Offline upgrade¶

Workbench supports offline upgrade. This guide will walk you through Workbench upgrade process.

Load image from installation package¶

You can load an image in two ways.

Load with chart-Syncer¶

If a mirror warehouse exists in the environment, it is recommended to use chart-syncer to synchronize images to the mirror warehouse, which is more efficient and convenient.

Create load-image.yaml as the chart-syncer profile

All parameters in the load-image.yaml file are mandatory. You need a private container registry and modify configurations as described below. See Official Doc for a detailed explanation of the chart-syncer profile.

chart repo installed

If chart repo is already install, use the following configuration to synchronize the image directly.

load-image.yaml

source:
  intermediateBundlesPath: amamba-offline # Relative path to executing chart-syncer command, **not** the relative path between this YAML file and the offline package
target:
  containerRegistry: 10.16.10.111 # image repo url
  containerRepository: amamba # the specific project in image repo
  repo:
    kind: HARBOR # Can be other supported Helm Chart repos, like ChartMuseum
    url: http://10.16.10.111/chartrepo/amamba #  change to chart repo url
    auth: # username/password
      username: "admin"
      password: "Harbor12345"
  containers:
    auth: # username/password
      username: "admin"
      password: "Harbor12345"

chart repo not installed

Chart-syncer also supports exporting a chart as a tgz file in a specified path if chart repo is not installed.

load-image.yaml

source:
    intermediateBundlesPath: amamba-offline #  Relative path to executing chart-syncer command, **not** the relative path between this YAML file and the offline package
target:
    containerRegistry: 10.16.10.111 # change to your image repo url
    containerRepository: release.daocloud.io/amamba # change to the project in your image repo
    repo:
      kind: LOCAL
      path: ./local-repo # loca path of chart
    containers:
      auth:
        username: "admin" # your username to access image repo
        password: "Harbor12345" # your password to access image repo

Run this command to sync the image.

charts-syncer sync --config load-image.yaml

Load with Docker/containerd¶

Decompress the tar package.
```
tar xvf amamba.bundle.tar
```
After the decompression, you will get 3 files:
- hints.yaml
- images.tar
- original-chart
Load the image from local to a Docker or containerd.
Dockercontainerd
docker load -i images.tar
ctr image import images.tar

Note

The image needs to be loaded via Docker or containerd to each node.
After the loading is complete, you should tag the image to keep version consistency.

Start Upgrade¶

Check if the helm repository of Workbench exists. amamba is the internal code for Workbench.
```
helm repo list | grep amamba
```
If nothing is returned or Error: no repositories to show is displayed, run the following command to add the Helm repository for Workbench.
```
helm repo add amamba http://{harbor url}/chartrepo/{project}
```

Update Workbench's Helm repository.

# outdated Helm versions may cause failure. If failed, try `helm update repo`
helm repo update amamba

Backup --set parameters. Before upgrading the global management version, it is recommended to run the following commands to back up --set parameters of the previous version.
```
helm get values amamba -n amamab-system -o yaml > amamba.bak.yaml
```

Select the version of Workbench you want to install (the latest version is recommended).

$ helm search  repo amamba-release-ci --versions |head
NAME                                   CHART VERSION        APP VERSION         DESCRIPTION                               
amamba-release-ci/amamba                0.14.0              0.14.0               Amamba is the entrypoint to DCE 5.0, provides de...

Modify registry and tag in amamba.bak.yaml.

Click to see sample YAML file

amamba.bak.yaml

amambaSyncer:
  resources:
    limits:
      cpu: 200m
      memory: 256Mi
    requests:
      cpu: 20m
      memory: 128Mi
apiServer:
  configMap:
    debug: true
  fromJar:
    image:
      registry: releas-ci.daocloud.io
      repository: docker/library/openjdk
      tag: 11.0-jre-slim
  image:
    registry: release-ci.daocloud.io
    repository: amamba/amamba-apiserver
  resources:
    limits:
      cpu: "2"
      memory: 2Gi
    requests:
      cpu: 20m
      memory: 150Mi
argo-cd:
  applicationSet:
    enabled: false
    image:
      repository: release-ci.daocloud.io/quay/argoproj/argocd
      tag: v2.4.12
  controller:
    image:
      repository: release-ci.daocloud.io/quay/argoproj/argocd
      tag: v2.4.12
    resources:
      limits:
        cpu: "2"
        memory: 2Gi
      requests:
        cpu: 100m
        memory: 256Mi
  dex:
    enabeld: true
    image:
      repository: release-ci.daocloud.io/ghcr/dexidp/dex
      tag: v2.32.0
    initImage:
      repository: release-ci.daocloud.io/quay/argoproj/argocd
      tag: v2.4.12
    resources:
      limits:
        cpu: 50m
        memory: 64Mi
      requests:
        cpu: 10m
        memory: 16Mi
  enabled: true
  notifications:
    enabled: false
  redis:
    enabled: true
    image:
      repository: release-ci.daocloud.io/docker/library/redis
      tag: 7.0.4-alpine
    metrics:
      enabled: false
    resources:
      limits:
        cpu: 200m
        memory: 128Mi
      requests:
        cpu: 5m
        memory: 16Mi
  repoServer:
    image:
      repository: release-ci.daocloud.io/quay/argoproj/argocd
      tag: v2.4.12
    resources:
      limits:
        cpu: 200m
        memory: 256Mi
      requests:
        cpu: 5m
        memory: 8Mi
  server:
    image:
      repository: release-ci.daocloud.io/quay/argoproj/argocd
      tag: v2.4.12
    resources:
      limits:
        cpu: 250m
        memory: 256Mi
      requests:
        cpu: 5m
        memory: 8Mi
    service:
      nodePortHttp: 31886
      nodePortHttps: 31887
      type: NodePort
argo-rollouts:
  controller:
    image:
      registry: release-ci.daocloud.io
      repository: quay/argoproj/argo-rollouts
      tag: v1.2.0
    resources:
      limits:
        cpu: 100m
        memory: 128Mi
      requests:
        cpu: 16m
        memory: 128Mi
  dashboard:
    enabled: true
    image:
      registry: release-ci.daocloud.io
      repository: quay/argoproj/kubectl-argo-rollouts
      tag: v1.2.0
    resources:
      limits:
        cpu: 100m
        memory: 128Mi
      requests:
        cpu: 10m
        memory: 16Mi
    service:
      nodePort: 30070
      type: NodePort
  enabled: true
devopsServer:
  enabled: true
  image:
    registry: release-ci.daocloud.io
    repository: amamba/amamba-devops-server
  resources:
    limits:
      cpu: "2"
      memory: 2Gi
    requests:
      cpu: 50m
      memory: 160Mi
global:
  amamba:
    imageTag: v0.13-dev-a8ca782a
  imageRegistry: release-ci.daocloud.io
mysql:
  busybox:
    repository: docker/busybox
    tag: 1.35.0
  image:
    repository: docker/mysql
    tag: 5.7.30
  persistence:
    size: 20Gi
  resources:
    limits:
      cpu: 500m
      memory: 512Mi
    requests:
      cpu: 20m
      memory: 256Mi
ui:
  image:
    tag: v0.15.0-dev-fd64789e
  resources:
    limits:
      cpu: 100m
      memory: 128Mi
    requests:
      cpu: 5m
      memory: 4Mi

Run the following command to upgrade

helm upgrade amamba . \
  -n amamba-system \
  -f ./amamba.bak.yaml