Skip to content

KEP-5142: Add SchedulerPopFromBackoffQ feature gate docs #49840

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.

Sign up for GitHub

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

Merged
merged 1 commit into from
Apr 8, 2025
Merged

KEP-5142: Add SchedulerPopFromBackoffQ feature gate docs #49840

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 15 additions & 0 deletions ...eference/command-line-tools-reference/feature-gates/SchedulerPopFromBackoffQ.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
---
title: SchedulerPopFromBackoffQ
content_type: feature_gate
_build:
list: never
render: false

stages:
- stage: beta
defaultValue: true
fromVersion: "1.33"
---

Improves scheduling queue behavior by popping pods from the backoffQ when the activeQ is empty.
This allows to process potentially schedulable pods ASAP, eliminating a penalty effect of the backoff queue.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is the gate for all kube-components, or is it only relevant to the scheduler?

My guess is that this is only relevant to the scheduler.
If this option is only for the kube-scheduler, why do we make it a feature gate (for cross-cutting features involving more than one component)?
Can we simply use the kube-scheduler config file for this?
Feature gates are "heavy switchces".
With no explicit documentation on which gates involve which component, a cluster operator may have to restart all kubelets and even kube-proxy on all nodes.
/hold

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, it's a feature gate only for the kube-scheduler. There are also other feature flags for another enhancements that similarly are designed only for a kube-scheduler. I'm not sure if moving from a feature gate to the config file is a good idea given that we are in the enhancements and code freezes.

How can an explicit documentation look like?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

/hold cancel
This issue has been sent to sig-architecture for guidance. I'm not anticipating a decision before the deadline this time, so I'm lifting the hold.