From 1e6eda86334344c0b846e37a49904c4afc6de055 Mon Sep 17 00:00:00 2001
From: jenkins <brad@bstein.dev>
Date: Sun, 23 Aug 2026 10:41:01 -0300
Subject: [PATCH] fix: converge load-spread rollouts

---
 infrastructure/core/coredns-deployment.yaml |  2 ++
 services/monitoring/helmrelease.yaml        | 11 ++++-------
 2 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/infrastructure/core/coredns-deployment.yaml b/infrastructure/core/coredns-deployment.yaml
index 0e8dafdb..586bd36d 100644
--- a/infrastructure/core/coredns-deployment.yaml
+++ b/infrastructure/core/coredns-deployment.yaml
@@ -95,6 +95,8 @@ spec:
                     operator: In
                     values:
                       - "true"
+                  - key: node-role.kubernetes.io/storage-backbone
+                    operator: DoesNotExist
           preferredDuringSchedulingIgnoredDuringExecution:
             - weight: 100
               preference:
diff --git a/services/monitoring/helmrelease.yaml b/services/monitoring/helmrelease.yaml
index 760b4da8..e77b7afe 100644
--- a/services/monitoring/helmrelease.yaml
+++ b/services/monitoring/helmrelease.yaml
@@ -70,7 +70,6 @@ metadata:
 spec:
   interval: 15m
   upgrade:
-    force: true
     disableWait: true
   chart:
     spec:
@@ -100,12 +99,10 @@ spec:
 
       persistentVolume:
         enabled: true
-        # A year of Atlas cardinality is ~210Gi today, and this 100Gi never
-        # matched the 200Gi volume actually bound in the cluster. That volume
-        # filled on 2026-08-18: VictoriaMetrics went read-only, ingestion
-        # stopped, and every public dashboard panel rendered "No data" while
-        # all 119 scrape targets stayed healthy.
-        size: 400Gi
+        # The claim has already been expanded to 400Gi. Keep the original
+        # StatefulSet template size here because Kubernetes forbids changing
+        # volumeClaimTemplates; the bound PVC retains its larger capacity.
+        size: 100Gi
       resources:
         requests:
           cpu: 500m