from all required OWNERS files. label Nov 30, 2021

k8s-ci-robot merged commit 37e00b2 into kubernetes:dev-1.23 Nov 30, 2021

	the `disabled` fields. This works with disabling default plugins, out-of-tree plugins,
	the `disabled` field. This works with disabling default plugins, out-of-tree plugins,

	### MultiPoint config
	## Plugins that apply to all extension points {#multipoint}

Add docs on scheduler MultiPoint config #30442

Add docs on scheduler MultiPoint config #30442

Conversation

damemi commented Nov 11, 2021

netlify bot commented Nov 11, 2021 • edited Loading

damemi left a comment

Choose a reason for hiding this comment

jlbutler commented Nov 11, 2021

nate-double-u commented Nov 15, 2021

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

sftim commented Nov 16, 2021

sftim commented Nov 16, 2021

Huang-Wei left a comment

Choose a reason for hiding this comment

nate-double-u left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

alculquicondor left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

damemi commented Nov 24, 2021

alculquicondor commented Nov 24, 2021

k8s-ci-robot commented Nov 24, 2021

nate-double-u left a comment

Choose a reason for hiding this comment

nate-double-u left a comment • edited Loading

Choose a reason for hiding this comment

damemi commented Nov 30, 2021

nate-double-u commented Nov 30, 2021

k8s-ci-robot commented Nov 30, 2021

jlbutler commented Nov 30, 2021

k8s-ci-robot commented Nov 30, 2021

netlify bot commented Nov 11, 2021 •

edited

Loading

nate-double-u left a comment •

edited

Loading