fix #1601 : enable cpu limit for "robusta-forwarder" service as current config cause cpu hogging

[rajpratik71]

fix #1601

Signed-off-by: Pratik Raj [email protected]

[coderabbitai]

Walkthrough

The forwarder Helm template now unconditionally sets the kubewatch container’s cpu limit from values. Default values set cpu limits for kubewatch and grafanaRenderer to 50m instead of null. Memory limit logic remains unchanged.

Changes

Cohort / File(s)	Summary
Forwarder Helm template helm/robusta/templates/forwarder.yaml	Changed templating to always emit resources.limits.cpu for kubewatch using .Values.kubewatch.resources.limits.cpu; memory limit logic unchanged.
Default values helm/robusta/values.yaml	Set kubewatch.resources.limits.cpu from ~ to 50m; set grafanaRenderer.resources.limits.cpu from ~ to 50m.

Sequence Diagram(s)

sequenceDiagram
  autonumber
  participant U as User/CI
  participant H as Helm
  participant T as forwarder.yaml (template)
  participant K as Kubernetes API
  participant S as Scheduler/Kubelet

  U->>H: helm install/upgrade
  H->>T: Render with values.yaml
  note over T: Unconditionally set kubewatch<br/>resources.limits.cpu = values.kubewatch.resources.limits.cpu
  T-->>H: Manifest (Deployment/Pod spec)
  H->>K: Apply manifest
  K->>S: Create/Run Pod
  S-->>S: Enforce CPU limit at runtime
  note over S: kubewatch container capped per configured CPU limit

Loading

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Pre-merge checks and finishing touches

❌ Failed checks (2 warnings)

Check name	Status	Explanation	Resolution
Linked Issues Check	⚠️ Warning	The changes add unconditional CPU limits for the kubewatch sidecar and update resource values for kubewatch and grafanaRenderer, but there is no evidence of setting a CPU limit for the primary robusta-forwarder container as required by issue #1601.	Update the forwarder template and values.yaml to include and honor CPU limit settings for the core robusta-forwarder container, ensuring the main forwarder process is constrained as outlined in issue #1601.
Out of Scope Changes Check	⚠️ Warning	The PR updates CPU limits for kubewatch and grafanaRenderer components, but only the kubewatch sidecar is part of the robusta-forwarder pods; the grafanaRenderer change is unrelated to limiting CPU for the forwarder service and falls outside the scope of issue #1601.	Remove or isolate the grafanaRenderer CPU limit change into a separate update and ensure any CPU limit modifications directly target the robusta-forwarder container and its sidecars as defined by the issue scope.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title Check	✅ Passed	The title clearly summarizes the primary change of enabling a CPU limit for the robusta-forwarder service to address CPU hogging, directly reflecting the main objective of the pull request.
Description Check	✅ Passed	The description references the relevant issue (#1601) and indicates the intent to fix CPU limit configuration, aligning with the changes made in the pull request.
Docstring Coverage	✅ Passed	No functions found in the changes. Docstring coverage check skipped.

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

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

• X
• Mastodon
• Reddit
• LinkedIn

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

[coderabbitai]

Actionable comments posted: 1

📜 Review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between b066ce1 and 9576d07.

📒 Files selected for processing (2)

• helm/robusta/templates/forwarder.yaml (1 hunks)
• helm/robusta/values.yaml (1 hunks)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: run_tests

[coderabbitai]

⚠️ Potential issue | 🔴 Critical

Restore conditional rendering for cpu limit

Emitting cpu: {{ .Values.kubewatch.resources.limits.cpu }} unconditionally breaks the common use case where operators set that value to null to remove the limit—Helm renders it as <no value>, yielding invalid YAML and failing the deploy. Please wrap the stanza with an if/with, as it was before, so the key is omitted when the value is unset.

Apply this diff to bring back the guard:

-          limits:
-            cpu: {{ .Values.kubewatch.resources.limits.cpu }}
+          limits:
+          {{- with .Values.kubewatch.resources.limits.cpu }}
+            cpu: {{ . }}
+          {{- end }}
             memory: {{ if .Values.isSmallCluster }}"64Mi"{{ else if .Values.kubewatch.resources.limits.memory }}{{ .Values.kubewatch.resources.limits.memory | quote }}{{ else }}{{ .Values.kubewatch.resources.requests.memory | quote }}{{ end }}

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	cpu: {{ .Values.kubewatch.resources.limits.cpu }}
	memory: {{ if .Values.isSmallCluster }}"64Mi"{{ else if .Values.kubewatch.resources.limits.memory }}{{ .Values.kubewatch.resources.limits.memory | quote }}{{ else }}{{ .Values.kubewatch.resources.requests.memory | quote }}{{ end }}
	limits:
	{{- with .Values.kubewatch.resources.limits.cpu }}
	cpu: {{ . }}
	{{- end }}
	memory: {{ if .Values.isSmallCluster }}"64Mi"{{ else if .Values.kubewatch.resources.limits.memory }}{{ .Values.kubewatch.resources.limits.memory | quote }}{{ else }}{{ .Values.kubewatch.resources.requests.memory | quote }}{{ end }}

🤖 Prompt for AI Agents

In helm/robusta/templates/forwarder.yaml around lines 74-75, the cpu limit key
is emitted unconditionally causing Helm to render "<no value>" when
.Values.kubewatch.resources.limits.cpu is null; wrap the cpu line in an if (or
with) guard that only renders the entire "cpu: ..." line when
.Values.kubewatch.resources.limits.cpu is set (preserve current indentation),
leaving the key out otherwise so the generated YAML remains valid.