allow optimizing guaranteed pods #83

maanas-23 · 2026-02-03T12:28:48Z

Summary by CodeRabbit

New Features
- Added optimizeGuaranteedPods configuration option to control whether guaranteed pods are eligible for optimization.
- CPU optimization for guaranteed pods is now supported when enabled via configuration.
Improvements
- Simplified memory request and limit optimization logic.

coderabbitai · 2026-02-03T12:29:07Z

Walkthrough

A new configuration option optimizeGuaranteedPods is introduced to control whether guaranteed pods should be optimized. The config field is defined in both the YAML file and Go struct, and the pod recommendation logic is refactored to respect this flag while simplifying memory handling and enabling CPU optimization for guaranteed pods.

Changes

Cohort / File(s)	Summary
Configuration Fields `config.yaml`, `pkg/config/config.go`	Added new boolean field `optimizeGuaranteedPods` to enable/disable optimization of guaranteed pods via configuration.
Pod Optimization Logic `pkg/task/taskApplyRecommendation.go`	Refactored guaranteed pod handling to check configuration flag instead of unconditionally skipping; removed memory equality synchronization branch; removed early return for guaranteed pod CPU changes, allowing CPU updates to proceed.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Poem

🐰 A carrot-shaped config flag now appears,
To guide which pods we optimize with care,
No more guaranteed pods left in the shadows,
Memory logic simplified, CPU runs free,
Configuration blooms—flexibility takes the stage! 🌱

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'allow optimizing guaranteed pods' directly and clearly summarizes the main change—enabling optimization of guaranteed pods via a new configuration flag.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

📝 Generate docstrings

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment
Commit unit tests in branch optimize-guaranteed

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

coderabbitai

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@pkg/task/taskApplyRecommendation.go`:
- Around line 318-321: The memory-clamping logic in applyMemoryRecommendation()
can lower a guaranteed container's request while its limit remains unchanged,
converting it to Burstable QoS on clusters that can't reduce limits; modify
applyMemoryRecommendation() to first detect guaranteed containers by checking if
OptimizeGuaranteedPods is true and currentMemoryLimit == currentMemoryRequest
(or request==limit) and skip any memory update (i.e., do not adjust
recommendedMemoryLimit/request) for those containers until limit-reduction
support exists, leaving the existing Math.Ceil clamp only for non-guaranteed
containers.

coderabbitai · 2026-02-03T12:37:56Z

pkg/task/taskApplyRecommendation.go

+	// TODO: will be possible from 1.34
 	if recommendedMemoryLimit < currentMemoryLimit {
 		recommendedMemoryLimit = math.Ceil(currentMemoryLimit)
 	}


⚠️ Potential issue | 🟠 Major

🧩 Analysis chain

🏁 Script executed:

wc -l pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 105

🏁 Script executed:

sed -n '310,330p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 1001

🏁 Script executed:

sed -n '300,340p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 1764

🏁 Script executed:

sed -n '450,475p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 1129

🏁 Script executed:

sed -n '280,350p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 3206

🏁 Script executed:

grep -n "func.*Apply\|OptimizeGuaranteedPods\|IsGuaranteedPod" pkg/task/taskApplyRecommendation.go | head -20

Repository: truefoundry/CruiseKube

Length of output: 1414

🏁 Script executed:

sed -n '1,50p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 1720

🏁 Script executed:

sed -n '153,210p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 2607

🏁 Script executed:

grep -n "applyMemoryRecommendation\|applyCPURecommendation" pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 448

🏁 Script executed:

sed -n '195,250p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 2209

🏁 Script executed:

sed -n '433,480p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 2272

🏁 Script executed:

grep -n "IsGuaranteedPod" pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 161

🏁 Script executed:

rg "IsGuaranteedPod" --context 3

Repository: truefoundry/CruiseKube

Length of output: 1075

🏁 Script executed:

rg "type PodInfo" pkg/task/utils/ -A 15

Repository: truefoundry/CruiseKube

Length of output: 1331

🏁 Script executed:

sed -n '314,342p' pkg/task/utils/node_stats.go

Repository: truefoundry/CruiseKube

Length of output: 48

🏁 Script executed:

grep -B 5 -A 10 "func.*IsGuaranteedPod" pkg/task/utils/node_stats.go

Repository: truefoundry/CruiseKube

Length of output: 560

🏁 Script executed:

sed -n '286,342p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 2557

🏁 Script executed:

sed -n '220,270p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 2383

🏁 Script executed:

grep -n "OptimizeGuaranteedPods" pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 161

🏁 Script executed:

rg "OptimizeGuaranteedPods" --context 2

Repository: truefoundry/CruiseKube

Length of output: 930

🏁 Script executed:

sed -n '354,420p' pkg/task/taskApplyRecommendation.go

Repository: truefoundry/CruiseKube

Length of output: 2824

🏁 Script executed:

grep -B 5 -A 15 "func.*applyCPURecommendation" pkg/task/taskApplyRecommendation.go | head -40

Repository: truefoundry/CruiseKube

Length of output: 783

Prevent QoS degradation for guaranteed containers when memory limits cannot be reduced.

When OptimizeGuaranteedPods is true, guaranteed containers (where request==limit) can reach applyMemoryRecommendation() and the memory clamping logic at lines 318-321 may allow the request to be lowered while the limit remains unchanged. This converts the container to Burstable QoS on clusters that don't yet support limit reduction. To ensure guaranteed containers stay protected, skip memory updates for containers where the current limit equals the current request until limit updates are supported.

Suggested guard

// TODO: will be possible from 1.34 if recommendedMemoryLimit < currentMemoryLimit { + if currentMemoryLimit == currentMemoryRequest { + logging.Infof(ctx, "Skipping memory recommendation for guaranteed pod %s/%s; limit cannot be decreased yet", rec.PodInfo.Namespace, rec.PodInfo.Name) + return false, true, nil + } recommendedMemoryLimit = math.Ceil(currentMemoryLimit) }

🤖 Prompt for AI Agents

In `@pkg/task/taskApplyRecommendation.go` around lines 318 - 321, The memory-clamping logic in applyMemoryRecommendation() can lower a guaranteed container's request while its limit remains unchanged, converting it to Burstable QoS on clusters that can't reduce limits; modify applyMemoryRecommendation() to first detect guaranteed containers by checking if OptimizeGuaranteedPods is true and currentMemoryLimit == currentMemoryRequest (or request==limit) and skip any memory update (i.e., do not adjust recommendedMemoryLimit/request) for those containers until limit-reduction support exists, leaving the existing Math.Ceil clamp only for non-guaranteed containers.

shubhamrai1993 · 2026-02-04T05:11:20Z

pkg/task/taskApplyRecommendation.go

 	currentMemoryLimitQuantity := currentContainerResources.Limits[corev1.ResourceMemory]
 	currentMemoryLimit := float64(currentMemoryLimitQuantity.Value()) / utils.BytesToMBDivisor

-	if currentMemoryRequest == currentMemoryLimit {


we need to mark a pod that has either equal request and limit for memory or cpu for eviction like here -

CruiseKube/pkg/task/applystrategies/adjust_amongst_pods_distributed_strategy.go

Line 110 in e93bb1f

sort.Slice(podInfosClone, func(i, j int) bool {

so that it can be evicted and recreated with corrected request and limit

maanas-23 and others added 2 commits February 3, 2026 17:58

allow optimizing guaranteed pods

b0d85ae

Merge branch 'main' into optimize-guaranteed

6d1cf74

coderabbitai bot reviewed Feb 3, 2026

View reviewed changes

shubhamrai1993 reviewed Feb 4, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

allow optimizing guaranteed pods #83

allow optimizing guaranteed pods #83

maanas-23 commented Feb 3, 2026 •

edited by coderabbitai bot

Loading

Uh oh!

coderabbitai bot commented Feb 3, 2026 •

edited

Loading

Uh oh!

coderabbitai bot left a comment

Uh oh!

coderabbitai bot Feb 3, 2026

Uh oh!

shubhamrai1993 Feb 4, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

allow optimizing guaranteed pods #83

Are you sure you want to change the base?

allow optimizing guaranteed pods #83

Conversation

maanas-23 commented Feb 3, 2026 • edited by coderabbitai bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Uh oh!

coderabbitai bot commented Feb 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Estimated code review effort

Poem

Uh oh!

coderabbitai bot left a comment

Choose a reason for hiding this comment

Uh oh!

coderabbitai bot Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

shubhamrai1993 Feb 4, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

maanas-23 commented Feb 3, 2026 •

edited by coderabbitai bot

Loading

coderabbitai bot commented Feb 3, 2026 •

edited

Loading