Air-Gapped Environment

In the air-gapped environment, you need to manually manage container images for EGS. You can use the following scripts to extract and push container images for air-gapped deployments:

1. airgap-image-pull.sh: Extracts all container images from Helm charts for kubeslice-egs-helm-ent-prod charts. It reads chart information from a configuration file and generates a comprehensive list of all container images used in the specified Helm charts.

2. airgap-image-push.sh: Pushes container images to a target repository. It reads a list of images and performs pull, tag, and push operations for AirGap deployments.

Prerequisites

Before you begin, ensure you have the following prerequisites:

• Helm installed on your local machine with version v3.0.0 or higher.
• Git installed on your local machine.
• yq installed on your local machine with version v4.0.0 or higher.
• bash shell to run the scripts with version v4.0.0 or higher.
• sed command-line utility installed on your local machine.
• Access to the target container image repository where you want to push the images.
• Network access to pull images from public registries (for the initial extraction step).

Clone the Repository

To copy the scripts, clone the EGS installation repository using the following command:

git clone https://github.com/kubeslice-ent/egs-installation.git

Navigate to the egs-installation/airgap-image-push directory:

cd egs-installation/airgap-image-push

Input File for Helm Charts

The airgap-image-pull.sh and airgap-image-push.sh scripts require an input file that specifies the Helm charts to process. The airgap-image-pull.sh script reads chart information from a text file named helm-chart-info.txt. This file must be specified using the HELM_CHART_INFO environment variable.

Field Descriptions

The helm-chart-info.txt file should contain the following fields in a space-separated format:

• repo-url: The URL of the Helm chart repository. The value must be https://kubeslice.aveshalabs.io/repository.
• repo-name: The name of the Helm chart repository. The value must be kubeslice-egs-helm-ent-prod.
• chart-name: The name of the Helm chart. It must be one of the following charts:
  • kubeslice-controller-egs
  • kubeslice-worker-egs
  • kubeslice-ui-egs
• chart-version: The version of the Helm chart. You can specify a specific version (for example, 1.15.4) or use * to indicate the latest version available in the repository.

Example Input Files

The following is an example of helm-chart-info.txt with specific Versions:

*****************<repo-url>*************** ******<repo-name>****** ****<chart-name>**** ****<chart-version>****
https://kubeslice.aveshalabs.io/repository kubeslice-egs-helm-ent-prod kubeslice-controller-egs 1.15.4
https://kubeslice.aveshalabs.io/repository kubeslice-egs-helm-ent-prod kubeslice-worker-egs 1.15.4
https://kubeslice.aveshalabs.io/repository kubeslice-egs-helm-ent-prod kubeslice-ui-egs 1.15.4

note

You can specify * for the <chart-version> field to indicate the latest version available in the repository.

info

The helm-chart-info.txt is an example file to illustrate the required format for the input file used by the scripts. The file is available in the airgap-image-push/ directory for reference. When running the scripts from within the airgap-image-push/ directory, use:

export HELM_CHART_INFO="helm-chart-info.txt"

The file name helm-chart-info.txt is a convention, but you can use any filename. Ensure you set the HELM_CHART_INFO environment variable to point to your input file.

Extract Container Images

The airgap-image-pull.sh script reads chart information from file named helm-chart-info.txt. This file should be specified using the HELM_CHART_INFO environment variable.

To extract container images from Helm charts, use the airgap-image-pull.sh script with the following command:

export HELM_CHART_INFO="helm-chart-info.txt"
bash airgap-image-pull.sh

The output is a file named helm-chart-images.txt that contains a list of all unique container images used in the specified Helm charts.

Push Container Images to Target Repository

The airgap-image-push.sh script reads a list of images from a file named helm-chart-info.txt. This file should be specified using the HELM_CHART_INFO environment variable.

Supported Registry Formats

Registry Type	REPO Format	Tagged Image Format
Docker Hub	docker.io/username	username/image:tag
Private Registry (Nexus)	registry.example.com	registry.example.com/path/image:tag

Prerequisites

Before pushing images to the target repository, ensure you have the following prerequisites:

• Docker installed on your local machine.

• Access credentials for the target container image repository. you may need to log in to the repository using Docker CLI.

  • Docker login command.

    docker login -u <username>

  • Privately hosted registries may require additional parameters such as --password and --email.

    docker login -u <username> --password <password> --email <email> <registry-url>

  Authentication is required to push images to private registries.

Push Images

To push container images to a target repository, use the airgap-image-push.sh script with the following commands:

1. Set the HELM_CHART_INFO environment variable to point to the helm-chart-images.txt file generated in the previous step.

2. Set the REPO variable for the target repository where you want to push the images.

3. Use the following helper command to copy the images from the helm-chart-images.txt file to a temporary file /tmp/images.txt in the required format:

   cat helm-chart-images.txt | sed 's/^/"/' | sed 's/$/"/' > /tmp/images.txt

4. Manually copy the images from the /tmp/images.txt file and update the IMAGES array in the airgap-image-push.sh script.

5. Run the airgap-image-push.sh script using the following command:

   bash airgap-image-push.sh