From 5d059786a132af407c0392417c8079d50482f2ee Mon Sep 17 00:00:00 2001
From: Viktor Barzin <vbarzin@gmail.com>
Date: Mon, 29 Jun 2026 14:46:58 +0000
Subject: [PATCH] rightsize: trim over-provisioned memory limits+requests
 (batch 1/N)

claude-breakglass 4Gi->512Mi, stirling-pdf 1536->512Mi, insta2spotify 2Gi->256Mi, recruiter-responder 768->256Mi. These idle/utility services had memory LIMITS sitting 4-15x above their 30d peak, inflating cluster limit-overcommit to 142% across the 5 post-node6 nodes. Burstable (request<limit), limits capped at ~peak x1.5 (never below peak), so no OOM risk (verified zero OOMKills cluster-wide in 30d). Reduces phantom limit overcommit + frees scheduler requests.

Follows the 3-reviewer adversarial review: raising limits on an already-overcommitted cluster worsens correlated node-OOM; the real fix is trimming the fat. Limits only lowered where peak is far below; tuned/DB/GPU limits untouched.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---
 stacks/claude-breakglass/main.tf   | 4 ++--
 stacks/insta2spotify/main.tf       | 4 ++--
 stacks/recruiter-responder/main.tf | 4 ++--
 stacks/stirling-pdf/main.tf        | 4 ++--
 4 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/stacks/claude-breakglass/main.tf b/stacks/claude-breakglass/main.tf
index ca700945..447d5fc8 100644
--- a/stacks/claude-breakglass/main.tf
+++ b/stacks/claude-breakglass/main.tf
@@ -284,10 +284,10 @@ resource "kubernetes_deployment" "breakglass" {
           resources {
             requests = {
               cpu    = "200m"
-              memory = "512Mi"
+              memory = "320Mi"
             }
             limits = {
-              memory = "4Gi"
+              memory = "512Mi"
             }
           }
         }
diff --git a/stacks/insta2spotify/main.tf b/stacks/insta2spotify/main.tf
index 5e1cc4ef..e6f2e538 100644
--- a/stacks/insta2spotify/main.tf
+++ b/stacks/insta2spotify/main.tf
@@ -178,11 +178,11 @@ resource "kubernetes_deployment" "insta2spotify" {
           }
           resources {
             limits = {
-              memory = "2Gi"
+              memory = "256Mi"
             }
             requests = {
               cpu    = "50m"
-              memory = "512Mi"
+              memory = "160Mi"
             }
           }
         }
diff --git a/stacks/recruiter-responder/main.tf b/stacks/recruiter-responder/main.tf
index c948992b..f9ea670b 100644
--- a/stacks/recruiter-responder/main.tf
+++ b/stacks/recruiter-responder/main.tf
@@ -298,8 +298,8 @@ resource "kubernetes_deployment" "recruiter_responder" {
           }
 
           resources {
-            requests = { cpu = "100m", memory = "384Mi" }
-            limits   = { memory = "768Mi" }
+            requests = { cpu = "100m", memory = "192Mi" }
+            limits   = { memory = "256Mi" }
           }
         }
       }
diff --git a/stacks/stirling-pdf/main.tf b/stacks/stirling-pdf/main.tf
index 506d2ce9..6d9395de 100644
--- a/stacks/stirling-pdf/main.tf
+++ b/stacks/stirling-pdf/main.tf
@@ -87,10 +87,10 @@ resource "kubernetes_deployment" "stirling-pdf" {
           resources {
             requests = {
               cpu    = "25m"
-              memory = "1536Mi"
+              memory = "320Mi"
             }
             limits = {
-              memory = "1536Mi"
+              memory = "512Mi"
             }
           }