Subscribe

UiPath AI Center

UiPath AI Center

Docker registry

Any docker registry compatible with Docker API v2 (for example, Artifactory, Azure Registry, etc.) can be used with AI Center to pull images from the kubernetes cluster. Every docker image in the AI Center installation can be represented as following:
<repository_url>/<project_id>/<image_name>:<image_tag>
The existing cluster installation supports two types of installation:

  • Online installation
  • Airgap installation

Online installation


In the case of online installation, installation artifacts are pulled directly from the UiPath hosted docker registry. These artifacts are used for completing the installation.
UiPath registry uses read-only user credentials. Use the following command to create secrets in the customer kubernetes cluster with UiPath provided credentials:

kubectl -n uipath create secret docker-registry aicenter-registry-secrets --docker-server="<uipath-registry>" --docker-username="<username>" --docker-password="<password>"

Airgap installation


In the case of airgap installation, you need to pull all installation artifacts (docker images) into their own registry. These artifacts are used for completing the installation.
You need to provide their registry imagePullSecrets, repository_url, and project_id to pull the images from.

kubectl -n uipath create secret docker-registry aicenter-registry-secrets --docker-server="<customer-docker-registry>" --docker-username="<username>" --docker-password="<password>"

Parameters


Input parameter

Required

Description

Default

global.registry.imagePullSecrets.name

Yes

Pull secrets to be placed in UiPath namespace to pull images from the registry to initiate installation.

aicenter-registry-secrets

global.registry.repository_url

Yes

Registry hostname to pull images from

global.registry.project_id

Yes

Registry project to formulate complete image names

aicenter

Updated 3 months ago

Docker registry


Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.