From 4a3566a50d651a2edc03841d51d662987c13759b Mon Sep 17 00:00:00 2001
From: Chris Werner Rau <cwr@teuto.net>
Date: Tue, 9 Dec 2025 14:57:38 +0100
Subject: [PATCH] fix(base-cluster/monitoring)!: grafana-tempo-distributed
 would need s3

Therefore we switch to the singleBinary chart.
Without this change the compactor doesn't work; meaning the traces are
never cleaned up...
---
 charts/base-cluster/README.md.gotmpl          | 22 +++++++
 .../templates/monitoring/alloy-gateway.yaml   |  2 +-
 .../monitoring/tracing/grafana-tempo.yaml     | 60 +++++++++----------
 charts/base-cluster/values.schema.json        | 30 ++++------
 charts/base-cluster/values.yaml               | 11 ++--
 5 files changed, 70 insertions(+), 55 deletions(-)

diff --git a/charts/base-cluster/README.md.gotmpl b/charts/base-cluster/README.md.gotmpl
index 203328ca68..a866977d06 100644
--- a/charts/base-cluster/README.md.gotmpl
+++ b/charts/base-cluster/README.md.gotmpl
@@ -393,4 +393,26 @@ ingress:
   provider: nginx
 ```
 
+### 8.x.x -> 9.0.0
+
+This release adds the capability to configure alertmanager to push it's alerts
+to multiple receivers, including email.
+
+To migrate to this new version you have to remove `.monitoring.prometheus.alertmanager.receivers.pagerduty.enabled`
+
+### 9.x.x -> 10.0.0
+
+This release adds support for [k8up](https://k8up.io) as a backup provider.
+
+If you are using velero, you need to migrate to using `.backup.provider.velero`.
+
+### 10.x.x -> 11.0.0
+
+This release fixes the grafana tempo deployment compaction / retention.
+
+For this we switched from the distributed deployment to the single binary deployment.
+
+Because of this you need to migrate to the `.monitoring.tracing.<field>` instead
+of `.monitoring.tracing.ingester.<field>`
+
 {{ .Files.Get "values.md"  }}
diff --git a/charts/base-cluster/templates/monitoring/alloy-gateway.yaml b/charts/base-cluster/templates/monitoring/alloy-gateway.yaml
index 8571eb2af3..2b022c9ced 100644
--- a/charts/base-cluster/templates/monitoring/alloy-gateway.yaml
+++ b/charts/base-cluster/templates/monitoring/alloy-gateway.yaml
@@ -98,7 +98,7 @@ spec:
 
           otelcol.exporter.otlp "tempo" {
             client {
-              endpoint = "grafana-tempo-distributor:4317"
+              endpoint = "grafana-tempo:4317"
 
               tls {
                 insecure = true
diff --git a/charts/base-cluster/templates/monitoring/tracing/grafana-tempo.yaml b/charts/base-cluster/templates/monitoring/tracing/grafana-tempo.yaml
index 9715c16e5c..7264753c12 100644
--- a/charts/base-cluster/templates/monitoring/tracing/grafana-tempo.yaml
+++ b/charts/base-cluster/templates/monitoring/tracing/grafana-tempo.yaml
@@ -9,7 +9,7 @@ metadata:
     app.kubernetes.io/part-of: monitoring
 spec:
   chart:
-    spec: {{- include "base-cluster.helm.chartSpec" (dict "repo" "grafana" "chart" "tempo-distributed" "context" $) | nindent 6 }}
+    spec: {{- include "base-cluster.helm.chartSpec" (dict "repo" "grafana" "chart" "tempo" "context" $) | nindent 6 }}
   interval: 1h
   driftDetection:
     mode: enabled
@@ -17,43 +17,43 @@ spec:
     - name: kube-prometheus-stack
       namespace: monitoring
   values:
-    {{- if .Values.global.imageRegistry }}
-    global:
-      image:
-        registry: {{ $.Values.global.imageRegistry }}
-    {{- end }}
-    ingester: {{- include "common.resourcesWithPreset" .Values.monitoring.tracing.ingester | nindent 6 }}
-      persistentVolumeClaimRetentionPolicy:
-        enabled: true
-        whenDeleted: Delete
-        whenScaled: Retain
-      replicas: &replicas 1
-      config:
-        replication_factor: *replicas
-      persistence:
-        enabled: true
-        size: {{ .Values.monitoring.tracing.ingester.persistence.size }}
     tempo:
+      {{- with .Values.global.imageRegistry }}
+      registry: {{ . }}
+      {{- end }}
+      resources: {{- include "common.resources" .Values.monitoring.tracing.resources | nindent 8 }}
       securityContext:
         privileged: false
         seLinuxOptions: {}
         seccompProfile:
           type: RuntimeDefault
-      podSecurityContext:
-        fsGroupChangePolicy: OnRootMismatch
-        supplementalGroups: []
-        sysctls: []
+      receivers:
+        otlp:
+          protocols:
+            grpc:
+              endpoint: "0.0.0.0:4317"
+      reportingEnabled: false
+
+    persistentVolumeClaimRetentionPolicy:
+      enabled: true
+      whenDeleted: Delete
+      whenScaled: Retain
+    replicas: 1
+    retention: 48h
+    persistence:
+      enabled: true
+      size: {{ .Values.monitoring.tracing.persistence.size }}
+    securityContext:
+      fsGroupChangePolicy: OnRootMismatch
+      supplementalGroups: []
+      sysctls: []
     metricsGenerator:
       enabled: true
-    traces:
-      otlp:
-        grpc:
-          enabled: true
-    metaMonitoring:
-      serviceMonitor:
-        enabled: true
-        labels:
-          monitoring/provisioned-by: base-cluster
+      remoteWriteUrl: "http://kube-prometheus-stack-prometheus:9090/api/v1/write"
+    serviceMonitor:
+      enabled: true
+      additionalLabels:
+        monitoring/provisioned-by: base-cluster
 ---
 apiVersion: v1
 kind: ConfigMap
diff --git a/charts/base-cluster/values.schema.json b/charts/base-cluster/values.schema.json
index c1d4a30f70..a197ccfdc0 100644
--- a/charts/base-cluster/values.schema.json
+++ b/charts/base-cluster/values.schema.json
@@ -871,30 +871,24 @@
             "enabled": {
               "type": "boolean"
             },
-            "ingester": {
+            "resourcesPreset": {
+              "$ref": "#/$defs/resourcesPreset"
+            },
+            "resources": {
+              "$ref": "#/$defs/resourceRequirements"
+            },
+            "persistence": {
               "type": "object",
               "properties": {
-                "resourcesPreset": {
-                  "$ref": "#/$defs/resourcesPreset"
-                },
-                "resources": {
-                  "$ref": "#/$defs/resourceRequirements"
-                },
-                "persistence": {
-                  "type": "object",
-                  "properties": {
-                    "size": {
-                      "$ref": "#/$defs/quantity"
-                    }
-                  },
-                  "additionalProperties": false
+                "size": {
+                  "$ref": "#/$defs/quantity"
                 }
               },
               "additionalProperties": false
             }
-          }
-        },
-        "additionalProperties": false
+          },
+          "additionalProperties": false
+        }
       },
       "additionalProperties": false
     },
diff --git a/charts/base-cluster/values.yaml b/charts/base-cluster/values.yaml
index 784084baa9..99dbfb1de7 100644
--- a/charts/base-cluster/values.yaml
+++ b/charts/base-cluster/values.yaml
@@ -111,7 +111,7 @@ global:
       charts:
         loki: 6.46.0
         alloy: 1.5.0
-        tempo-distributed: 1.57.0
+        tempo: 1.24.1
       condition: "{{ and .Values.monitoring.prometheus.enabled (or .Values.monitoring.loki.enabled .Values.monitoring.tracing.enabled) }}"
     external-dns:
       url: https://kubernetes-sigs.github.io/external-dns
@@ -347,11 +347,10 @@ monitoring:
           effect: NoSchedule
   tracing:
     enabled: false
-    ingester:
-      resourcesPreset: small
-      resources: {}
-      persistence:
-        size: 10Gi
+    resourcesPreset: small
+    resources: {}
+    persistence:
+      size: 10Gi
 
 descheduler:
   enabled: true