feat: add HPA (horizontal pod autoscaler) support to performance config by tangxiaoxin06 · Pull Request #21 · wellCh4n/oops

tangxiaoxin06 · 2026-04-15T08:48:19Z

Summary

Adds optional HPA support to per-environment performance config
Backend creates/updates/deletes autoscaling/v2 HPA targeting the app's StatefulSet
When autoscaling is enabled, OOPS stops setting spec.replicas to avoid fighting the HPA controller
UI exposes a checkbox + min/max replicas + CPU/memory target utilization inputs
Translations added for zh/en

Motivation

great fit for a PaaS. HPA is a K8s primitive that should be first-class.

Requirements

Target cluster must have metrics-server installed for HPA to compute utilization. Otherwise HPA stays in <unknown> state.

Test plan

Backend compiles on JDK 21
Deploy an app with autoscaling disabled — behavior identical to before
Enable autoscaling, set min=1 max=5 CPU=70, redeploy → kubectl get hpa -n <ns> shows the HPA
Disable autoscaling → HPA deleted
Frontend renders checkbox; replicas input hidden when autoscaling is on

Replicas were a static value, forcing operators to manually adjust them and redeploy under traffic spikes. This change adds an optional autoscaling block per-environment that drives a Kubernetes HPA (autoscaling/v2) targeting the application StatefulSet. Backend: - Add AutoscalingConfig (enabled/min/max/CPU/memory targets) on ApplicationPerformanceConfig.EnvironmentConfig (JSON blob, no schema migration needed). - In ArtifactDeployTask, skip writing spec.replicas when autoscaling is enabled to avoid fighting the HPA controller, and create/update or delete the HPA resource accordingly. Frontend: - Expose an "Autoscaling (HPA)" checkbox with min/max replicas and CPU/memory target utilization inputs in the performance tab; replicas input hides while autoscaling is on. - Add matching types, zod schema and zh/en translations. Requires metrics-server in the target cluster for HPA to function. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Copilot

Pull request overview

Adds per-environment Horizontal Pod Autoscaler (HPA) configuration support so apps can optionally be scaled by Kubernetes HPA instead of a fixed replica count.

Changes:

Extend performance env config model/types/schema to include an optional autoscaling block (enabled/min/max/CPU+memory targets).
Update the UI to toggle autoscaling and show autoscaling-specific inputs while hiding the replicas input when enabled.
Update backend deployment to create/update/delete an autoscaling/v2 HPA targeting the app’s StatefulSet and stop setting spec.replicas when autoscaling is enabled.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
web/locales/zh.ts	Adds new zh translations for autoscaling fields/hint.
web/locales/en.ts	Adds new en translations for autoscaling fields/hint.
web/lib/api/types.ts	Extends API types with `AutoscalingConfig`.
web/app/apps/schema.ts	Extends zod schema to validate autoscaling fields.
web/app/apps/components/application-performance-info.tsx	Adds autoscaling UI controls and conditional rendering of replicas input.
src/main/java/com/github/wellch4n/oops/task/ArtifactDeployTask.java	Creates/patches HPA when enabled; avoids setting StatefulSet replicas under HPA.
src/main/java/com/github/wellch4n/oops/data/ApplicationPerformanceConfig.java	Adds persisted autoscaling config fields to the env config model.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-15T10:02:53Z

+        var cfg = perfEnvConfig.getAutoscaling();
+        int minReplicas = cfg.getMinReplicas() == null ? 1 : cfg.getMinReplicas();
+        int maxReplicas = cfg.getMaxReplicas() == null ? Math.max(minReplicas, 10) : cfg.getMaxReplicas();
+        if (maxReplicas < minReplicas) {
+            maxReplicas = minReplicas;
+        }
+
+        List<io.fabric8.kubernetes.api.model.autoscaling.v2.MetricSpec> metrics = new ArrayList<>();
+        if (cfg.getTargetCpuUtilization() != null) {
+            metrics.add(buildResourceMetric("cpu", cfg.getTargetCpuUtilization()));
+        }
+        if (cfg.getTargetMemoryUtilization() != null) {
+            metrics.add(buildResourceMetric("memory", cfg.getTargetMemoryUtilization()));
+        }
+        if (metrics.isEmpty()) {
+            metrics.add(buildResourceMetric("cpu", 70));
+        }


minReplicas, maxReplicas, and target utilizations are taken directly from config without validation/clamping (other than max < min). If values are out of range (e.g. utilization <=0 or >100), the HPA patch will fail and break deploys. Consider validating these fields server-side (clamp to valid ranges or throw a descriptive exception) before building the HPA spec.

Copilot · 2026-04-15T10:02:53Z

+                      <Input
+                        type="number"
+                        min={1}
+                        placeholder="1"
+                        className="w-24"
+                        value={field.value ?? ""}
+                        onChange={e => field.onChange(e.target.value === "" ? undefined : parseInt(e.target.value, 10))}
+                      />


parseInt(e.target.value, 10) can produce NaN for certain valid <input type="number"> intermediate values (e.g. "-", "e"), which then gets stored in RHF state and may cause React warnings (value becomes NaN) and broken validation. Consider converting via valueAsNumber and mapping NaN to undefined (or validating/sanitizing the string before parsing).

Copilot · 2026-04-15T10:02:54Z

+                      <Input
+                        type="number"
+                        min={1}
+                        placeholder="10"
+                        className="w-24"
+                        value={field.value ?? ""}
+                        onChange={e => field.onChange(e.target.value === "" ? undefined : parseInt(e.target.value, 10))}
+                      />


Same NaN issue as above: parseInt(e.target.value, 10) on a type="number" input can yield NaN, which then becomes the controlled value and can trigger React warnings / invalid form state. Prefer valueAsNumber + Number.isNaN guard (or sanitize before parsing).

Copilot · 2026-04-15T10:02:54Z

+                          value={field.value ?? ""}
+                          onChange={e => field.onChange(e.target.value === "" ? undefined : parseInt(e.target.value, 10))}


Same NaN issue as above for the utilization input: parseInt can yield NaN for intermediate/invalid number strings, and then value={field.value ?? ""} will pass NaN through as a controlled value. Guard against NaN (e.g. valueAsNumber + isNaN -> undefined).

Suggested change

value={field.value ?? ""}

onChange={e => field.onChange(e.target.value === "" ? undefined : parseInt(e.target.value, 10))}

value={typeof field.value === "number" && !Number.isNaN(field.value) ? field.value : ""}

onChange={e => {

const value = e.target.valueAsNumber

field.onChange(Number.isNaN(value) ? undefined : value)

}}

Copilot · 2026-04-15T10:02:54Z

+                        <Input
+                          type="number"
+                          min={1}
+                          max={100}
+                          placeholder="—"
+                          className="pr-8"
+                          value={field.value ?? ""}
+                          onChange={e => field.onChange(e.target.value === "" ? undefined : parseInt(e.target.value, 10))}
+                        />


Same NaN issue as above for the utilization input: parseInt can return NaN, which is then stored in form state and fed back into the controlled input value. Consider handling NaN explicitly (map to undefined) to avoid warnings and validation problems.

Copilot · 2026-04-15T10:02:54Z

+      minReplicas: z.number().int().min(1).optional(),
+      maxReplicas: z.number().int().min(1).optional(),
+      targetCpuUtilization: z.number().int().min(1).max(100).optional(),
+      targetMemoryUtilization: z.number().int().min(1).max(100).optional(),


The schema allows autoscaling.minReplicas and autoscaling.maxReplicas independently, but doesn't enforce maxReplicas >= minReplicas. Since the backend silently clamps maxReplicas up to minReplicas, users can save an invalid combination without feedback. Consider adding a cross-field refinement so the UI can show a validation error when both are set and max < min.

Suggested change

targetMemoryUtilization: z.number().int().min(1).max(100).optional(),

targetMemoryUtilization: z.number().int().min(1).max(100).optional(),

}).superRefine((autoscaling, ctx) => {

if (

autoscaling.minReplicas !== undefined &&

autoscaling.maxReplicas !== undefined &&

autoscaling.maxReplicas < autoscaling.minReplicas

) {

ctx.addIssue({

code: z.ZodIssueCode.custom,

message: "Max replicas must be greater than or equal to min replicas",

path: ["maxReplicas"],

})

}

Copilot · 2026-04-15T10:02:55Z

+            } catch (Exception e) {
+                log.warn("Failed to delete HPA for {}/{}: {}", namespace, applicationName, e.getMessage());


When autoscaling is disabled, HPA deletion errors are swallowed for all exception types. If deletion fails for reasons other than NotFound (e.g. RBAC/network), the old HPA may remain and continue reconciling replicas, which can conflict with setting StatefulSet.spec.replicas. Consider only ignoring NotFound, and failing the deploy task (or surfacing a clear error) for other deletion failures.

Suggested change

} catch (Exception e) {

log.warn("Failed to delete HPA for {}/{}: {}", namespace, applicationName, e.getMessage());

} catch (io.fabric8.kubernetes.client.KubernetesClientException e) {

if (e.getCode() == 404) {

log.debug("HPA for {}/{} does not exist, nothing to delete", namespace, applicationName);

} else {

log.error("Failed to delete HPA for {}/{}:", namespace, applicationName, e);

throw e;

}

} catch (RuntimeException e) {

log.error("Failed to delete HPA for {}/{}:", namespace, applicationName, e);

throw e;

wellCh4n · 2026-04-15T17:00:34Z

Well done. However, as this was unplanned, I need to run some local tests before merging it into the codebase. It will take some time, but I'll get it done shortly.

wellCh4n requested a review from Copilot April 15, 2026 09:57

Copilot started reviewing on behalf of wellCh4n April 15, 2026 09:58 View session

Copilot AI reviewed Apr 15, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: add HPA (horizontal pod autoscaler) support to performance config #21

feat: add HPA (horizontal pod autoscaler) support to performance config #21
tangxiaoxin06 wants to merge 1 commit into
wellCh4n:mainfrom
tangxiaoxin06:feat/performance-hpa

tangxiaoxin06 commented Apr 15, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

wellCh4n commented Apr 15, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

		value={field.value ?? ""}
		onChange={e => field.onChange(e.target.value === "" ? undefined : parseInt(e.target.value, 10))}

-                          value={field.value ?? ""}
-                          onChange={e => field.onChange(e.target.value === "" ? undefined : parseInt(e.target.value, 10))}
+                          value={typeof field.value === "number" && !Number.isNaN(field.value) ? field.value : ""}
+                          onChange={e => {
+                            const value = e.target.valueAsNumber
+                            field.onChange(Number.isNaN(value) ? undefined : value)
+                          }}

-      targetMemoryUtilization: z.number().int().min(1).max(100).optional(),
+      targetMemoryUtilization: z.number().int().min(1).max(100).optional(),
+    }).superRefine((autoscaling, ctx) => {
+      if (
+        autoscaling.minReplicas !== undefined &&
+        autoscaling.maxReplicas !== undefined &&
+        autoscaling.maxReplicas < autoscaling.minReplicas
+      ) {
+        ctx.addIssue({
+          code: z.ZodIssueCode.custom,
+          message: "Max replicas must be greater than or equal to min replicas",
+          path: ["maxReplicas"],
+        })
+      }

		} catch (Exception e) {
		log.warn("Failed to delete HPA for {}/{}: {}", namespace, applicationName, e.getMessage());

-            } catch (Exception e) {
-                log.warn("Failed to delete HPA for {}/{}: {}", namespace, applicationName, e.getMessage());
+            } catch (io.fabric8.kubernetes.client.KubernetesClientException e) {
+                if (e.getCode() == 404) {
+                    log.debug("HPA for {}/{} does not exist, nothing to delete", namespace, applicationName);
+                } else {
+                    log.error("Failed to delete HPA for {}/{}:", namespace, applicationName, e);
+                    throw e;
+                }
+            } catch (RuntimeException e) {
+                log.error("Failed to delete HPA for {}/{}:", namespace, applicationName, e);
+                throw e;

Conversation

tangxiaoxin06 commented Apr 15, 2026

Summary

Motivation

Requirements

Test plan

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

wellCh4n commented Apr 15, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants