Change chart directory structure

2026-01-19 11:11:33 +09:00
parent 113bcfdf5b
commit 1c0c62291f
5947 changed files with 4134 additions and 1 deletions
@@ -0,0 +1,53 @@
+# Rancher Monitoring CRD 버전 갱신 가이드
+
+- 해당 차트는 `rancher-monitoring`을 배포 전에 배포되어야 하는 crd 파일을 관리하는 차트이다.
+- 갱신 작업 시 `rancher-monitoring`에 대한 갱신 작업도 같이 진행되어야 한다.
+
+## 1. git 작업 환경 구성
+
+- 서비스 카탈로그 git 다운로드
+```
+$ git clone https://github.com/paasup/service-catalog.git
+```
+
+- 작업 브랜치로 체크아웃
+```
+$ git checkout -b update-monitoring/104.1.2+up57.0.3 
+```
+
+## 2. helm chart upgrade
+
+### 1) 파일 삭제
+
+- 수정 또는 추가한 파일을 제외한 나머지 파일을 삭제한다.
+  ``` sh
+  # chart 디렉토리로 이동
+  cd ~/service-catalog/charts/rancher-monitoring-crd
+  
+  # 파일 삭제 전 삭제할 파일 목록 확인
+  find . -mindepth 1 \( -name "CUSTOM-README.md" -o -name "BUILD-README.md"  -o -name "custom-values.yaml" \) -prune -o -print
+  
+  # 파일 삭제
+  find . -mindepth 1 \(-name "CUSTOM-README.md" -o -name "BUILD-README.md"  -o -name "custom-values.yaml" \) -prune -o -exec rm -rf {} +
+  ```
+
+### 2) 차트 다운로드
+
+- 해당 차트는 rancher에서 관리 중인 [charts](https://github.com/rancher/charts)에서 다운로드한다.
+- rancher/charts 클론
+  ``` sh
+  # 작업 디렉토리로 이동
+  $ cd ~/
+
+  # rancher 버전 별로 branch가 관리된다.
+  # rancher 버전에 맞추어 필요한 버전의 chart를 클론한다.
+  $ git clone -b release-v2.9 https://github.com/rancher/charts
+  ```
+- 파일 복사
+  ``` sh
+  $ cd service-catalog/charts/rancher-monitoring-crd/
+  $ cp -r ~/charts/charts/rancher-monitoring-crd/104.1.2+up57.0.3/* ./
+  ```
+## 3. git push 및 tag 추가
+- **rancher-monitoring에 대한 갱신 작업 진행 후 git push를 진행한다.**
+- **push 방법은 rancher-monitoring의 BUILD-README.md를 참조하여 진행한다.**
@@ -0,0 +1,26 @@
+
+# Rancher Monitoring CRD 배포
+
+## 1. 배포 방법
+
+- `rancher-monitoring`을 배포하기 전 해당 차트를 배포해야 한다.
+- 명령어
+
+  ``` sh
+  $ helm upgrade rancher-monitoring-crd --install \
+    --namespace cattle-monitoring-system --create-namespace \
+    ../rancher-monitoring-crd -f ./custom-values.yaml
+  ```
+
+
+## 2.custom-values.yaml 설명
+
+- custom-values.yaml에 정의된 값에 대한 설명이다.
+
+### 1) 오프라인 설정
+
+- private 환경 배포시 사용한다.
+
+| Name                   | 설명                                                         | 기본값 |
+| ---------------------- | ------------------------------------------------------------ | ------ |
+| `global.cattle.systemDefaultRegistry` | 오프라인 설치 시에 설정. <br />paasup 설치시에는 "paasup.io"로 설정 | `""`   |
@@ -0,0 +1,10 @@
+annotations:
+  catalog.cattle.io/certified: rancher
+  catalog.cattle.io/hidden: "true"
+  catalog.cattle.io/namespace: cattle-monitoring-system
+  catalog.cattle.io/release-name: rancher-monitoring-crd
+apiVersion: v2
+description: Installs the CRDs for rancher-monitoring.
+name: rancher-monitoring-crd
+type: application
+version: 104.1.2+up57.0.3
@@ -0,0 +1,24 @@
+# rancher-monitoring-crd
+A Rancher chart that installs the CRDs used by rancher-monitoring.
+
+## How does this chart work?
+
+This chart marshalls all of the CRD files placed in the `crd-manifest` directory into a ConfigMap that is installed onto a cluster alongside relevant RBAC (ServiceAccount, ClusterRoleBinding, ClusterRole, and PodSecurityPolicy).
+
+Once the relevant dependent resourcees are installed / upgraded / rolled back, this chart executes a post-install / post-upgrade / post-rollback Job that:
+- Patches any existing versions of the CRDs contained within the `crd-manifest` on the cluster to set `spec.preserveUnknownFields=false`; this step is required since, based on [Kubernetes docs](https://kubernetes.io/docs/tasks/extend-kubernetes/custom-resources/custom-resource-definitions/#field-pruning) and a [known workaround](https://github.com/kubernetes-sigs/controller-tools/issues/476#issuecomment-691519936), such CRDs cannot be upgraded normally from `apiextensions.k8s.io/v1beta1` to `apiextensions.k8s.io/v1`.
+- Runs a `kubectl apply` on the CRDs that are contained within the crd-manifest ConfigMap to upgrade CRDs in the cluster
+
+On an uninstall, this chart executes a separate post-delete Job that:
+- Patches any existing versions of the CRDs contained within `crd-manifest` on the cluster to set `metadata.finalizers=[]`
+- Runs a `kubectl delete` on the CRDs that are contained within the crd-manifest ConfigMap to clean up the CRDs from the cluster
+
+Note: If the relevant CRDs already existed in the cluster at the time of install, this chart will absorb ownership of the lifecycle of those CRDs; therefore, on a `helm uninstall`, those CRDs will also be removed from the cluster alongside this chart.
+
+## Why can't we just place the CRDs in the templates/ directory of the main chart?
+
+In Helm today, you cannot declare a CRD and declare a resource of that CRD's kind in templates/ without encountering a failure on render.
+
+## [Helm 3] Why can't we just place the CRDs in the crds/ directory of the main chart?
+
+The Helm 3 `crds/` directory only supports the installation of CRDs, but does not support the upgrade and removal of CRDs, unlike what this chart facilitiates.
@@ -0,0 +1,4 @@
+# global:
+#  cattle:
+#   # offline 배포 시 설정 
+#   systemDefaultRegistry: "paasup.io"
@@ -0,0 +1,30 @@
+# Rancher
+
+{{- define "system_default_registry" -}}
+{{- if .Values.global.cattle.systemDefaultRegistry -}}
+{{- printf "%s/" .Values.global.cattle.systemDefaultRegistry -}}
+{{- end -}}
+{{- end -}}
+
+# Windows Support
+
+{{/*
+Windows cluster will add default taint for linux nodes,
+add below linux tolerations to workloads could be scheduled to those linux nodes
+*/}}
+
+{{- define "linux-node-tolerations" -}}
+- key: "cattle.io/os"
+  value: "linux"
+  effect: "NoSchedule"
+  operator: "Equal"
+{{- end -}}
+
+{{- define "linux-node-selector" -}}
+{{- if semverCompare "<1.14-0" .Capabilities.KubeVersion.GitVersion -}}
+beta.kubernetes.io/os: linux
+{{- else -}}
+kubernetes.io/os: linux
+{{- end -}}
+{{- end -}}
+
@@ -0,0 +1,102 @@
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: {{ .Chart.Name }}-create
+  namespace: {{ .Release.Namespace }}
+  labels:
+    app: {{ .Chart.Name }}
+  annotations:
+    "helm.sh/hook": post-install, post-upgrade, post-rollback
+    "helm.sh/hook-delete-policy": before-hook-creation, hook-succeeded, hook-failed
+spec:
+  template:
+    metadata:
+      name: {{ .Chart.Name }}-create
+      labels:
+        app: {{ .Chart.Name }}
+    spec:
+      serviceAccountName: {{ .Chart.Name }}-manager
+      securityContext:
+        runAsNonRoot: false
+        runAsUser: 0
+      containers:
+        - name: create-crds
+          image: {{ template "system_default_registry" . }}{{ .Values.image.repository }}:{{ .Values.image.tag }}
+          imagePullPolicy: IfNotPresent
+          command:
+          - /bin/sh
+          - -c
+          - >
+            echo "Applying CRDs...";
+            mkdir -p /etc/crd;
+            base64 -d /etc/config/crd-manifest.tgz.b64 | tar -xzv -C /etc/crd;
+            kubectl replace -Rf /etc/crd || kubectl create -Rf /etc/crd;
+            echo "Done!"
+          volumeMounts:
+            - name: crd-manifest
+              readOnly: true
+              mountPath: /etc/config
+      restartPolicy: OnFailure
+      nodeSelector: {{ include "linux-node-selector" . | nindent 8 }}
+      {{- if .Values.nodeSelector }}
+      {{- toYaml .Values.nodeSelector | nindent 8 }}
+      {{- end }}
+      tolerations: {{ include "linux-node-tolerations" . | nindent 8 }}
+      {{- if .Values.tolerations }}
+      {{- toYaml .Values.tolerations | nindent 8 }}
+      {{- end }}
+      volumes:
+      - name: crd-manifest
+        configMap:
+          name: {{ .Chart.Name }}-manifest
+---
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: {{ .Chart.Name }}-delete
+  namespace: {{ .Release.Namespace }}
+  labels:
+    app: {{ .Chart.Name }}
+  annotations:
+    "helm.sh/hook": pre-delete
+    "helm.sh/hook-delete-policy": before-hook-creation, hook-succeeded, hook-failed
+spec:
+  template:
+    metadata:
+      name: {{ .Chart.Name }}-delete
+      labels:
+        app: {{ .Chart.Name }}
+    spec:
+      serviceAccountName: {{ .Chart.Name }}-manager
+      securityContext:
+        runAsNonRoot: false
+        runAsUser: 0
+      containers:
+        - name: delete-crds
+          image: {{ template "system_default_registry" . }}{{ .Values.image.repository }}:{{ .Values.image.tag }}
+          imagePullPolicy: IfNotPresent
+          command:
+          - /bin/sh
+          - -c
+          - >
+            echo "Deleting CRDs...";
+            mkdir -p /etc/crd;
+            base64 -d /etc/config/crd-manifest.tgz.b64 | tar -xzv -C /etc/crd;
+            kubectl delete --ignore-not-found=true -Rf /etc/crd;
+          volumeMounts:
+            - name: crd-manifest
+              readOnly: true
+              mountPath: /etc/config
+      restartPolicy: OnFailure
+      nodeSelector: {{ include "linux-node-selector" . | nindent 8 }}
+      {{- if .Values.nodeSelector }}
+      {{- toYaml .Values.nodeSelector | nindent 8 }}
+      {{- end }}
+      tolerations: {{ include "linux-node-tolerations" . | nindent 8 }}
+      {{- if .Values.tolerations }}
+      {{- toYaml .Values.tolerations | nindent 8 }}
+      {{- end }}
+      volumes:
+      - name: crd-manifest
+        configMap:
+          name: {{ .Chart.Name }}-manifest
@@ -0,0 +1,8 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: {{ .Chart.Name }}-manifest
+  namespace: {{ .Release.Namespace }}
+data:
+  crd-manifest.tgz.b64:
+  {{- .Files.Get "files/crd-manifest.tgz" | b64enc | indent 4 }}
@@ -0,0 +1,76 @@
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRole
+metadata:
+  name: {{ .Chart.Name }}-manager
+  labels:
+    app: {{ .Chart.Name }}-manager
+rules:
+- apiGroups:
+  - apiextensions.k8s.io
+  resources:
+  - customresourcedefinitions
+  verbs: ['create', 'get', 'patch', 'delete', 'update', 'list']
+{{- if .Values.global.cattle.psp.enabled }}
+- apiGroups: ['policy']
+  resources: ['podsecuritypolicies']
+  verbs:     ['use']
+  resourceNames:
+  - {{ .Chart.Name }}-manager
+{{- end }}
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRoleBinding
+metadata:
+  name: {{ .Chart.Name }}-manager
+  labels:
+    app: {{ .Chart.Name }}-manager
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: ClusterRole
+  name: {{ .Chart.Name }}-manager
+subjects:
+- kind: ServiceAccount
+  name: {{ .Chart.Name }}-manager
+  namespace: {{ .Release.Namespace }}
+---
+apiVersion: v1
+kind: ServiceAccount
+metadata:
+  name: {{ .Chart.Name }}-manager
+  namespace: {{ .Release.Namespace }}
+  labels:
+    app: {{ .Chart.Name }}-manager
+---
+{{- if .Values.global.cattle.psp.enabled }}
+apiVersion: policy/v1beta1
+kind: PodSecurityPolicy
+metadata:
+  name: {{ .Chart.Name }}-manager
+  namespace: {{ .Release.Namespace }}
+  labels:
+    app: {{ .Chart.Name }}-manager
+spec:
+  privileged: false
+  allowPrivilegeEscalation: false
+  hostNetwork: false
+  hostIPC: false
+  hostPID: false
+  runAsUser:
+    rule: 'RunAsAny'
+  seLinux:
+    rule: 'RunAsAny'
+  supplementalGroups:
+    rule: 'MustRunAs'
+    ranges:
+      - min: 1
+        max: 65535
+  fsGroup:
+    rule: 'MustRunAs'
+    ranges:
+      - min: 1
+        max: 65535
+  readOnlyRootFilesystem: false
+  volumes:
+    - 'configMap'
+    - 'secret'
+{{- end }}
@@ -0,0 +1,7 @@
+#{{- if gt (len (lookup "rbac.authorization.k8s.io/v1" "ClusterRole" "" "")) 0 -}}
+#{{- if .Values.global.cattle.psp.enabled }}
+#{{- if not (.Capabilities.APIVersions.Has "policy/v1beta1/PodSecurityPolicy") }}
+#{{- fail "The target cluster does not have the PodSecurityPolicy API resource. Please disable PSPs in this chart before proceeding." -}}
+#{{- end }}
+#{{- end }}
+#{{- end }}
@@ -0,0 +1,17 @@
+# Default values for rancher-monitoring-crd.
+# This is a YAML-formatted file.
+# Declare variables to be passed into your templates.
+
+global:
+  cattle:
+    psp:
+      enabled: false
+    systemDefaultRegistry: ""
+
+image:
+  repository: rancher/shell
+  tag: v0.2.1
+
+nodeSelector: {}
+
+tolerations: []