@leverege/build-tools 2.109.0 → 2.110.1

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@leverege/build-tools",
-  "version": "2.109.0",
+  "version": "2.110.1",
   "description": "A collection of build / support tools for Leverege developers",
   "main": "index.js",
   "repository": {

package/src/helm-charts/prom-operator/helmup.plugin

@@ -8,7 +8,7 @@ showInstalling "The Prometheus Operator (kube-prometheus-stack)"
 
 # https://artifacthub.io/packages/helm/prometheus-community/kube-prometheus-stack
 OCI_CHART="oci://ghcr.io/prometheus-community/charts/kube-prometheus-stack"
-[ -z "$PROMETHEUS_STACK_CHART_VERSION" ] && PROMETHEUS_STACK_CHART_VERSION="85.3.3"
+[ -z "$PROMETHEUS_STACK_CHART_VERSION" ] && PROMETHEUS_STACK_CHART_VERSION="86.2.3"
 helm upgrade $NS --install prometheus-stack $OCI_CHART \
                  --values  prom-operator/prometheus-stack.yaml \
                  --version $PROMETHEUS_STACK_CHART_VERSION $HELM_WHAT

package/src/helm-charts/prom-operator/prometheus-stack.yaml.ovh

@@ -342,7 +342,7 @@ alertmanager:
           - alertname="GKEMaintenanceInProgress"
           - maintenance="true"
         target_matchers:
-          - alertname=~"KubernetesNodeNotReady|KubernetesPodNotHealthy|KubernetesPodCrashLooping|frequent_app_restarts"
+          - alertname=~"KubernetesNodeNotReady|KubernetesPodNotHealthy|KubernetesPodCrashLooping|frequent_app_restarts|KubernetesStatefulsetDown|KubernetesDaemonsetMisscheduled"
 
     route:
       receiver: slack # default for everything

package/src/helm-charts/prom-operator/rules/cnpg-rules.yaml

@@ -68,7 +68,7 @@ spec:
         summary: Checks the number of database conflicts
       expr: |-
         rate(cnpg_pg_stat_database_deadlocks[5m]) > 0
-      for: 1m
+      for: 3m
       labels:
         severity: warning
     - alert: ReplicaFailingReplication
@@ -77,7 +77,7 @@ spec:
         summary: Checks if the replica is failing to replicate
       expr: |-
         cnpg_pg_replication_in_recovery > cnpg_pg_replication_is_wal_receiver_up
-      for: 1m
+      for: 3m
       labels:
         severity: warning
 #
@@ -85,7 +85,7 @@ spec:
 #
     - alert: PostgresqlDown
       expr: cnpg_collector_up == 0
-      for: 1m
+      for: 3m
       labels:
         severity: critical
       annotations:
@@ -94,9 +94,9 @@ spec:
 
     - alert: PostgresqlRestarted
       expr: time() - cnpg_pg_postmaster_start_time < 60
-      for: 2m
+      for: 0m
       labels:
-        severity: critical
+        severity: warning
       annotations:
         summary: Postgresql restarted (instance {{ $labels.instance }})
         description: "Postgresql restarted\n  VALUE = {{ $value }}\n  LABELS = {{ $labels }}"

package/src/helm-charts/prom-operator/rules/elasticsearch-rules.yaml

@@ -76,7 +76,7 @@ spec:
                 LABELS = {{ $labels }}
         - alert: ElasticsearchHealthyNodes
           expr: elasticsearch_cluster_health_number_of_nodes < 3
-          for: 0m
+          for: 5m
           labels:
             severity: critical
           annotations:
@@ -87,7 +87,7 @@ spec:
                 LABELS = {{ $labels }}
         - alert: ElasticsearchHealthyDataNodes
           expr: elasticsearch_cluster_health_number_of_data_nodes < 3
-          for: 0m
+          for: 5m
           labels:
             severity: critical
           annotations:
@@ -142,7 +142,7 @@ spec:
                 LABELS = {{ $labels }}
         - alert: ElasticsearchUnassignedShards
           expr: elasticsearch_cluster_health_unassigned_shards > 0
-          for: 0m
+          for: 5m
           labels:
             severity: critical
           annotations:

package/src/helm-charts/prom-operator/rules/kubestate-rules.yaml

@@ -154,7 +154,7 @@ spec:
                 LABELS = {{ $labels }}
         - alert: KubernetesStatefulsetDown
           expr: kube_statefulset_replicas != kube_statefulset_status_replicas_ready > 0
-          for: 3m # awesome was 1m
+          for: 5m # awesome was 1m; bumped to 3m, then 5m for GKE rolling updates
           labels:
             severity: critical
           annotations:
@@ -228,8 +228,11 @@ spec:
                 VALUE = {{ $value }}
                 LABELS = {{ $labels }}
         - alert: KubernetesReplicasetReplicasMismatch
-          expr: kube_replicaset_spec_replicas != kube_replicaset_status_ready_replicas
-          for: 15m # awesome was => 10m
+          expr: |
+            (kube_replicaset_spec_replicas != kube_replicaset_status_ready_replicas)
+            * on (namespace, replicaset) group_left()
+            kube_replicaset_owner{owner_kind!="Deployment"}
+          for: 15m # awesome was => 10m; Deployment-owned RSes excluded (covered by KubernetesDeploymentReplicasMismatch)
           labels:
             severity: warning
           annotations:
@@ -306,7 +309,7 @@ spec:
                 LABELS = {{ $labels }}
         - alert: KubernetesDaemonsetMisscheduled
           expr: kube_daemonset_status_number_misscheduled > 0
-          for: 1m
+          for: 5m
           labels:
             severity: critical
           annotations:
@@ -339,7 +342,7 @@ spec:
                 LABELS = {{ $labels }}
         - alert: KubernetesApiServerErrors
           expr: sum(rate(apiserver_request_total{job="apiserver",code=~"(?:5..)"}[1m])) by (instance, job) / sum(rate(apiserver_request_total{job="apiserver"}[1m])) by (instance, job) * 100 > 3
-          for: 2m
+          for: 5m
           labels:
             severity: critical
           annotations:

package/src/helm-charts/prom-operator/rules/redis-rules.yaml

@@ -10,7 +10,7 @@ spec:
       rules:
         - alert: RedisDown
           expr: redis_up == 0
-          for: 0m
+          for: 2m
           labels:
             severity: critical
           annotations:
@@ -21,7 +21,7 @@ spec:
                 LABELS = {{ $labels }}
         - alert: RedisMissingMaster
           expr: (count(redis_instance_info{role="master"}) or vector(0)) < 1
-          for: 0m
+          for: 2m
           labels:
             severity: critical
           annotations:
@@ -43,7 +43,7 @@ spec:
                 LABELS = {{ $labels }}
         - alert: RedisDisconnectedSlaves
           expr: count without (instance, job) (redis_connected_slaves) - sum without (instance, job) (redis_connected_slaves) - 1 > 0
-          for: 0m
+          for: 2m
           labels:
             severity: critical
           annotations:
@@ -54,7 +54,7 @@ spec:
                 LABELS = {{ $labels }}
         - alert: RedisReplicationBroken
           expr: delta(redis_connected_slaves[1m]) < 0
-          for: 0m
+          for: 2m
           labels:
             severity: critical
           annotations:

package/src/helm-charts/prom-operator/rules/traefik-rules.yaml

@@ -10,7 +10,7 @@ spec:
       rules:
         - alert: TraefikServiceDown
           expr: count(traefik_service_server_up) by (service) == 0
-          for: 0m
+          for: 2m
           labels:
             severity: critical
           annotations:

package/src/helm-charts/valkey/helmup.bootstrap

@@ -0,0 +1,12 @@
+#!/bin/bash
+#
+# The network policy sets:
+#
+#   ingressNSMatchLabels: # this is why default ns gets labelled
+#     redis: external
+# 
+# which should be specifically set in redis client service helm charts but
+# for now we set it hear without being forced to upgraed al of the legacy
+# helm charts. This setting will eventually end up in the leverege base chart.
+# 
+kubectl label namespace default redis=external

package/src/helm-charts/valkey/valkey-local.yaml

@@ -14,6 +14,14 @@ auth:
   enabled: false
   sentinel: true
 
+networkPolicy:
+  enabled: true
+  allowExternal: false
+  ingressNSMatchLabels: # this is why default ns gets labelled
+    redis: external
+  ingressNSPodMatchLabels:
+    redis-client: true
+
 primary:
   replicaCount: 1
 

package/src/helm-charts/velero/helmup.bootstrap

@@ -11,7 +11,7 @@ function installVeleroEnvironment() {
   gcloud config set project $GCP_PROJECT_ID
 
   ## The major chart version will determine the bucket suffix
-  [ -z "$VELERO_CHART_VERSION" ] && VELERO_CHART_VERSION="11"
+  [ -z "$VELERO_CHART_VERSION" ] && VELERO_CHART_VERSION="12"
 
   ## Create a bucket
   BUCKET="$GCP_PROJECT_ID-velero-$VELERO_CHART_VERSION"

package/src/helm-charts/velero/helmup.plugin

@@ -5,8 +5,8 @@ showInstalling "Velero Backup System"
 # We use a specific version for the velero chart because we need to control
 # the version of the locally installed velero client application.
 #
-VELERO_CHART_VERSION="11.4.0"
-EXPECTED_CLIENT_VERSION="v1.18.0"
+VELERO_CHART_VERSION="12.0.3"
+EXPECTED_CLIENT_VERSION="v1.18.1"
 
 addHelmRepo vmware-tanzu https://vmware-tanzu.github.io/helm-charts
 

package/src/init-my-chart/Chart.yaml.hbs

@@ -6,5 +6,5 @@ apiVersion: v2
 icon: https://raw.githubusercontent.com/kubernetes/kubernetes/master/logo/logo.svg
 dependencies:
   - name: leverege
-    version: "^1.2.0"
+    version: "^1.3.0"
     repository: "oci://us-docker.pkg.dev/leverege-registry/base-charts"

package/src/init-my-chart/leaf-chart-templates/templates/loadbalancerservice.yaml

@@ -0,0 +1 @@
+{{ include "leverege.loadbalancerservice" . }}

package/src/init-my-chart/leaf-chart-templates/values.schema.json

@@ -204,8 +204,32 @@
       "required": ["enabled"],
       "properties": {
         "enabled": { "type": "boolean" },
-        "livenessProbe": { "type": "object", "additionalProperties": true },
-        "readinessProbe": { "type": "object", "additionalProperties": true }
+        "livenessProbe": {
+          "type": "object",
+          "additionalProperties": true,
+          "description": "Pod: Liveness probe override. Fields are merged with base defaults — only specify fields you want to change.",
+          "properties": {
+            "httpGet": { "type": "object", "additionalProperties": true },
+            "initialDelaySeconds": { "type": "integer", "minimum": 0 },
+            "periodSeconds": { "type": "integer", "minimum": 1 },
+            "timeoutSeconds": { "type": "integer", "minimum": 1 },
+            "failureThreshold": { "type": "integer", "minimum": 1 },
+            "successThreshold": { "type": "integer", "minimum": 1 }
+          }
+        },
+        "readinessProbe": {
+          "type": "object",
+          "additionalProperties": true,
+          "description": "Pod: Readiness probe override. Fields are merged with base defaults — only specify fields you want to change.",
+          "properties": {
+            "httpGet": { "type": "object", "additionalProperties": true },
+            "initialDelaySeconds": { "type": "integer", "minimum": 0 },
+            "periodSeconds": { "type": "integer", "minimum": 1 },
+            "timeoutSeconds": { "type": "integer", "minimum": 1 },
+            "failureThreshold": { "type": "integer", "minimum": 1 },
+            "successThreshold": { "type": "integer", "minimum": 1 }
+          }
+        }
       }
     },
 

package/src/init-my-chart/values.yaml.hbs

@@ -35,8 +35,9 @@ fullnameOverride: ""
 partOf: {{hbsPartOf}} # leverege-stack or project-x - OPTIONAL
 
 # overwhelm will auto inject project_id, host and ingress into service
-# When type is LoadBalancer, metrics are stripped from the main service and a
-# companion <fullname>-internal ClusterIP service is rendered automatically.
+# When type is LoadBalancer, the main service remains ClusterIP (stable internal
+# name, http + metrics) and a companion <fullname>-loadbalancer service is rendered
+# to handle external traffic (http only, no metrics exposure).
 service:
   project_id: set-in-values-yaml
   preemptible: {{hbsServicePreemptible}} # true or false - OPTIONAL
@@ -58,8 +59,19 @@ serviceMonitor:
   namespace: "default" # namespace where service to scrape resides
 
 # liveness and readiness probe to prove k8s health
+# Individual fields are merged with base defaults — only specify what you need to change.
+# Base defaults: liveness => path=/livenessProbe, initialDelaySeconds=30, periodSeconds=60
+#                readiness => path=/healthz,       initialDelaySeconds=30, periodSeconds=10
+# Unset fields fall back to Kubernetes defaults (timeoutSeconds: 1, failureThreshold: 3,
+# successThreshold: 1). See https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/
 probes:
   enabled: true
+  # livenessProbe:
+  #   timeoutSeconds: 10    # k8s default: 1
+  #   failureThreshold: 5   # k8s default: 3
+  # readinessProbe:
+  #   timeoutSeconds: 5     # k8s default: 1
+  #   failureThreshold: 5   # k8s default: 3
 
 # this tells calico that redis access is needed
 redis:

package/src/init-my-chart/leaf-chart-templates/templates/internalservice.yaml

@@ -1 +0,0 @@
-{{ include "leverege.internalservice" . }}