[juno-node] Feat/upload juno data file (#130)

* feat: add chart to backup juno data and upload to cloudflare r2 * feat: add chart to backup juno data and upload to cloudflare r2 - add new line at the end of file * feat: add chart to backup juno data and upload to cloudflare r2 - change readme and change typo. * feat: add chart to backup juno data and upload to cloudflare r2 - auto generate README.md and helm-docs v1.13.1 changed. * Fix: change dynamic name * Fix: enhance to replace pod with job. * Fix: remove tail space * Fix: add space * Fix: add space to correct syntax error * Fix: change rclone's secret from local file to secret managers * Fix: change schedule * Fix: add blank line and add externalsecret-common.yaml * Fix: add relative path ofdataFromKey * Fix: add relative path ofdataFromKey * Fix: change DB size * Revert "Fix: change DB size" This reverts commit 9e58710. * Fix: change DB size -v * Fix: test purpose to exclude sst files * Fix: test purpose to exclude sst files * Fix: test purpose to exclude sst files * Update externalsecret-common.yaml Signed-off-by: PhilexWong <[email protected]> * Update externalsecret-common.yaml Signed-off-by: PhilexWong <[email protected]> * Update externalsecret-common.yaml Signed-off-by: PhilexWong <[email protected]> * Fix: test purpose to exclude sst files - revert * Fix: rename the jar file * Fix: rename the jar file -exclude sst file for testing * feat: add retention function * feat: add retention function * feat: add retention function - add values * feat: add retention function * feat: add retention function * Remove -z param during tar package * Fix: Remove -z param during tar package * Fix: add values.yaml files * Fix: Remove -z param during tar package * Fix: Remove -z param during tar package * Fix: add values.yaml files * Fix: change format. --------- Signed-off-by: PhilexWong <[email protected]>
NethermindEth · Sep 6, 2024 · e12dc3a · e12dc3a
1 parent 80f0af7
commit e12dc3a
Show file tree

Hide file tree

Showing 5 changed files with 376 additions and 5 deletions.
diff --git a/charts/juno-node/Chart.yaml b/charts/juno-node/Chart.yaml
@@ -1,6 +1,6 @@
 apiVersion: v2
 name: juno-chart
-version: 0.1.4
+version: 0.1.5
 appVersion: "1"
 description: A Helm chart for deploying Juno service
 maintainers:

diff --git a/charts/juno-node/README.md b/charts/juno-node/README.md
@@ -1,6 +1,6 @@
 # juno-chart
 
-![Version: 0.1.4](https://img.shields.io/badge/Version-0.1.4-informational?style=flat-square) ![AppVersion: 1](https://img.shields.io/badge/AppVersion-1-informational?style=flat-square)
+![Version: 0.1.5](https://img.shields.io/badge/Version-0.1.5-informational?style=flat-square) ![AppVersion: 1](https://img.shields.io/badge/AppVersion-1-informational?style=flat-square)
 
 A Helm chart for deploying Juno service
 
@@ -27,6 +27,15 @@ A Helm chart for deploying Juno service
 | args.--ws | string | `"true"` |  |
 | args.--ws-host | string | `"0.0.0.0"` |  |
 | args.--ws-port | string | `"6061"` |  |
+| backupJunoDataJob.backupSchedule | string | `"*/20 * * * *"` |  |
+| backupJunoDataJob.cleanupSchedule | string | `"*/40 * * * *"` |  |
+| backupJunoDataJob.dataSource | string | `"juno-sepolia-pv-ssd-juno-sepolia-0"` |  |
+| backupJunoDataJob.enabled | bool | `true` |  |
+| backupJunoDataJob.endpoint | string | `"https://12345543.r2.cloudflarestorage.com"` |  |
+| backupJunoDataJob.key | string | `"key-1234"` |  |
+| backupJunoDataJob.network | string | `"sepolia"` |  |
+| backupJunoDataJob.secret | string | `"secret-12345"` |  |
+| backupJunoDataJob.storageSize | string | `"200Gi"` |  |
 | batchjob.enabled | bool | `false` |  |
 | batchjob.schedule | string | `"* */1 * * *"` |  |
 | deployment.healthCheck.enabled | bool | `false` |  |
@@ -84,6 +93,7 @@ A Helm chart for deploying Juno service
 | serviceAccount.enabled | bool | `false` |  |
 | serviceAccount.gcpServiceAccount | string | `"[email protected]"` |  |
 | serviceAccount.name | string | `"juno-pgo"` |  |
+| svc.externalTrafficPolicy | string | `""` |  |
 | svc.globalStaticInternalIpName | string | `""` |  |
 | svc.globalStaticIpName | string | `""` |  |
 | svc.ingress.enabled | bool | `true` |  |
@@ -155,4 +165,4 @@ A Helm chart for deploying Juno service
 | taintsToleration.tolerations.network | string | `"juno"` |  |
 
 ----------------------------------------------
-Autogenerated from chart metadata using [helm-docs v1.12.0](https://github.com/norwoodj/helm-docs/releases/v1.12.0)
+Autogenerated from chart metadata using [helm-docs v1.13.1](https://github.com/norwoodj/helm-docs/releases/v1.13.1)
diff --git a/charts/juno-node/templates/externalsecret-common.yaml b/charts/juno-node/templates/externalsecret-common.yaml
@@ -0,0 +1,20 @@
+{{- if .Values.secret }}
+{{- with .Values.secret.data }}
+apiVersion: external-secrets.io/v1beta1
+kind: ExternalSecret
+metadata:
+  name: {{ $.Values.deployment.projectName }}-external-secret
+  namespace: {{ $.Values.deployment.namespace }}
+spec:
+  refreshInterval: {{ $.Values.secret.data.refreshInterval }}
+  secretStoreRef:
+    name: {{ $.Values.secret.data.secretStoreName }}
+    kind: {{ $.Values.secret.data.secretStoreKind }}
+  target:
+    name: {{ $.Values.secret.data.targetName }}
+    creationPolicy: {{ $.Values.secret.data.targetCreationPolicy }}
+  dataFrom:
+    - extract:
+        key: {{ $.Values.secret.data.dataFromKey }} # name of the secret in secret manager (GCP secret manager)
+{{- end }}
+{{- end }}
diff --git a/charts/juno-node/templates/juno-data-backup-cronjob.yaml b/charts/juno-node/templates/juno-data-backup-cronjob.yaml
@@ -0,0 +1,309 @@
+{{- if .Values.backupJunoDataJob.enabled -}}
+# Service Account for the Backup Job
+apiVersion: v1
+kind: ServiceAccount
+metadata:
+  name: {{ .Values.deployment.namespace }}-backup-junodata-sa
+  namespace: {{ .Values.deployment.namespace }}
+---
+
+# Role for Backup Job with necessary permissions
+apiVersion: rbac.authorization.k8s.io/v1
+kind: Role
+metadata:
+  name: {{ .Values.deployment.namespace }}-backup-junodata-role
+  namespace: {{ .Values.deployment.namespace }}
+rules:
+  - apiGroups: [ "", "apps","batch"]
+    resources: ["pods", "jobs", "persistentvolumeclaims"]
+    verbs: ["get", "list","create", "update", "patch", "delete"]
+---
+# RoleBinding to bind Role with ServiceAccount
+apiVersion: rbac.authorization.k8s.io/v1
+kind: RoleBinding
+metadata:
+  name: {{ .Values.deployment.namespace }}-backup-junodata-rolebinding
+  namespace: {{ .Values.deployment.namespace }}
+subjects:
+  - kind: ServiceAccount
+    name: {{ .Values.deployment.namespace }}-backup-junodata-sa
+    namespace: {{ .Values.deployment.namespace }}
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: Role
+  name: {{ .Values.deployment.namespace }}-backup-junodata-role
+---
+
+# Secret to store R2 Cloud credentials
+apiVersion: v1
+kind: Secret
+metadata:
+  name: {{ .Values.deployment.namespace }}-rclone-config
+  namespace: {{ .Values.deployment.namespace }}
+stringData:
+  rclone.conf: |
+    [R2]
+    type = s3
+    provider = Cloudflare
+    env_auth = true
+    endpoint = https://d1cc7d59ae8f8dc2b1aa530c41b5c6ec.r2.cloudflarestorage.com
+---
+# ConfigMap for cloning disk manifest
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: {{ .Values.deployment.namespace }}-cloning-disk-manifest
+  namespace: {{ .Values.deployment.namespace }}
+data:
+  cloning-disk-manifest.yaml: |
+    apiVersion: v1
+    kind: PersistentVolumeClaim
+    metadata:
+      name: {{ .Values.deployment.namespace }}-pv-ssd-snapshot
+      namespace: {{ .Values.deployment.namespace }}
+    spec:
+      dataSource:
+        name: {{ .Values.backupJunoDataJob.dataSource }}
+        kind: PersistentVolumeClaim
+      accessModes:
+      - ReadWriteOnce
+      storageClassName: premium-rwo
+      resources:
+        requests:
+          storage: {{ .Values.backupJunoDataJob.storageSize }}
+    ---
+    apiVersion: v1
+    kind: PersistentVolumeClaim
+    metadata:
+      name: {{ .Values.deployment.namespace }}-juno-data-backup-pvc
+      namespace: {{ .Values.deployment.namespace }}
+    spec:
+      accessModes:
+        - ReadWriteOnce
+      storageClassName: premium-rwo
+      resources:
+        requests:
+          storage: {{ .Values.backupJunoDataJob.storageSize }}
+---
+# ConfigMap for cloning juno manifest
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: {{ .Values.deployment.namespace }}-cloning-juno-manifest
+  namespace: {{ .Values.deployment.namespace }}
+data:
+  cloning-juno-manifest.yaml: |
+    apiVersion: batch/v1
+    kind: Job
+    metadata:
+      name: {{ .Values.deployment.namespace }}-juno-data-archival-job
+      namespace: {{ .Values.deployment.namespace }}
+    spec:
+      ttlSecondsAfterFinished: 60
+      template:
+        spec:
+          serviceAccountName: {{ .Values.deployment.namespace }}-backup-junodata-sa
+          volumes:
+          - name: juno-data-volume
+            persistentVolumeClaim:
+              claimName: {{ .Values.deployment.namespace }}-pv-ssd-snapshot
+          - name: {{ .Values.deployment.namespace }}-rclone-config
+            secret:
+              secretName: {{ .Values.deployment.namespace }}-rclone-config
+          - name: tar-backup-volume
+            persistentVolumeClaim:
+              claimName: {{ .Values.deployment.namespace }}-juno-data-backup-pvc
+          initContainers:
+          - name: juno-archival-tar
+            image: busybox
+            command: ["/bin/sh", "-c"]
+            args:
+            - |
+              rm -rf /mnt/juno-tar-backup/*.tar &&
+              rm -rf /mnt/data/*.tar &&
+              tar -cf /mnt/juno-tar-backup/juno_{{ .Values.backupJunoDataJob.network }}_{{ .Values.deployment.imagetag }}_$(date +\%Y\%m\%d).tar --exclude=./lost+found -C /mnt/data .  && sleep 10
+            volumeMounts:
+            - name: juno-data-volume
+              mountPath: /mnt/data
+            - name: tar-backup-volume
+              mountPath: /mnt/juno-tar-backup
+          containers:
+          - name: rclone-upload-container
+            image: rclone/rclone:latest
+            env:
+              - name: AWS_ACCESS_KEY_ID
+                valueFrom:
+                  secretKeyRef:
+                    name: {{ .Values.secret.data.targetName }}
+                    key: r2_access_key_id
+              - name: AWS_SECRET_ACCESS_KEY
+                valueFrom:
+                  secretKeyRef:
+                    name: {{ .Values.secret.data.targetName }}
+                    key: r2_secret_access_key
+            command: ["/bin/sh", "-c"]
+            args:
+            - |
+              apk add --no-cache curl &&
+              apk add --no-cache jq &&
+                latestBlockNumber=$(curl --location '{{ .Values.backupJunoDataJob.junoFreeEndpoint }}/{{ .Values.backupJunoDataJob.network }}-juno' --header 'Content-Type: application/json' --data '{ "jsonrpc": "2.0","method": "starknet_blockNumber", "id": 1}' | jq '.result') &&
+              echo "latestBlockNumber is $latestBlockNumber" &&
+              mv /mnt/juno-tar-backup/juno_{{ .Values.backupJunoDataJob.network }}_{{ .Values.deployment.imagetag }}*.tar /mnt/juno-tar-backup/juno_{{ .Values.backupJunoDataJob.network }}_{{ .Values.deployment.imagetag }}_$latestBlockNumber.tar &&
+              echo "/mnt/juno-tar-backup/juno_{{ .Values.backupJunoDataJob.network }}_{{ .Values.deployment.imagetag }}_$latestBlockNumber.tar" &&
+              rclone copy /mnt/juno-tar-backup/*.tar R2:/{{ .Values.backupJunoDataJob.bucketName }}/{{ .Values.backupJunoDataJob.network }}
+            volumeMounts:
+            - name: {{ .Values.deployment.namespace }}-rclone-config
+              mountPath: /config/rclone
+            - name: tar-backup-volume
+              mountPath: /mnt/juno-tar-backup
+          restartPolicy: OnFailure
+---
+# CronJob for Backup Task
+apiVersion: batch/v1
+kind: CronJob
+metadata:
+  name: {{ .Values.deployment.namespace }}-backup-junodata-cronjob
+  namespace: {{ .Values.deployment.namespace }}
+spec:
+  schedule: "{{ .Values.backupJunoDataJob.backupSchedule }}"
+  concurrencyPolicy: Forbid
+  successfulJobsHistoryLimit: 1
+  failedJobsHistoryLimit: 1
+  jobTemplate:
+    spec:
+      completions: 1
+      ttlSecondsAfterFinished: 30
+      template:
+        spec:
+          serviceAccountName: {{ .Values.deployment.namespace }}-backup-junodata-sa
+          restartPolicy: Never
+          initContainers:
+            - name: copy-disk-kubectl-container
+              image: bitnami/kubectl:latest
+              command: ["/bin/sh"]
+              args: ["-c", "kubectl apply -f /cloning-disk-manifest/cloning-disk-manifest.yaml"]
+              volumeMounts:
+                - name: cloning-disk-manifest-volume
+                  mountPath: /cloning-disk-manifest
+          containers:
+            - name: clone-juno-kubectl-container
+              image: bitnami/kubectl:latest
+              command: ["/bin/sh"]
+              args: ["-c", "kubectl apply -f /cloning-juno-manifest/cloning-juno-manifest.yaml"]
+              volumeMounts:
+                - name: cloning-juno-manifest-volume
+                  mountPath: /cloning-juno-manifest
+          volumes:
+            - name: cloning-disk-manifest-volume
+              configMap:
+                name: {{ .Values.deployment.namespace }}-cloning-disk-manifest
+            - name: cloning-juno-manifest-volume
+              configMap:
+                name: {{ .Values.deployment.namespace }}-cloning-juno-manifest
+---
+# CronJob for Cleaning up Completed Pods and PVCs
+apiVersion: batch/v1
+kind: CronJob
+metadata:
+  name: {{ .Values.deployment.namespace }}-delete-used-pvc
+  namespace: {{ .Values.deployment.namespace }}
+spec:
+  schedule: "{{ .Values.backupJunoDataJob.cleanupSchedule }}"
+  concurrencyPolicy: Forbid
+  successfulJobsHistoryLimit: 1
+  failedJobsHistoryLimit: 1
+  jobTemplate:
+    spec:
+      completions: 1
+      ttlSecondsAfterFinished: 30
+      template:
+        spec:
+          serviceAccountName: {{ .Values.deployment.namespace }}-backup-junodata-sa
+          restartPolicy: OnFailure
+          containers:
+            - name: kubectl-container
+              image: bitnami/kubectl:latest
+              command:
+                - "/bin/bash"
+                - "-c"
+                - |
+                  # Delete PVC if not used
+                  describe_output=$(kubectl describe pvc {{ .Values.deployment.namespace }}-pv-ssd-snapshot)
+                  if echo "$describe_output" | grep -q "Used By:[[:space:]]*<none>"; then
+                    echo "Deleting {{ .Values.deployment.namespace }}-pv-ssd-snapshot..."
+                    kubectl delete pvc {{ .Values.deployment.namespace }}-pv-ssd-snapshot
+                    sleep 30
+                  fi
+                  describe_output=$(kubectl describe pvc {{ .Values.deployment.namespace }}-juno-data-backup-pvc)
+                  if echo "$describe_output" | grep -q "Used By:[[:space:]]*<none>"; then
+                    echo "Deleting {{ .Values.deployment.namespace }}-juno-data-backup-pvc..."
+                    kubectl delete pvc {{ .Values.deployment.namespace }}-juno-data-backup-pvc
+                    sleep 30
+                  fi
+---
+# CronJob for Cleaning up Completed Pods and PVCs
+apiVersion: batch/v1
+kind: CronJob
+metadata:
+  name: {{ .Values.deployment.namespace }}-r2-retention-cronjob
+  namespace: {{ .Values.deployment.namespace }}
+spec:
+  schedule: "0 0 * * */2"
+  jobTemplate:
+    spec:
+      completions: 1
+      ttlSecondsAfterFinished: 30
+      template:
+        spec:
+          restartPolicy: OnFailure
+          containers:
+            - name: {{ .Values.deployment.namespace }}-r2-retention
+              image: ubuntu:latest
+              env:
+                - name: RETENTION_LIMIT
+                  value: "{{ .Values.backupJunoDataJob.retentionLimit }}"
+                - name: API_TOKEN
+                  valueFrom:
+                    secretKeyRef:
+                      name: {{ .Values.secret.data.targetName }}
+                      key: r2_api_token
+                - name: ACCOUNT_ID
+                  value: "d1cc7d59ae8f8dc2b1aa530c41b5c6ec"
+                - name: BUCKET_NAME
+                  value:  "{{ .Values.backupJunoDataJob.bucketName }}"
+              command:
+                - /bin/sh
+                - -c
+                - |
+                  #!/bin/sh
+                  mkdir -p /var/lib/apt/lists/partial
+                  apt-get update && apt-get install -y curl jq
+                  # Constants
+                  API_TOKEN="$API_TOKEN"
+                  RETENTION_LIMIT="$RETENTION_LIMIT"
+                  ACCOUNT_ID="$ACCOUNT_ID"
+                  BUCKET_NAME="$BUCKET_NAME"
+
+                  # Construct the Cloudflare API URL with account ID and bucket name
+                  CLOUDFLARE_API_URL="https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/r2/buckets/$BUCKET_NAME/objects?prefix={{ .Values.backupJunoDataJob.network }}/"
+                  # Get the list of objects with the specified prefix
+                  objects=$(curl -s -X GET "$CLOUDFLARE_API_URL" -H "Authorization: Bearer $API_TOKEN" | jq -r '.result')
+
+                  # Check if the number of objects exceeds the retention limit
+                   object_count=$(echo "$objects" | jq length)
+                  echo "total backup number is $object_count"
+
+                   if [ "$object_count" -le "$RETENTION_LIMIT" ]; then
+                    echo "exiting...."
+                    exit 0
+                   fi
+                  delete_number=$((object_count - RETENTION_LIMIT))
+                  # Sort the objects by last_modified date and delete the oldest ones
+                  echo "$objects" | jq -r '.[] | [.key, .last_modified] | @tsv' | sort -k2 | head -n "$delete_number" | while IFS=$'\t' read -r key last_modified; do
+                  delete_url="https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/r2/buckets/$BUCKET_NAME/objects/${key}tar"
+                  echo "Deleting ${key}tar at $delete_url"
+                  delete_response=$(curl -s -X DELETE "$delete_url" -H "Authorization: Bearer $API_TOKEN")
+                  echo "Delete response: $delete_response"
+                  done
+---
+{{- end -}}