SMOODEV-1965: smooai-next chart — optional hpa.memoryTarget for memory scale-out#9
Merged
Merged
Conversation
…y scale-out The chart's HPA hardcoded a single CPU Resource metric, so memory-pressured Next.js workloads (smooai-web) had no way to scale out — they rode individual pods into a V8 heap-limit OOM (SMOODEV-1964). Add an optional hpa.memoryTarget value that appends a second memory Resource metric (averageUtilization % of the memory request). Default 0 = CPU-only, fully backward-compatible (rendered output unchanged for existing consumers). Chart bumped 0.2.1 -> 0.2.2. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Why
Follow-up to SMOODEV-1964 (smooai-web OOM). The chart's
templates/hpa.yamlhardcoded a single CPU Resource metric, so a memory-pressured Next.js workload had no way to scale out — it rode individual pods into a V8 heap-limit OOM. The 1964 hotfix raised the per-pod memory ceiling; this gives the autoscaler a memory signal too.What
hpa.memoryTargetvalue. When set (e.g.80), the HPA gets a second memory Resource metric (averageUtilizationas a % of the memory request).0= CPU-only, fully backward-compatible — rendered output is byte-identical for existing consumers (verified viahelm template).The HPA scales up when either cpu or memory is hot, and down only when both are cool — so the consumer must set
memoryTargetcomfortably above steady-statemem-used ÷ mem-requestor it pins at max (documented invalues.yaml).Verification
helm lintpasses.helm templatewith default → CPU-only metric (unchanged).helm template --set hpa.memoryTarget=80→ both cpu (70) + memory (80) metrics.Next
After merge + chart publish (release
workflow_dispatch), smooai consumes0.2.2and setshpa.memoryTarget: 80ink8s/charts/web.ts(separate PR in the smooai monorepo).🤖 Generated with Claude Code