Skip to content

Offline Upgrade of Middleware - Kafka Module

This page explains how to install or upgrade the middleware - kafka module after downloading it from the Download Center.

Info

The term mcamel used in the following commands or scripts is an internal development codename for the middleware module.

Loading Images from the Installation Package

You can load the images using one of the following methods. It is recommended to use chart-syncer to sync the images to your container registry when it is available, as it is more efficient and convenient.

Syncing Images to Container Registry using chart-syncer

  1. Create load-image.yaml file.

    Note

    All parameters in this YAML file are required. You need to have a private container registry and modify the relevant configuration.

    If you already have a chart repo installed in your environment, chart-syncer also supports exporting the chart as a tgz file.

    load-image.yaml
    source:
  intermediateBundlesPath: mcamel-offline # The relative path to execute the charts-syncer command, not the relative path between this YAML file and the offline package
target:
  containerRegistry: 10.16.10.111 # Change this to your container registry URL
  containerRepository: release.daocloud.io/mcamel # Change this to your container registry
  repo:
    kind: HARBOR # It can also be any other supported Helm Chart repository type
    url: http://10.16.10.111/chartrepo/release.daocloud.io # Change this to the chart repo URL
    auth:
      username: "admin" # Your container registry username
      password: "Harbor12345" # Your container registry password
  containers:
    auth:
      username: "admin" # Your container registry username
      password: "Harbor12345" # Your container registry password

    If you don't have a chart repo installed in your environment, chart-syncer also supports exporting the chart as a tgz file and storing it in the specified path.

    load-image.yaml
    source:
  intermediateBundlesPath: mcamel-offline # The relative path to execute the charts-syncer command, not the relative path between this YAML file and the offline package
target:
  containerRegistry: 10.16.10.111 # Change this to your container registry URL
  containerRepository: release.daocloud.io/mcamel # Change this to your container registry
  repo:
    kind: LOCAL
    path: ./local-repo # Local path to the chart
  containers:
    auth:
      username: "admin" # Your container registry username
      password: "Harbor12345" # Your container registry password

  2. Run the command to sync the images.

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

Directly Loading with Docker or containerd

Unpack and load the image files.

  1. Extract the tar archive.

    tar -xvf mcamel-kafka_0.8.1_amd64.tar
cd mcamel-kafka_0.8.1_amd64
tar -xvf mcamel-kafka_0.8.1.bundle.tar

    After successful extraction, you will get 3 files:

    • hints.yaml
    • images.tar
    • original-chart

  2. Load the images to Docker or containerd from local.

    docker load -i images.tar

    ctr -n k8s.io image import images.tar

Note

Docker or containerd image loading operation needs to be performed on each node. After loading, tag the images to match the registry and repository used during installation.

Upgrade

There are two ways to perform the upgrade. You can choose the corresponding upgrade method based on the preconditions:

  1. Check if the helm repository exists.

    helm repo list | grep kafka

    If the result is empty or shows the following prompt, proceed to the next step; otherwise, skip the next step.

    Error: no repositories to show

  2. Add the helm repository.

    helm repo add mcamel-kafka http://{harbor url}/chartrepo/{project}

  3. Update the helm repository.

    helm repo update mcamel/mcamel-kafka # If the helm version is too low and it fails, try executing __helm update repo__

  4. Select the version you want to install (recommended to install the latest version).

    helm search repo mcamel/mcamel-kafka --versions

    [root@master ~]# helm search repo mcamel/mcamel-kafka --versions
NAME                            CHART VERSION   APP VERSION     DESCRIPTION               
mcamel/mcamel-kafka     0.8.1           0.8.1           A Helm chart for Kubernetes
...

  5. Back up the --set parameters.

    Before upgrading to a new version, it is recommended to execute the following command to back up the --set parameters of the old version.

    helm get values mcamel-kafka -n mcamel-system -o yaml > mcamel-kafka.yaml

  6. Run helm upgrade .

    Before upgrading, it is recommended to update the global.imageRegistry field in the mcamel-kafka.yaml file with the current container registry address.

    export imageRegistry={your_image_repository}

    helm upgrade mcamel-kafka mcamel/mcamel-kafka \
  -n mcamel-system \
  -f ./mcamel-kafka.yaml \
  --set global.imageRegistry=$imageRegistry \
  --version 0.8.1

  1. Back up the --set parameters.

    Before upgrading to a new version, it is recommended to execute the following command to back up the --set parameters of the old version.

    helm get values mcamel-kafka -n mcamel-system -o yaml > mcamel-kafka.yaml

  2. Run helm upgrade .

    Before upgrading, it is recommended to update the bak.yaml file by replacing global.imageRegistry with your current container registry address.

    export imageRegistry={your_image_repository}

    helm upgrade mcamel-kafka . \
  -n mcamel-system \
  -f ./mcamel-kafka.yaml \
  --set global.imageRegistry=${imageRegistry} \
  --set console_image.registry=${imageRegistry} \ 
  --set operator_image.registry=${imageRegistry}

Comments