Uploading SpeechSense images

The basic Stackland package does not include SpeechSense images. Make sure to upload the images to the cluster before activating the component.

Getting startedGetting started

1. Make sure your cluster service account can access protected images in the container registry.
2. Download the latest sladm version as described in our Installation guide.

Cluster with internet accessCluster with internet access

If your cluster has internet access, pull the SpeechSense images directly from the container registry:

sladm pull \
  --image-bundle speechsense \
  --config config/

Where:

• --image-bundle: Name of the protected component. To pull SpeechSense images, specify speechsense.

The command will create a directory with container images in OCI format.

Air-gapped clusterAir-gapped cluster

If the cluster has no internet access, pull the images to an internet-connected machine, transfer them into the internal infrastructure, and then push them to the cluster’s local registry.

Pulling images on an internet-connected machinePulling images on an internet-connected machine

On an internet-connected machine, run this command:

sladm pull \
  --image-bundle speechsense \
  --config config/

Transferring artifacts into the internal infrastructureTransferring artifacts into the internal infrastructure

Transfer the following to a machine with cluster access:

• Directory with container images in OCI format.
• Binary file named sladm.

The transfer method depends on the security policies in your organization: removable media, secure file storage systems, or isolated network segments.

Pushing images to the local registryPushing images to the local registry

On a machine with cluster access, run this command:

sladm push \
  --local-registry \
  --kubeconfig=<kubeconfig_path> \
  --image-bundle-folder <path_to_directory_with_images>

Where:

• --local-registry: Indicates the use of the cluster’s local registry.
• --kubeconfig: Path to the kubeconfig file for access to the cluster.
• --image-bundle-folder: Path to the directory with container images.

If you only need to install SpeechSense on some of the cluster nodes, you can specify a list of target nodes for copying images in the following format: --target-registry <host_name_or_IP_1>:5001,<host_name_or_IP_2>:5001.

What's nextWhat's next

• Configuring and activating SpeechSense: Set up and activate SpeechSense in your cluster.